Zbirka / Series OPERA INSTITUTI ARCHAEOLOGICI SLOVENIAE 40 Uredniki zbirke / Editors of the series Jana Horvat, Andrej Pleterski, Anton Velušček Jana Horvat, Irena Lazar, Manjša rimska naselja na slovenskem prostoru Andrej Gaspari (ur. / eds.) Minor Roman settlements in Slovenia Recenzenta / Reviewed by Janez Dular, Ivan Šprajc Prevod / Translation Andreja Maver, Meta Osredkar, Gregor Pobežin, Lucija Jelenko Jezikovni pregled / Language Editors Urška Kosec, Špela Križ, Jana Volk, Terry T. Jackson Tehnična ureditev / Technical Editors Andreja Dolenc Vičič, Mateja Belak Oblikovanje ovitka / Front cover design Tamara Korošec Priprava slikovnega gradiva / Preparation of illustrations Drago Valoh, Mateja Belak Prelom / DTP Mateja Belak Založnik / Publisher Založba ZRC Zanj / Represented by Aleš Pogačnik Izdajatelj / Issued by ZRC SAZU, Inštitut za arheologijo Zanj / Represented by Anton Velušček Tisk / Printed by Present d. o. o., Ljubljana Naklada / Print run 500 izvodov / copies Izid knjige sta podprla / Javna agencija za raziskovalno dejavnost RS (Slovenian Research Agency), Published with the support of Znanstvenoraziskovalni center SAZU (Research Centre of the Slovenian Academy of Sciences and Arts) Ljubljana 2020; prva izdaja, prvi natis / first edition, first print Prva e-izdaja knjige (pdf) je pod pogoji licence Creative Commons 4.0 CC-BY-NC-SA prosto dostopna tudi v elektronski obliki (pdf) / First e-edition of the book (pdf) is freely available in e-form (pdf) under the Creative Commons 4.0 CC-BY-NC-SA. DOI: https://doi.org/10.3986/9789610502586 CIP - Kataložni zapis o publikaciji Narodna in univerzitetna knjižnica, Ljubljana 904(497.4-2)«652« MANJŠA rimska naselja na slovenskem prostoru = Minor Roman settlements in Slovenia / uredniki, edited by Jana Horvat, Irena Lazar, Andrej Gaspari ; [prevod Andreja Maver ... et al.]. - 1. izd., 1. natis. - Ljubljana : Založba ZRC, 2020. - (Opera Instituti Archaeologici Sloveniae, ISSN 1408-5208 ; 40) ISBN 978-961-05-0257-9 1. Vzp. stv. nasl. 2. Horvat, Jana, 1959­COBISS.SI-ID 303610624 ISBN 978-961-05-0258-6 (pdf) COBISS.SI ID 303643904 Raziskava je vključena v program (P6-0064 (B)), ki ga sofinancira Javna agencija za raziskovalno dejavnost Republike Slovenije iz državnega proračuna. / The authors acknowledge the financial support from the Slovenian Research Agency (research core funding No. P6-0064 (B)). © 2020, ZRC SAZU, Inštitut za arheologijo, Založba ZRC MANJŠA RIMSKA NASELJA NA SLOVENSKEM PROSTORU MINOR ROMAN SETTLEMENTS IN SLOVENIA Uredniki / Edited by Jana Horvat Irena Lazar Andrej Gaspari LJUBLJANA 2020 VSEBINA / CONTENTS Predgovor / Preface (Jana HORVAT, Irena LAZAR, Andrej GASPARI) ........................................................................ 7 Fluvio Frigido, Castra – Ajdovščina (Tina ŽERJAL, Vesna TRATNIK) .......................................................................... 9 Fluvio Frigido, Castra – Ajdovščina. Raziskave / Investigations 2017-2019 (Maruša UREK, Ana KOVAČIČ)......47 Ad Pirum - Hrušica (Peter KOS) ....................................................................................................................................... 61 Longaticum – Logatec (Ahac ŠINKOVEC) ....................................................................................................................... 77 Nauportus - Vrhnika (Jana HORVAT) ............................................................................................................................... 93 Vipava (Vesna TRATNIK) ................................................................................................................................................ 113 Gradišče nad Knežakom (Boštjan LAHARNAR, Edisa LOZIĆ, Alenka MIŠKEC) .................................................. 123 Ulaka (Andrej GASPARI) ................................................................................................................................................. 141 Ig (Lucija GRAHEK, Anja RAGOLIČ) ............................................................................................................................ 173 Mengeš (Milan SAGADIN) .............................................................................................................................................. 187 Carnium – Kranj (Milan SAGADIN) .............................................................................................................................. 201 Šmartno pri Cerkljah (Špela TOMAŽINČIČ, Draško JOSIPOVIČ) .......................................................................... 213 Blagovica (Ana PLESTENJAK) ........................................................................................................................................ 231 Atrans – Trojane (Janja ŽELEZNIKAR, Julijana VISOČNIK) ..................................................................................... 249 Šempeter v Savinjski dolini (Irena LAZAR) ................................................................................................................... 295 Colatio – Stari trg pri Slovenj Gradcu (Saša DJURA JELENKO) ................................................................................. 305 Zagrad (Saša DJURA JELENKO) ..................................................................................................................................... 325 Slovenska Bistrica (Mira STRMČNIK GULIČ) ............................................................................................................. 339 Ančnikovo gradišče (Zvezdana MODRIJAN) ................................................................................................................ 349 Praetorium Latobicorum – Trebnje (Uroš BAVEC) ....................................................................................................... 363 Romula – Ribnica (Irena LAZAR) ................................................................................................................................... 387 Primerjalni pregled manjših rimskih naselij / Minor Roman settlements – comparative overview (Jana HORVAT) ............................................................................................................................................................. 403 PREDGOVOR / PREFACE Monografija je posvečena manjšim rimskim na­seljem, ki ležijo na območju današnje Slovenije. Gre za prehoden in geografsko raznolik prostor, na katerega so segale tri velike upravne enote rimske države: Italija ter provinci Norik in Zgornja Panonija. Strnjena naselja, ki jih predstavljamo, so po veli­kosti in pomenu vmesni člen med avtonomnimi mesti in razpršeno poselitvijo podeželja, katere osnova so bile vile rustike. V zadnjih desetletjih je arheologija z velikimi zaščitnimi izkopavanji pridobila obsežne in pomembne nove podatke o tovrstnih naseljih, ki pa do zdaj večinoma še niso bili poglobljeno analizirani in objavljeni. Potreba po zbranem vedenju se je npr. izrazito pokazala v sodelovanju z mednarodno skupino Adriaticum mare pri projektu Informatiziranega atlasa antičnega Jadrana (AdriAtlas). Tri ustanove, Inštitut za arheologijo Znanstvenoraziskovalnega centra Slovenske akademije znanosti in umetnosti, Inštitut za arheologijo in dediščino Fakultete za humanistične študije Univerze na Primorskem ter Oddelek za arheologijo Filozofske fakultete Univerze v Ljubljani, so leta 2015 dale pobudo, da bi vedenje o manjših naseljih zbrali, ustrezno ovre­dnotili in dvignili na višjo raven. Delo je dozorelo v štirih letih. Šestindvajset avtorjev prihaja iz različnih ustanov, od univerz, raziskovalnih inštitutov in muzejev do Zavoda za varstvo kulturne dediščine Slovenije, ali pa gre za posameznike, ki delujejo kot samostojni raziskovalci. V samostojnih poglavjih predstavljajo dvajset naselij različnega tipa in stopnje raziskanosti. Avtorji so najboljši poznavalci oziroma aktivni raziskovalci najdišč, ki jih obravnavajo. Prav za potrebe monografije so se poglobili v primarno dokumentacijo novejših raziskav in poročila, v drobno gradivo in razpršene starejše objave. Zgoščene predsta­vitve naselij bolj ali manj sledijo skupnemu konceptu. Podatki so umeščeni v prostor, podprti s kartami in načrti, vsebinsko primerljivi in jasno ovrednoteni. Vsa­ko poglavje vsebuje podatke o legi naselja in njegovem antičnem imenu, kratko zgodovino raziskav, morebitno obljudenost lokacije v prazgodovini, predstavitev antič­nih literarnih virov in epigrafskih spomenikov. Osrednji del je usmerjen v pregled arheoloških ostankov rimske The book discusses the minor settlements that dotted the territory of present-day Slovenia in the Roman period. This geographically diverse territory was crossed by important lines of communication and divided between three large administrative units of the Roman state: Italy and the provinces of Noricum and Upper Pannonia. The compact minor settlements represent a link, in both size and significance, between the autonomous towns and the dispersed settlement of the countryside with countryside villas as its backbone. In recent dec­ades, the large-scale rescue excavations across Slovenia have yielded vast and important data on these minor settlements, though they have for the most part not yet been analysed in detail and published. Clearly, there is a great desire for the information these excavations brought to light, but the consequent knowledge became even more desirable during the collaboration with the international Adriaticum Mare group on the AdriAtlas or Computerised Atlas of the Antique Adriatic project. In 2015, three institutions in Slovenia (Institute of Archaeology ZRC SAZU; Institute for Archaeology and Heritage, Faculty of Humanities, University of Pri­morska; Department of Archaeology, Faculty of Arts, University of Ljubljana) came together with the shared goal of advancing our knowledge on the minor Roman settlements in Slovenia by appropriately analysing and evaluating the data, new and old. The book is the result of four years of work and research. Twenty-six authors wrote contributions on individual settlements. They come from different institu­tions – both universities, research institutes, museums, the Institute for the Protection of Cultural Heritage of Slovenia – and from the ranks of independent research­ers, and present twenty settlements of different types and degrees of investigation in separate chapters. The authors are either those with the most in-depth knowledge on a specific site or those who actively investigated it. For the purposes of this book, they examined the excava­tion records and reports, the recovered small finds and different earlier publications, and presented their results in a comprehensive and clear manner roughly following dobe: v topografijo, infrastrukturo, stavbe, grobišča in premične ostanke posebnega pomena. Sledijo podatki o statusu naselja, o družbenem položaju posameznih prebivalcev, njihovih administrativnih ali vojaških funk-cijah, poklicih in etnični pripadnosti. Poglavje zaokroža oris zgodovinskega razvoja naselja. Želimo si, da bi delo omogočilo primerjave med posameznimi naselji, da se bodo razjasnila nekatera vprašanja o njihovem gospodarskem in družbenem pomenu ter o njihovi vlogi v poselitveni sliki širšega prostora med Jadranom in Donavo. Predvsem pa si želimo, da bi spodbudilo nove analize in objave gradiva ter bilo z njimi kmalu preseženo. Radi bi se zahvalili mnogim. Najprej Francisu Tassauxu iz Bordeauxa, duši združenja Adriaticum mare, za prijazno spodbudo. Poleg obeh glavnih recen­zentov, Janeza Dularja in Ivana Šprajca, so h kakovosti posameznih besedil mnogo pripomogli številni kolegi, s katerimi so se posvetovali avtorji ali uredniki. Mateja Belak, Andreja Dolenc Vičič, Dragotin Valoh in Tamara Korošec (vsi ZRC SAZU – Inštitut za arheologijo) so poskrbeli za enotno podobo knjige. Prevodi v angle-ščino so delo Mete Osredkar, Andreje Maver, Gregorja Pobežina in Lucije Jelenko. Jana Horvat, Irena Lazar, Andrej Gaspari a common concept. The information they obtained is located in space, illustrated with maps and plans, clearly conveyed and properly evaluated. The presentation of each settlement opens with its location and name in Antiquity, possible habitation traces from prehistory, mentions in ancient literary texts and documents, and recovered epigraphic evidence. The next, main part offers an overview of the archaeological remains from the Roman period: topography, infrastructure, build­ings, cemeteries and portable remains of particular significance. This is followed by the information on the status of a settlement, social standing of its inhabit­ants, their administrative or military functions, as well as professional or ethnical background. All is brought together in an outline of the historical development of each settlement. It is our wish and aim that the information in this book enables and incites comparisons between individual sites in order to shed light on certain issues pertaining to their economic and social role in the set­tlement of the wider area between the Adriatic and the Danube. Even more importantly, it is our wish that the book would serve as a stepping stone for further research and better knowledge on the subject. The book is a result of a concerted effort of authors and numerous other individuals. First of all, our thanks go to Francis Tassaux from Bordeaux, the soul of the Adriaticum Mare Association, for his kind incentive. In addition to the two main peer reviewers, Janez Du-lar and Ivan Šprajc, numerous colleagues shared their knowledge and offered professional advice to both the authors and the editors. Mateja Belak, Andreja Dolenc Vičič, Dragotin Valoh and Tamara Korošec (all ZRC SAZU – Institute of Archaeology) made sure that the book of numerous contributions functions as a whole. Meta Osredkar, Andreja Maver, Gregor Pobežin and Lucija Jelenko translated the contributions into English. Jana Horvat, Irena Lazar, Andrej Gaspari Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 9–46 FLUVIO FRIGIDO, CASTRA – AJDOVŠČINA Tina ŽERJAL, Vesna TRATNIK Izvleček Nastanek rimske naselbine v današnji Ajdovščini, postaje ob itinerarski cesti Akvileja–Emona, postavljamo v 1. st. pr. n. št. V Antoninskem itinerariju in na Tabuli Peutingeriani je v 3. st. n. št. poimenovana Fluvio Frigido, na Jeruzalem­skem itinerariju je v 4. st. vpisana kot mutatio Castra. Raziskave so pokazale, da rimska poselitev v 1. in 2. st. ni bila gosto strnjena le v srednjeveškem jedru Ajdovščine, ampak je obsegala še dobršen del okolice. Nagrobni spomeniki iz tega časa omenjajo rimske državljane, osvobojence ali predstavnike višjega sloja, staroselske prebivalce, ki so pridobili državljanstvo ter imeli pomembne administrativne funkcije. V poznem 3. st. je bilo ob sotočju Hublja in Lokavščka zgrajeno obzidje s stolpi – Castra. Utrdba je imela pomembno vlogo v sklopu zapor Claustra Alpium Iuliarum. Prisotnost vojaške posadke potrjujejo posamične najdbe. Življenje v utrdbi in njeni okolici se je vsaj v skromni obliki nadaljevalo tudi v drugi polovici 5. st. ter z zgodnje srednjeveško poselitvijo še v 6. ali 7. st. Ključne besede: Italija (10. regija), Ajdovščina, Fluvio Frigido, Castra, rimska doba, naselbina, utrdba, mansio/statio, Claustra Alpium Iuliarum Abstract The Roman settlement in the modern town of Ajdovščina stood at the main road connecting Aquileia and Emona and is believed to have been established in the 1st century BC. It is called Fluvio Frigido in the 3rd-century Antonine Itinerary and Tabula Peutingeriana, while the Jerusalem Itinerary from the 4th century marks it as mutatio Castra. Archaeological investigations have shown that the Roman settlement of the 1st and 2nd centuries extended across the medieval centre of Ajdovščina and a good part of the surrounding area. Funerary monuments mention Roman citizens, freedmen and members of higher classes, as well as individuals of indigenous origin who acquired Roman citizenship and held various administrative offices. In the late 3rd century, a fortification wall with towers – Castra – was constructed at the confluence of the Hubelj and Lokavšček streams. This fortress played an important role within the Claustra Alpium Iuliarum bar­rier system and the recovered small finds confirm the presence of an army garrison. Life in the fortress and its vicinity continued, albeit to a diminished extent, to the second half of the 5th century, with early medieval habitations even in the 6th and 7th centuries. Keywords: Italy (Regio X), Ajdovščina, Fluvio Frigido, Castra, Roman period, settlement, fortress, mansio/statio, Claustra Alpium Iuliarum https://doi.org/10.3986/9789610502586_01 Sl. 1: Panoramski pogled na Ajdovščino in vzhodno obzidje poznorimske utrdbe. Fig. 1: Panoramic view of the modern town of Ajdovščina and the eastern fortification wall of the Late Roman fortress. (Foto / Photo: Leo Caharija, Primorske novice) GEOGRAFSKI IN HISTORIČNI ORIS GEOGRAPHIC AND HISTORICAL OUTLINE Ajdovščina (sl. 1) leži na vzhodnem delu rodovitne Vipavske doline, pomembnim prehodom od morja in Ajdovščina (Fig. 1) lies at the eastern part of the furlanske ravnice do jugovzhodnega alpskega sveta. Na fertile valley of the River Vipava, which is an important tem delu se Alpe znižajo in nudijo ugodne naravne po-route connecting the Adriatic and the Friuli Plain with vezave proti nekdanjima provincama Noriku in Panoniji. the south-eastern Alps. It is also the area where the Alps are lowest and allow a relatively easy passage towards the Roman provinces of Noricum and Pannonia. Toponim današnjega kraja je izpeljanka iz besede “ajd” v pomenu pogan, velikan.1 Slovanski prišleki so verjetno antične ruševine pripisali mitičnim bitjem Ajdom. Poimenovanja v drugih jezikih so izpeljanke slovenskega imena naselja: italijansko Aidussina, nemško Heidenschaft. Rimska država je ta prostor, ki so ga poseljevala karnijska ljudstva,2 osvojila v zadnjih desetletjih 2. st. pr. n. št. Najverjetneje je tako vsaj že v prvih dveh desetletjih 1. st. pr. n. št. Vipavska dolina je pripadala provinci Galiji Cisalpini, fiskalno in administrativno pa upravnemu teri­toriju latinske kolonije Aquileia. Provinca Galija Cisalpina je bila leta 42 pr. n. št. priključena Italiji.3 Pod Avgustom je akvilejski ager upravno sodil v Regio X Italije in kasneje 4 v poznorimsko provinco Venetia et Histria. Rimsko naselje v Ajdovščini se je razprostiralo v bližini sotočja Lokavščka in Hublja. Po dolini so vodile karavanske poti (prazgodovinska jantarna pot) iz Akvi­leje preko Okre v smeri Navporta.5 V avgustejskem obdobju so zgradili krak ceste preko Julijskih Alp.6 Odsek preko prelaza Hrušica (Ad Pirum) je potovanje do Emone skrajšal za cel dan. Cestna postaja v Ajdovščini je bila tako zadnja pred strmim vzponom proti prelazu. ZGODOVINA RAZISKAV Mogočne ruševine poznorimske trdnjave so vzbu­dile zanimanje starinoslovcev že v 17. stoletju. Leta 1624 je Philip Clüvert Cluverius ob opisu poteka rimske ceste postajo Castra umestil ob prehod čez reko Frigidus in posredno v Vipavsko dolino. Martin Baučer (l. 1663) je prvi opisal ostanke utrdbe in jih pripisal antični Hemo­ni.7 Johan Ludwig Schönleben (1681) je povzel njegov 1 S. v. ajd, Slovar Slovenskega knjižnega jezika 1, Ljubljana 1970, 17; Svoljšak et al. 2013, 7. 2 O etnični pripadnosti staroselskega prebivalstva Vipav­ske doline lahko sklepamo na osnovi več virov, ki kažejo na področje mešanja različnih skupin Venetov, Karnov, morda Histrov in Japodov; omembe Karnijskih Alp, Venetskih Alp in Panonskih Alp kot predhodnic Julijskih Alp; ostankov venetskega jezika; gentilnih imen severnojadranskega jezi­kovnega substrata (po Alföldy 1978; 1999 ipd.). Glej obširno literaturo v Zaccaria 2007b, 243–244; 2009a-b. 3 O vključitvi ozemlja v Galijo Cisalpino glej Šašel 1975– 1976; 1984, 802; Zaccaria 2007a, 129; 2009a-b; Horvat, Bav­dek 2009. 4 O mejah tržaškega agra: Zaccaria 1992, 152, 163–164. O mejah akvilejskega agra: Zaccaria 2007a; Šašel Kos 2002a (za Ajdovščino posebej str. 378, 382); Šašel Kos 2002b. 5 Strabon, Geografija (IV 6, 10; VII 5, 2). Zbrani in pre­vedeni odlomki: Horvat, Bavdek 2009, 19-22. Rufius Festus, Brev. 7; Šašel 1975–76, 604, 612; Bosio 1991, 209–210; Vedaldi Iasbez 1994, 72–73. 7 Pri tem je mišljena Emona. The toponym of the modern town derives from the Slavic word ajd meaning a pagan or a giant.1 It is likely that when the Slavs settled the area, the ruins of Roman buildings were still well visible, making a powerful im­pression, as if made by the mythical creatures the Slavs called Ajdi. The names for the modern town in other languages, Aidussina in Italian and Heidenschaft in Ger­man, are derived from the Slovenian name. The Romans conquered the area, inhabited by the Carni people,2 in the final decades of the 2nd century BC. The Vipava Valley came under Cisalpine Gaul, most probably already in the first two decades of the 1st century BC, forming part of the territory of the Latin colony of Aquileia in terms of tax collection and administration. In 42 BC, Cisalpine Gaul became an integral part of Italy.3 Under Augustus, Aquileian territory belonged to Italy’s regio X, in the Late Roman period to the province of Venetia et Histria.4 The Roman settlement at Ajdovščina was located at the confluence of the Hubelj and Lokavšček streams, along the caravan routes (including the ancient Amber Route) leading from Aquileia across Ocra towards Nauportus.5 In the Augustan period, a new section of the road traversing the Julian Alps6 was constructed. The road across the Hrušica (Ad Pirum) Pass shortened the jour­ney from Aquileia to Emona by one whole day. The road station in Ajdovščina was the last stop before the steep ascent towards the pass. HISTORY OF RESEARCH The ruins of the Late Roman fortress have attracted the attention of antiquities enthusiasts and scholars from as early as the 17th century onwards. In 1624, Philip Clüvert Cluverius described the route of the Roman roads 1 S. v. ajd, Slovar Slovenskega knjižnega jezika 1, Ljubljana 1970, 17; Svoljšak et al. 2013, 7. 2 The mention of Carnian, Venetic and Pannonian Alps as predecessors to the Julian Alps, vestiges of the Venetic lan­guage, as well as the gentilicia of the North Adriatic linguis­tic substratum (according to Alföldy 1978; 1999 and others) suggest that the Vipava Valley was an area where different groups of the Veneti, Carni, possibly Histri and Japodes mixed. See extensive literature on the subject in Zaccaria 2007b, 243–244; 2009a-b. 3 For the incorporation of the area into Cisalpine Gaul, see Šašel 1975–76; 1984, 802; Zaccaria 2007a, 129; 2009a-b; Horvat, Bavdek 2009. 4 For the borders of the territory of Tergeste, see Zac­caria 1992, 152, 163–164; for the borders of the territory of Aquileia, see Zaccaria 2007a; Šašel Kos 2002a, in connection with Ajdovščina in particular pp. 378, 382; Šašel Kos 2002b. 5 Strabo, Geography (IV 6, 10; VII 5, 2). For collected passages translated into Slovenian, see Horvat, Bavdek 2009, 19-22. 6 Rufius Festus, Brev. 7; Šašel 1975–76, 604, 612; Bosio 1991, 209–210; Vedaldi Iasbez 1994, 72–73. opis, Janez Vajkard Valvasor (1689, knjiga 11, 272) pa si je Ajdovščino sam ogledal in podal svoja opažanja.8 S preučevanjem antičnih virov, itinerarijev in ceste med Akvilejo in Emono so raziskovalci v 19. stoletju povezali cestno postajo Fluvio Frigido in mutatio Castra z ostanki poznorimske Ajdovščine ter fluvius Frigidus s Hubljem in Vipavo. Prve upodobitve talne zasnove utrdbe so bile shematske. Peter Hitzinger9 je domneval pravilen kvadrat s šestnajstimi stolpi. Alfons Müllner je izčrpno popisal rimske ostaline v Ajdovščini, tako ostanke utrdbe kakor drobne najdbe. Opisal je tudiplato s prazgodovinsko naselbino Gradišče (Školj) in ledino Mirce, kamor je umestil zgodnjerimsko postajo Fluvio Frigido.10 O še vidnih ostankih utrdbe in stolpov so poročali tudi Paolo de Bizzarro, konservator c. k. komisije za spomeniško varstvo na Dunaju za Goriško (1888) in raziskovalci zapornega sistema Claustra Al-pium Iuliarum: Pietro Kandler (1849, 1863, 1871, 1905), konservator c. k. komisije za spomeniško varstvo na Dunaju za Istro/Primorje, Alberto Puschi (1901; 1902), Pietro Sticotti (1934; 1937), Alessandro Stucchi (1946).11 Potek javne ceste med Akvilejo in Emono skozi Ajdovščino so raziskovali: Anton von Premerstein in Simon Rutar (1899), Albert Hermann Otto Cuntz (1902) in Puschi (1903, 1905; njegove beležke je objavil Stucchi 1948). O rimskih najdbah na Mircah je pisal Rutar (1895), o Gradišču nad Ajdovščino Carlo Mar-chesetti (1903).12 Med prvo svetovno vojno (leta 1913 in 1916) je prva arheološka izkopavanja v mestu vodil Walter Schmid, s pomočjo Karla Picka. Svoje izsledke je objavil v poročilu skupaj z grafično predlogo rimskega tabora nepravilne ovalne oblike s trinajstimi stolpi. Označil je tudi obrambni jarek in vhod ob zahodni strani utrdbe, lokacijo najdbe miljnika in nekaterih grobov na zaho­dnem in vzhodnem grobišču.13 Po prvi svetovni vojni je leta 1920 v Ajdovščini sondirala italijanska vojska pod vodstvom I. Garibaldija, ki ni poznal Schmidovih zaključkov in je ponovno do-mneval pravilno kvadratno zasnovo obzidja.14 Natanč­neje je topografijo rimske Ajdovščine povzel Stucchi.15 8 Obširni pregledi v Šašel 1970; Petru 1971, 97–99; 1975b; Osmuk, Svoljšak, Žbona Trkman 1994, 15–21; Svoljšak et al. 2013, 21–22 (z obsežno razlago dela starejših raziskovalcev in pomembnimi reprodukcijami najstarejših kartografskih podob). 9 Hitzinger 1855; 1861, 47. 10 Müllner 1879, 131-133; 1889 (s karto in prvo neorto­gonalno risbo obzidja). 11 Natančneje o samih prispevkih teh avtorjev s citati beri pri: Šašel 1970; Petru 1971, 97–99; Šašel, Petru 1971, 50–52. 12 Marchesetti 1903, 91. 13 Glavno delo Pick, Schmid 1922–1924 s karto (repro­dukcija karte Svoljšak et al. 2013, 26), ostali prispevki Schmid 1922–1924; 1923–1924; Pick, Schmid 1916. 14 Stucchi 1946, 33–34; Svoljšak et al. 2013, 28. 15 Stucchi 1946. and placed the road station of Castra at the river cross­ing of the Frigidus, i.e. in the Vipava Valley. The first to describe the remains of the fortress was Martin Baučer in 1663, attributing them to the ancient town of Hemona.7 Johan Ludwig Schönleben (1681) reiterated his descrip­tion, while Johann Weichard Valvasor (1689) inspected Ajdovščina personally.8 The study of ancient texts, itineraries and the road connecting Aquileia and Emona led researchers in the 19th century to attribute the remains of the Late Roman Ajdovščina to the road station of Fluvio Frigido and mu-tatio Castra, and to identify fluvius Frigidus as the streams of Hubelj and Vipava. The first plans of the fortress were schematic. Peter Hitzinger,9 for example, presumed a regular square plan with sixteen towers. Alfons Müllner described the Roman remains at Ajdovščina in detail, both the architectural remains of the fortress and the recovered artefacts. He also described the plateau hostingthe prehistoric hillfort at Gradišče (Školj) and the Mirce site, the latter seen as the Early Roman Fluvio Frigido station.10 Others reported on the still visible remains of the fortress with its towers: Paolo de Bizzarro, conserva­tor of the K. K. Zentralkommission für Denkmalpflege in Vienna for the Gorizia area (1888), and researchers of the Claustra Alpium Iuliarum barrier system: Pietro Kandler (1849, 1863, 1871, 1905), conservator of the K. K. Zentralkommission für Denkmalpflege in Vienna forthe littoral area (Österreichisches Küstenland), Alberto Puschi (1901; 1902), Pietro Sticotti (1934; 1937) and Alessandro Stucchi (1946).11 Anton von Premerstein and Simon Rutar (1899), Al­bert Hermann Otto Cuntz (1902) and Puschi (1903, 1905; his notes published in Stucchi 1948) investigated the main road between Aquileia and Emona through Ajdovščina. Rutar (1895) wrote of the Roman finds from Mirce, Carlo Marchesetti (1903) of Gradišče above Ajdovščina.12 The first archaeological excavations took place dur­ing WWI (in 1913 and 1916), led by Walter Schmid with the help of Karl Pick. Schmid published his findings in a report accompanied by a plan of the Roman fort of an irregularly oval layout with thirteen towers. On the plan, he marked the defensive ditch and the entrance in the west side, the findspot of a milestone and several graves in the western and eastern cemeteries.13 7 Meaning Emona. 8 More extensive overviews in Šašel 1970; Petru 1971, 97–99; 1975b; Osmuk, Svoljšak, Žbona Trkman 1994, 15–21; Svoljšak et al. 2013, 21–22: with a detailed discussion of the work of previous researchers and with important reproduc­tions of the earliest cartographic records. 9 Hitzinger 1855; 1861, 47. 10 Müllner 1879, 131-133; 1889: with a map and the first non-orthogonal drawing of the fortification walls. 11 For more details on the contributions of these authors, see Šašel 1970; Petru 1971, 97–99; Šašel, Petru 1971, 50–52. 12 Marchesetti 1903, 91. 13 The main publication in Pick, Schmid 1922–1924 with Carlo Gregorutti in Sticotti sta objavila glavne epigrafske spomenike.16 Pomembne podatke najdemo tudi pri zapisih uči­telja Pavla Plesničarja, ki je v Ajdovščini služboval med obema vojnama.17 Stare predmete iz Ajdovščine je v sredini prejšnjega stoletja zbiral in hranil domačin StipeŠtekar, ki je bil prisoten pri mnogih odkritjih grobov in rimskih zidov.18 V drugi polovici 20. stoletja je obzidje v Ajdovščini raziskoval Peter Petru (Zavod RS za spomeniško var-stvo), ki je deloma izboljšal Schmidov načrt in poudaril večkotno obliko utrdbe. Skupaj z Dragom Svoljšakom (Goriški muzej) sta v šestdesetih letih izkopavala na več lokacijah znotraj srednjeveškega mesta.19 V tem obdo­bju so nastali tudi temeljni pregledni enciklopedični prispevki o rimski Ajdovščini.20 Jaroslav Šašel je podal več člankov o rimskih cestah in rimski zgodovini jugo­vzhodnoalpskega prostora na osnovi antičnih literarnih virov, v katerih je omenil Ajdovščino in izpostavil neka­tere njene epigrafske spomenike.21 Skupaj s Petrujem sta raziskovala pomen poznorimskega zapornega sistema 22 Claustra Alpium Iuliarum. Ajdovščino je od sedemdesetih let dalje sistematič­no raziskovala arheologinja Nada Osmuk (ZVKDS OE Nova Gorica), ob posameznih priložnostih tudi BeatričeŽbona Trkman (Goriški muzej). Načrtno konserva­torsko delo in sondažne raziskave na odsekih obzidja, spremljanje izkopov za gradbena dela in posamezna izkopavanja so prispevali k bolj jasnemu vpogledu v tlorisno zasnovo utrdbe. Končno je bil razjasnjen tudi potek obzidja ob južni stranici utrdbe in raziskana je bila stavba s termami, del zahodnega grobišča23 in prostor izven obzidja.24 Novejša sondiranja so povečini objavlje­na v reviji Varstvo spomenikov.25 V letih 2018 in 2019 16 Gregorutti 1892; Sticotti, Inscr. It. X 4 – vključene v teritorij mesta Tergeste; Sticotti 1906; 1908. 17 Plesničar 1998 (ponatis besedil P. Plesničar, Kulturni odmevi 1958 in 1959). 18 Štekar 1960–1961, 200; Žbona Trkman 1990a. 19 Petru 1960–1961; 1965; 1969; 1971; 1972; 1974; 1975b; 1976; Petru, Svoljšak 1965a–b; 1980; Svoljšak 1968; 1968– 1969a–b. D. Svoljšak se je bolj posvetil prazgodovinskim najdbam in Gradišču (Svoljšak 1965; 1967; 1988–1989) ter zgodnjesrednjeveškim odkritjem (Svoljšak 1970–1971; Svolj­šak, Knific 1976). 20 Kastelic 1987, 28-29; Šašel 1970; Petru 1971, 97-99; 1975b. 21 Šašel 1970; 1971; 1973; 1974; 1975-1976, 77, 90; 1980; 1984. 22 Šašel, Petru 1971, 50–52; Petru 1975a. 23 Osmuk 1977; 1979; 1986a–b; 1987a–b; 1988; 1990a–b; 1991; 1992; 1993a–c; 1996; 1999a–d; 2003; 2005a–c; Žbona Trkman 1982; 1986; 1987; 1990a–b; 1993; Svoljšak, Žbona Trkman 1986; s preglednimi deli Osmuk 1991; 1997; Osmuk, Svoljšak, Žbona Trkman 1994; Svoljšak et al. 2013. 24 Tratnik, Žerjal 2017. 25 Tratnik 2007, 15–17; Fabec, Tratnik 2009, 19–21; Fa-bec 2017. Strnjen pregled in seznam vseh raziskav po dru- In 1920, the Italian army conducted trial trenching here under Italo Garibaldi, who was unfamiliar with Schmid’s findings and presumed a regularly square layout of the fortification wall.14 Stucchi published a more detailed topography of Roman Ajdovščina,15 Carlo Gregorutti and Sticotti published the main epigraphic monuments.16 Important information can also be found in the notes of Pavel Plesničar, a teacher who worked at Ajdovščina between the two world wars.17 Stipe Štekar was a local who collected and kept antiquities from Ajdovščina around the middle of the 20th century, he was also present at numer­ous discoveries of Roman graves and walls.18 Peter Petru (Slovenian institute for heritage protec­tion) investigated the fortification wall in the second half of the 20th century and made improvements on Schmid’s plan, emphasising the polygonal form of the fort. In col­laboration with Drago Svoljšak (Regional Museum of Gorica), they excavated at several sites within the medi­eval town in the 1960s.19 Encyclopaedic articles on Roman Ajdovščina were written in this period.20 Jaroslav Šašel wrote several articles on the Roman roads and history in the south-eastern Alpine region based on ancient texts, in which he also touched upon Ajdovščina and described some of the inscriptions from the area.21 Together with Petru, they investigated the significance of the Late Roman Claustra Alpium Iuliarum barrier system.22 From the 1970s onwards, Nada Osmuk (Institute for the Protection of Cultural Heritage of Slovenia, Regional Office Nova Gorica, hereinafter ZVKDS) sys­tematically investigated Ajdovščina, occasionally alsoBeatriče Žbona Trkman (Regional Museum of Gorica). Planned conservation work and trial trenching at some sections of the fortification wall, watching briefs during construction works and several excavations have of­fered additional insight into the layout of the fort. They also reliably identified the fortification wall in the south, a map (reproduced in Svoljšak et al. 2013, 26), other con­tributions in Schmid 1922–1924; 1923–1924; Pick, Schmid 1916. 14 Stucchi 1946, 33–34; Svoljšak et al. 2013, 28. 15 Stucchi 1946. 16 Gregorutti 1892; Sticotti, Inscr. It. X 4 - included into the territory of Tergeste; Sticotti 1906; 1908. 17 Plesničar 1998 (reprint of Plesničar, Kulturni odmevi 1958 and 1959). 18 Štekar 1960–1961, 200, Žbona Trkman 1990a. 19 Petru 1960–1961; 1965; 1969; 1971; 1972; 1974; 1975b; 1976; Petru, Svoljšak 1965a-b; 1980; Svoljšak 1968; 1968– 1969a-b. Svoljšak dedicated more attention to the prehistoric finds and to Gradišče (Svoljšak 1965; 1967; 1988-1989); and also to the early medieval finds (Svoljšak 1970–71; Svoljšak, Knific 1976.). 20 Kastelic 1987, 28-29; Šašel 1970; Petru 1971, 97-99; 1975b. 21 Šašel 1970; 1971; 1973; 1974; 1975-1976, 77, 90; 1980; 1984. 22 Šašel, Petru 1971, 50–52; Petru 1975a. so potekale obsežne raziskave v središču Ajdovščine, ki so odkrile pomembne nove stavbe in strukture.26 Z numizmatičnimi najdbami in pomenom naselja Castra v poznorimskem obdobju se je ukvarjal Peter Kos.27 Antične vire je zbral Rajko Bratož,28 epigrafske spomenike in romanizacijo prostora sta obravnavalaMarjeta Šašel Kos29 in Claudio Zaccaria,30 amfore in uvoženo posodje Verena Vidrih Perko.31 V novejšem obdobju so bile izpeljane študije rimskodobne Ajdo­vščine in njene vpetosti v severnojadranski prostor, od romanizacije dalje.32 Pomembna so tudi dela poljudne narave, ki podajajo širšo sliko družbenega konteksta in zadnje ugotovitve raziskovalcev.33 V širših pregledih rimskega in poznorimskega obdobja v Sloveniji se pov­zemajo glavna vedenja o rimski naselbini v Ajdovščini.34 PRAZGODOVINA Na dobro naravno zavarovanem skalnem platoju, severno nad Ajdovščino, se razprostira prazgodovinska naselbina (kaštelir), poimenovana Gradišče ali Školj (232 m). Na severu je bil plato utrjen s suho grajenim obrambnim zidom.35 V sredini naselbine je obsežen kal, domnevno naravni vodni zbiralnik, kjer so med arhe­ološkimi raziskavami odkrili odlomke t. i. kaštelirske keramike.36 Naselbina je datirana v pozno bronasto in starejšo železno dobo.37 gi svetovni vojni do leta 2013: Svoljšak et al. 2013, 82–84. Zgodnjeslovansko grobišče v Šturjah, s posameznimi rimski-mi najdbami: Brezigar, Josipovič 2015; Brezigar 2017; 2018. 26 Rozman, Urek, Kovačič 2019. Podrobneje Urek, Kova- čič 2020, v tej knjigi.27 Kos 1986; 1997; 2012; 2014a–c; 2017. 28 Bratož 2000, 29-31; 2014; 2018. 29 Šašel Kos 1997a, št. 176; 2002a, 378; 2004, št. 176. 30 Zaccaria 2007a–b; 2009a–b; 1992, 164, 235–236, 238– 239. 31 Vidrih Perko 1994, 88–96; 1997; 2000, 442. 32 Vidrih Perko, Žbona Trkman 2003–2004; 2004; 2005. 33 Osmuk, Svoljšak, Žbona Trkman 1994; Svoljšak et al. 2013. 34 Vuga 1993a–b; Ciglenečki 1997, 193, 197 s poudarkom na trasi cest v okolici Ajdovščine in poznorimskih točkah ob njih; Ciglenečki 1999a, 290–292; 1999b, 318–319; 2015, 394–395, 402–403; Horvat 1999, 225–226, 232; Šašel Kos 1999a, 262. Poljudni pregled arheologije Vipavske doline: Vuga 1984; Bratina 2013. 35 Müllner 1879, 130 (= 1996, 150–151); Marchesetti 1903, 91; Petru 1975b, 121; nekateri raziskovalci sicer omen-jajo tri stranice obzidja – npr. Petru 1965, 134. 36 Svoljšak 1965 (rekognosciranje Goriškega muzeja leta 1965); 1967 (sondiranje Goriškega muzeja leta 1966). 37 Petru 1975b, 121; Svoljšak 1988–1989, 367-386, po­sebej 376, št. 13; Mlinar 1999, 22-30; Bratina 2010, 167; na­zadnje Svoljšak et al. 2013, 7-8. unearthed a building with a bath complex, part of the west cemetery23 and the extra muros area.24 Recent trial trenching results have for the most part been published in the Varstvo spomenikov journal.25 The last and extensive investigations in the centre of Ajdovščina took place in 2018 and 2019, unearthing important new buildings and structures from the Roman period.26 Peter Kos studied the numismatic evidence and significance of Castra in the Late Roman period.27 Rajko Bratož collected the literary evidence on the subject.28 Marjeta Šašel Kos29 and Claudio Zaccaria30 studied the epi-graphic evidence and the Romanisation of the area. Verena Vidrih Perko analysed the amphorae and other imported pottery goods.31 Recently, studies have been published on the Roman-period Ajdovščina and its significance within the North Adriatic area from the Romanisation process onwards.32 There is a number of important publications in­tended for the general public, which provide a wider social context and the most recent findings of the experts.33 The basic knowledge of the Roman settlement in Ajdovščina is also given in the overviews of the Roman and Late Roman periods in Slovenia.34 PREHISTORY The rocky plateau north of Ajdovščina, calledGradišče or Školj (232 m asl), holds a prehistoric hillfort. It is naturally well-protected with an additional drywall fortification on the north side.35 It had a vast water hole 23 Osmuk 1977; 1979; 1986a–b; 1987a–b; 1988; 1990a–b; 1991; 1992; 1993a–c; 1996; 1999a–d; 2003; 2005a–c; Žbona Trkman 1982; 1986; 1987; 1990a–b; 1993; Svoljšak, Žbona Trkman 1986; owerviews in Osmuk 1991; 1997; Osmuk, Svoljšak, Žbona Trkman 1994; Svoljšak et al. 2013. 24 Tratnik, Žerjal 2017. 25 Tratnik 2007, 15–17; Fabec, Tratnik 2009, 19–21; Fa-bec 2017. For a concise overview and list of all investigations after WWII up to 2013, see Svoljšak et al. 2013, 82–84. For the Early Slavic cemetery at Šturje, with stray Roman finds, see Brezigar, Josipovič 2015; Brezigar 2017; 2018. 26 Rozman, Urek, Kovačič 2019. For more details, see Urek, Kovačič 2020, in this book. 27 Kos 1986; 1997; 2012; 2014a-c; 2017. 28 Bratož 2000, 29-31; 2014; 2018. 29 Šašel Kos 1997a, No. 176; 2002a, 378; 2004, No. 176. 30 Zaccaria 2007a-b; 2009a-b; 1992, 164, 235–236, 238– 239. 31 Vidrih Perko 1994, 88–96; 1997; 2000, 442. 32 Vidrih Perko, Žbona Trkman 2003–2004; 2004; 2005. 33 Osmuk, Svoljšak, Žbona Trkman 1994; Svoljšak et al. 2013. 34 Vuga 1993a-b; Ciglenečki 1997, 193, 197 with an emphasis on the roads in the Ajdovščina area and the Late Roman sites along them; Ciglenečki 1999a, 290–292; 1999b, 318–319; 2015, 394–395, 402–403; Horvat 1999, 225–226, 232; Šašel Kos 1999a, 262. For a general overview of the ar­chaeology in the Vipava Valley, see Vuga 1984; Bratina 2013. 35 Müllner 1879, 130 (= 1996, 150–151); Marchesetti Med arheološkimi izkopavanji v Ajdovščini so bili tako znotraj kot tudi izven poznorimske utrdbe najdeni posamezni odlomki prazgodovinske lončenine.38 ANTIČNI LITERARNI VIRI Rimska naselbina v Ajdovščini oz. cestna postaja Fluvio Frigido je omenjena v Antoninskem itinerariju in na Tabuli Peutingeriani v 3. st.39 V Antoninskem itinerariju je navedena razdalja 36 milj ali 54 kilomet­rov od Akvileje in 22 milj od Logatca (Longaticum). Na Tabuli Peutingeriani je omenjena le razdalja 15 milj do Hrušice (Ad Pirum), cestna postaja pa je zapisana ob istoimenski reki fluvius Frigidus, ki izvira pod gorovjem in se nato izliva pri Akvileji v jezero ali laguno. Tok reke je skiciran shematično, enači se z reko Vipavo. Potok Hubelj ob Ajdovščini, kot največji pritok Vipave, je bil dolgo časa razumljen za njen glavni izvir. V spodnjem toku je kot fluvius Frigidus verjetno upodobljen tudi spodnji tok reke Soče od sotočja z Vipavo do izliva v morje. Ime reke je izpisano z rdečo barvo.40 V Jeruzalemskem itinerariju iz 4. st. je vpisana postaja mutatio Castra, inde surgunt Alpes Iuliae, ad Pirum summas Alpes (kjer se prično vzpenjati Alpe, do kraja Ad Pirum na vrhu Alp) in sicer 12 milj od kraja Ad Fornulos oziroma 35 milj do Akvileje in 9 milj od Hrušice.41 To poimenovanje odraža dvojno, trgovsko in vojaško oz. obrambno funkcijo v poznorimskem obdobju utrjene naselbine.42 Na upodobitvi zapornega sistema tractus Italiae circa Alpes, v kopijah priročnika Notitia Dignitatum iz 5. stoletja je upodobljena trdnjava s stolpi, vhodom in stavbami v notranjosti, med katerimi je v ospredju vidna stavba s peristilom. Nekateri raziskovalci v njej vidijo 43 upodobitev trdnjave Castra. Pavel Diakon konec 8. stoletja opisuje v svojem delu o zgodovini Langobardov (Historia Langobar­dorum) bitko med Langobardi in Avari leta 664 pri kraju Flovius.44 38 Bratina 2010, 167; Tratnik, Žerjal 2017, 250-251 in Urek, Kovačič 2020. 39 It. Ant. 128.7; Tab. Peut. III, 5; Šašel 1970; Petru 1971; Šašel 1975, 75–78, 90–92; Petru 1975b; Bosio 1973; 1991, 201–211; Vedaldi Iasbez 1994, 444-446; Bratož 2000, 31; Svoljšak 2000; Calzolari 2000. 40 Vedaldi Iasbez 1994, 129-131; 2000, 303; Bratož 2000, 27-28; kar razloži izostanek omemb reke Soče v antični lite-raturi, o novih dognanjih glej Vedaldi Iasbez 2000, 303-309. 41 It. Burd. 560, 2; Šašel 1975, 74–87; 1971, 24; Petru 1971; Vedaldi Iasbez 1994, 444-446; Bratož 2000, 30–31; Svoljšak 2000; Calzolari 2000; Svoljšak et al. 2013, 15. 42 Šašel 1970; 1975, 90–91; Petru 1971; 1975b; Bosio 1973; 1991, 201–211; Vedaldi Iasbez 1994, 444-446. 43 Svoljšak et al. 2013, 16; Kos 2014a; drugačna interpre­tacija, da je upodobljena Akvileja: Bratož 2014, 194-197.44 Šašel 1971, 43; Svoljšak et al. 2013, 17; Osmuk, Svoljšak, Žbona Trkman 1994, 14. at the centre that was presumably used as a natural wa­ter cistern. In it, archaeologists unearthed sherds of the Castellieri pottery.36 The hillfort was inhabited in the Late Bronze and Early Iron Ages.37 Several sherds of prehistoric pottery were also found below the hill, during the archaeological excavations within the Late Roman fortress and outside it.38 LITERARY EVIDENCE The Roman settlement in Ajdovščina, i.e. the road station of Fluvio Frigido appears in the Antonine Itinerary and Tabula Peutingeriana, both dating to the 3rd centu­ry.39 In the former, it is shown at a distance of 36 Roman miles or 54 kilometres from Aquileia and 22 miles from Longaticum (Logatec). Tabula Peutingeriana only gives the distance of 15 miles to Ad Pirum (Hrušica), while the station is marked beside the river of the same name, fluvius Frigidus, which is depicted as springing at the foot of the adjacent mountain and flowing into a lake or lagoon at Aquileia. The river is drawn in a very schematic line and is identified with the River Vipava. The stream of Hubelj, the largest tributary of the Vipava and flowing near Ajdovščina, was long seen as its main source. In its lower reaches, the fluvius Frigidus depiction probably incorporates the lower reaches of the Soča/Isonzo from its confluence with the Vipava to the Adriatic. The name of the river is written in red.40 The Jerusalem Itinerary from the 4th century fea­tures mutatio Castra, inde surgunt Alpes Iuliae, ad Pirum summas Alpes (where the Julian Alps begin to rise, Ad Pirum on the summit of the Alps), located 12 miles from Ad Fornulos, 35 miles from Aquileia and 9 miles from Hrušica.41 The name signifies a double, trading and 1903, 91; Petru 1975b, 121; some researchers mention three sides of the fortification walls, e.g. Petru 1965, 134. 36 Svoljšak 1965 (reconnaissance of the Regional Mu­seum of Gorica); 1967 (trial trenching of the Regional Mu­seum of Gorica in 1966). 37 Petru 1975b, 121; Svoljšak 1989, 367-386, particularly 376, No. 13; Mlinar 1999, 22-30; Bratina 2010, 167; most re­cently in Svoljšak et al. 2013, 7-8. 38 Bratina 2010, 167; Tratnik, Žerjal 2017, 250-251 and Urek, Kovačič 2020. 39 It. Ant. 128.7; Tab. Peut. III, 5; Šašel 1970; Petru 1971; Šašel 1975, 75–78, 90–92; Petru 1975b; Bosio 1973; 1991, 201–211; Vedaldi Iasbez 1994, 444-446; Bratož 2000, 31; Svoljšak 2000; Calzolari 2000. 40 Vedaldi Iasbez 1994, 129-131; 2000, 303; Bratož 2000, 27-28; which would explain the absence of mentioning the River Soča/Isonzo in ancient texts; for new findings on the subject, see Vedaldi Iasbez 2000, 303-309. 41 It. Burd. 560, 2; Šašel 1975, 74–87; 1971, 24; Petru 1971; Vedaldi Iasbez 1994, 444-446; Bratož 2000, 30–31; Svoljšak 2000; Calzolari 2000; Svoljšak et al. 2013, 15. Posredno je območje Ajdovščine opisano v šte­vilnih rimskih literarni virih, ki opisujejo veliko bitko med cesarjem Teodozijem I. in uzurpatorjem Evge­nijem iz leta 394, poimenovano bitka pri Frigidu.45 Sistem Claustra Alpium Iuliarum, katerega del naj bi bila tudi utrdba v Ajdovščini, je skupno omenjen v vsaj 40 virih.46 EPIGRAFSKI VIRI Epigrafsko gradivo, najdeno v 19. stoletju, je zbral C. Gregorutti,47 deloma je bilo objavljeno v petem zvezku Theodora Mommsena Corpus inscriptionum Latinarum (CIL V 715, 716) in prvem dodatku, kjer jih je med spomenike akvilejskega teritorija uvrstil Ettore Pais (Suppl. It. 58, 59, 60). P. Sticotti je večino vključil v objavo napisov iz tržaškega agra (Inscr. It. X 4, 339–344, 346, 379).48 Posamezne je upošteval Giovanni Battista Brusin med napisi akvilejskega teritorija (Inscr. Aqins.. 2785). Nekatere je ponovno pregledal in kritično pre­tehtal C. Zaccaria v zvezku Supplementa italica n. s. 10, kjer jih je izločil iz tržaškega teritorija.49 Med temi sta dva nagrobnika danes pogrešana (Katalog napisov št. 5 in 6), trije nagrobniki so hranjeni v lapidariju palače At-tems – Musei provinciali di Gorizia (Katalog napisov št. 1, 3 in 4), ostale hranijo v Goriškem muzeju Kromberk (Katalog napisov št. 2, 7, 8, 9, 10), miljnik pa v Narodnem muzeju Slovenije (Katalog napisov št. 11). Epigrafski spomeniki odkriti (tudi ponovno) po drugi svetovni vojni so bili vključeni v korpuse napisov iz Jugoslavije (ILJug 451, ILJug 1221a–c) ali objavljeni drugje (Katalog napisov št. 7).50 45 Vedaldi Iasbez 1994, 129-131; nazadnje Bratož 2014, 160-186 in Bratož 2018 z obsežno starejšo literaturo in razla-go posameznih virov. 46 Šašel 1971; Kos 2014a–c; Bratož 2014, 96–99, 127ss, 187–208. 47 Gregorutti 1892. 48 Pred tem jih je objavil v dveh člankih Sticotti 1906; 1908. 49 Zaccaria 1992, za meje tržaškega agra glej 164, 168, za posamezne spomenike 235–236, 238–239. 50 Natančneje glej Katalog napisov. Mlajši pregled epi­grafskih spomenikov Svoljšak et al. 2013, 12, 56–58, 63. military/defensive function of the settlement fortified in the Late Roman period.42 The depictions of the tractus Italiae circa Alpes barrier system in the copies of the Notitia Dignitatum manual from the 5th century show a fortress with towers, an entrance and buildings in the interior that include one with a peristyle in the foreground. Some scholars see this building as representing the Castra fortress.43 Towards the end of the 8th century, Paul the Deacon in his Historia Langobardorum writes of a battle between the Langobards and the Avars that took place in 664 near a place called Flovius.44 Indirectly, the Ajdovščina area appears in numer­ous Roman texts that describe the great battle fought in 394 between the Emperor Theodosius I and the usurper Eugenius, called the Battle of the Frigidus.45 The Claustra Alpium Iuliarum barrier system, of which the fortress in Ajdovščina formed part, is mentioned in at least forty different ancient texts.46 EPIGRAPHIC EVIDENCE Gregorutti collected the inscriptions that came to light in the 19th century.47 They were in part published in the fifth volume of Theodore Mommsen’s Corpus Inscriptionum Latinarum (CIL V 715, 716) and its first supplement, where Ettore Pais included them among the monuments from the territory of Aquileia (Suppl. It. 58, 59, 60). In contrast, Sticotti ascribed most of them to the territory of Tergeste (Inscr. It. X 4, 339–344, 346, 379).48 Giovanni Battista Brusin considered some inscriptions under the territory of Aquileia (Inscr. Aq. 2785). In Sup-plementa Italica n. s. 10, Zaccaria re-examined and criti­cally evaluated certain inscriptions, did not treat them as belonging to the territory of Tergeste.49 These include two tombstones now missing (Catalogue of inscriptions Nos. 5 and 6), three tombstones held in the stone collec­tion of the Palazzo Attems - Musei provinciali di Gorizia 42 Šašel 1970; 1975, 90–91; Petru 1971; 1975b; Bosio 1973; 1991, 201–211; Vedaldi Iasbez 1994, 444-446. 43 Svoljšak et al. 2013, 16; Kos 2014a; for a different inter­pretation, that the depiction actually represents Aquileia, see Bratož 2014, 194-197. 44 Šašel 1971, 43; Svoljšak et al. 2013, 17; Osmuk, Svoljšak, Žbona Trkman 1994, 14. 45 Vedaldi Iasbez 1994, 129-131; most recently in Bratož 2014, 160-186 and Bratož 2018 with an extensive list of ear­lier literature and a discussion on individual texts. 46 Šašel 1971; Kos 2014a-c; Bratož 2014, 96–99, 127ff, 187–208. 47 Gregorutti 1892. 48 He had previously published them in two articles: Sti­cotti 1906; 1908. 49 Zaccaria 1992, for the boundaries of the ager of Ter­geste, see pp. 164, 168, for individual monuments, see pp. 235–236, 238–239. CESTA Javna cesta via publica Aquileia–Emona51 naj bi sle­dila stari prazgodovinski karavanski “jantarni” poti preko Okre proti vzhodu, ki je opisana pri Strabonu.52 Rimski kolonizatorji so začetne odseke iz Akvileje domnevno začeli graditi kmalu po izgradnji ceste Via Annia (druga polovica 2. ali začetek 1. st. pr. n. št.),53 odsek prek Julijskih Alp mimo Hrušice (mansio Ad Pirum summas Alpes) pa so dogradili verjetno v avgustejskem obdobju.54 O poteku javne ceste mimo Ajdovščine je veliko domnev. Zaradi najdbe miljnika, lokacije zahodnega grobišča in domnevnega vhoda v trdnjavo, so jo nekateri enačili s srednjeveškim in novoveškim potekom glavne ceste iz srednjeveškega jedra Ajdovščine proti Gorici oz. Ljubljani.55 Puschi je domneval, da je cesta vodila sever-no od utrdbe ter ob sotočju zavila okoli obzidja proti jugu. Schmid in Stucchi sta to tezo ovrgla in domnevala potek javne ceste južno od utrdbe,56 kar je domneval tudi Petru in drugi za njim. Javna cesta mimo rimskega naselja naj bi potekala približno v smeri Goriške ceste, kar potrjujeta grobišče Mirce na zahodu in grobišče ob mostu čez Hubelj na vzhodu (sl. 2: 12 in 19).57 Šele konec 20. stoletja so bili ostanki javne ceste odkriti na dveh mestih.58 Na zahodni strani Ajdovščine, na ledini Na Gmajni (približno 1 km od obzidja), je bilo odkrito rahlo napeto prodnato cestišče, široko 4,2 m, z jarki za odvodnjavanje na obeh straneh.59 Trasa rimske ceste, ki je vodila preko sedla vzhodno od Vipavskega Križa, je bila potrjena še ob potoku Jevšček (okoli 2 km zahodno od obzidja).60 51 Šašel 1975, 75-78, pril. XII; Bosio, 1991, 201–211; Bratož 2000, 29-31; Šašel 1973; Šašel Kos 1997b, 24. 52 Strabon, Geografija IV 6, 10 in VII 5, 2; Šašel 1974; Šašel Kos 1990, 17-19; Vedaldi Iasbez 1994, 96-100; Horvat, Bavdek 2009, 21; Gaspari 2017, 127–128. 53 Bosio, 1991, 202; Vedaldi Iasbez 1994, 445-446 (z da­tacijo v konec 1. st. pr. n. št.); Bratož 2000, 29; Šašel 1973; Šašel Kos 1997b, 24. 54 Vedaldi Iasbez 1994, 72–73: izgradnja se datira v leto 10 n. št. na osnovi omembe Rufija Festa (glej opombo 6). 55 Takrat imenovana Goriška cesta, današnja Gregorči­čeva ulica. Glej franciscejski kataster – Svoljšak et al. 2013, 22–23. Osmuk 1997, 122. 56 Cuntz 1902, 150; Puschi 1902; 1903; Schmid 1923-1924, postavlja potek med Ustjem in Ajdovščino (250 m južno od utrdbe); Stucchi 1948, 22 (takoj južno od utrdbe). 57 Petru 1965, 134; Osmuk 1997, 122; Osmuk, Svoljšak, Žbona Trkman 1994, 8. 58 Izris poteka rimskih cest v okolici Ajdovščine: Svoljšak et al. 2013, 80. 59 Gradnja obvoznice leta 1998: Osmuk 1999e; 2003, k. o. Vipavski Križ, parc. št. 1070/152. Le 0,4 m pod današnjo po­vršino. 60 Gradnja obvoznice leta 1998: Osmuk 1999e, 7; 2003, k. o. Vipavski Križ, parc. št. 1024, 1026. (Catalogue of inscriptions Nos. 1, 3 and 4), others kept in the Goriški muzej Kromberk, museum collection in Ajdovščina (Catalogue of inscriptions Nos. 2, 8-10) and the milestone held in the Narodni muzej Slovenije (Catalogue of inscriptions No. 11). The inscriptions discovered or rediscovered after WWII were published in Inscriptiones Latinae Jugoslaviae (ILJug 451, ILJug 1221a-c) or elsewhere (Catalogue of inscription No. 7).50 ROMAN ROAD The via publica from Aquileia to Emona51 is believed to have traced the prehistoric Amber Route leading east­wards across the Ocra Pass, a description of which can be found in Strabo.52 The Romans presumably began building the initial sections from Aquileia soon after having finished the Via Annia (second half of the 2nd or early 1st century BC),53 while the section across the Julian Alps that led across Hrušica (mansio Ad Pirum summas Alpes) and avoided Ocra was constructed in probably the Augustan period.54 A number of hypotheses have been put forward in connection with the part of this public road that passed Ajdovščina. The findspot of a milestone, the location of the west cemetery and the presumed entrance to the for­tress led some to suggest the route was the same as that of the medieval and post-medieval main road leading from the centre of Ajdovščina towards Gorica on the one side and towards Ljubljana on the other.55 In contrast, Puschi presumed the Roman road led north of the fortress and around the fortification wall at the confluence to then run southwards. Schmid and Stucchi rejected this hypothesis and rather supposed it ran south of the fort,56 which is a supposition that Petru and others after him shared. The Roman road was believed to have run past the settlement 50 For more details, see the Catalogue of inscriptions. For a recent overview of the epigraphic monuments, see Svoljšak et al. 2013, 12, 56–58, 63. 51 Šašel 1975, 75-78, App. XII; Bosio, 1991, 201–211; Bratož 2000, 29-31; Šašel 1973; Šašel Kos 1997b, 24. 52 Strabo, Geography IV 6, 10; VII 5, 2; Šašel 1974; Šašel Kos 1990, 17-19; Vedaldi Iasbez 1994, 96-100; Horvat, Bavdek 2009, 21; Gaspari 2017, 127–128. 53 Bosio, 1991, 202; Vedaldi Iasbez 1994, 445-446 (with a dating to the end of the 1st century BC); Bratož 2000, 29; Šašel 1973; Šašel Kos 1997b, 24. 54 Vedaldi Iasbez 1994, 72–73: the construction is dated to AD 10 based on a note by Rufius Festus – see Fn. 6. 55 Then called Goriška cesta, now Gregorčičeva ulica. See the Franziscean cadastre – Svoljšak et al. 2013, 22–23; Osmuk 1997, 122. 56 Cuntz 1902, 150; Puschi 1902; 1903; Schmid 1923-24, who places it between Ustje and Ajdovščina: 250 m south of the fort; Stucchi 1948, 22: just south of the fort. Petru je domneval lokacijo mostu in prehoda čez reko južno od današnjega mostu čez Hubelj,61 najver­jetneje na osnovi lege južnega grobišča (sl. 2: 19), ki je prikazano na Schmidovi skici.62 Na območju Šturij, vzhodno od Hublja, se je verjetno itinerarska cesta razcepila na staro pot preko Okre63 in avgustejski odcep s strmim vzponom na Col in Hrušico.64 Ceste v naselju so bolje poznane šele iz novejših raziskav.65 Domneva se, da je bil vhod v poznorimsko trdnjavo približno v liniji današnje Gregorčičeve ulice ob 1. stolpu, poimenovanem Varda (sl. 2).66 V bližini je bil vzidan miljnik iz let 305/306 (Katalog napisov, št. 11). Njegova prvotna lokacija ni znana.67 Schmid je vhod narisal med 2. in 3. stolpom.68 Bližino ceste nakazuje tudi lega zahodnega grobišča. Na severnem delu Ceste 5. maja sta bili izkopani dve širši utrjeni nasutji, ki bi lahko bili del poti v naselju (sl. 2: 14).69 Vhodi v trdnjavo na drugih stranicah obzidja niso poznani.70 Med raziskovalci velja, da sta bili ob gradnji utrd-be strugi Lokavščka in Hublja regulirani.71 Pomena reke fluvius Frigidus kot prometne žile ne poznamo. Pomembnost reke se odraža v prevzemu imena za na­selbino in v upodobitvi reke na Tabuli Peutingeriani ter v omembah v virih.72 61 Petru 1965, 134. 62 Pod nekdanjim internatom. Pick, Schmid 1922–1924, 302, sl. 148. Reprodukcija: Svoljšak et al. 2013, 25.63 Horvat, Bavdek 2009. 64 Puschi 1903; Stucchi 1946, 29; Bosio 1991, 207–209; Šašel 1975, 75–78, pril. XII. Domnevni potek ceste in lokacija ostankov na Colu: Svoljšak et al. 2013, 80. 65 Med izkopavanji v letih 2017, 2018 in 2019 so bile od­krite tudi tlakovane ulice znotraj utrdbe, npr. tlakovana ulica, ki poteka od 1. stolpa proti vzhodu. Rozman, Urek, Kovačič 2019 in prispevek Urek, Kovačič 2020. 66 Osmuk 1997, 122. Omenjeno odkrito cestišče v letih 2017 in 2018 bi to potrjevalo – Rozman, Urek, Kovačič 2019, sl. 1; Urek, Kovačič 2020. 67 Katalog napisov št. 11. 68 Pick, Schmid 1922–1924 s karto. Z novejšimi raziska­ vami lokacija vhoda ni bila potrjena – Osmuk 1997, 122.69 Tratnik, Žerjal 2017, 250, 254. 70 Dolgo so domnevali, da so vrata v vzhodnem obzidju, ki so vodila na most proti Šturjam, rimska, toda izkazalo se je, da so srednjeveškega nastanka – Petru 1965, 138; Osmuk 1997, 122, idr. 71 Osmuk 1997, 121–122. 72 Vedaldi Iasbez 1994, 444-446. Vidrih Perko, Žbona Trkman 2004: izčrpno o domnevah rečnega prometa surovin in izdelkov iz Vipavske doline v Akvilejo. between the old main road towards Gorica and the railway, as indicated by the western and eastern cemeteries.57 The first physical evidence of the road came to light at the end of the 20th century, at two sites west of the Roman settlement.58 At the Na Gmajni site (roughly a kilometre from the fortification wall), a slightly cambered gravel road surface was unearthed, 4.2 m wide and flanked by drainage ditches on either side.59 Another section, across the saddle east of Vipavski Križ, was found near the Jevšček stream (some 2 km west of the fortification wall).60 Petru posited a bridge and river crossing south of the modern-day bridge across the Hubelj,61 most likely based on the location of the south cemetery (Fig. 2: 19) as shown in Schmid’s sketch.62 At Šturje east of the Hubelj, the road probably forked, with one leg continuing along the ancient road towards and across Ocra63 and, from the Augustan period on­wards, the other leg climbing steeply to Col and Hrušica.64 The roads and streets within the settlement are only better known through recent investigations.65 It is pre­sumed that the entrance to the Late Roman fortress stood roughly in the line of the modern street of Gregorčičeva ulica, at Tower 1, called Varda (Fig. 2).66 Found nearby was a milestone from 305/306 (Catalogue of inscriptions, No. 11); its original location is unknown.67 Schmid drew the entrance between Towers 2 and 3.68 The location of the west cemetery also points to a road in proximity. In the north part of the street of Cesta 5. maja, two wide and compact deposits came to light that might represent the remains of a paved street in the settlement (north of the 57 Petru 1965, 134; Osmuk 1997, 122; Osmuk, Svoljšak, Žbona Trkman 1994, 8. 58 For marked locations of the Roman roads in the Ajdovščina area, see Svoljšak et al. 2013, 80. 59 Bypass construction in 1998: Osmuk 1999e; 2003, Vipavski Križ cadastral municipality, Lot No. 1070/152; only 0.4 m under the current surface. 60 Bypass construction in 1998: Osmuk 1999e 7; 2003, Vipavski Križ cadastral municipality, Lot Nos. 1024, 1026.61 Petru 1965, 134. 62 Under the former boarding school. Pick, Schmid 1922–1924, 302, Fig. 148. Reproduction: Svoljšak et al. 2013, 25. 63 Horvat, Bavdek 2009. 64 Puschi 1903; Stucchi 1946, 29; Bosio 1991, 207–209; Šašel 1975, 75–78, App. XII; for the presumed location of the road and its remains at Col, see Svoljšak et al. 2013, 80. 65 Paved streets within the fortress were also unearthed during the 2017, 2018 and 2019 excavations, for example one leading eastwards from Tower 1. Rozman, Urek, Kovačič 2019 and the contribution by Urek, Kovačič 2020. 66 Osmuk 1997, 122. The road surface found in 2017 and 2018 confirmed it – Rozman, Urek, Kovačič 2019, Fig. 1; Urek, Kovačič 2020. 67 Catalogue of inscriptions No. 11. 68 Pick, Schmid 1922–1924 with map. The recent investi­gations have not corroborated the location of the entrance – Osmuk 1997, 122. NASELBINA ZGODNJERIMSKA POSELITEV – FLUVIO FRIGIDO Müllner je prvi opisal rimske ruševine na ledini Mirce ali na Mircah. Toponim verjetno izvira iz latinske besede murus. Takoj izven jugozahodnega obzidja in juž-no od današnje Gregorčičeve ulice naj bi se na travnikih in poljih raztezali ostanki antičnih zidov, rimske opeke in lončenine. Obsegali naj bi 20 ha veliko površino, v obliki kvadrata, s stranico dolžine 450 m. Ruševine je pripisal civilni naselbini ob utrdbi in zgodnjerimski cestni postaji (statio/mansio) Fluvio Frigido.73 Tudi na karti Walterja Schmida se Mirce raztezajo na jugozaho­dni strani obzidja. Na tem prostoru so bili z novejšimi raziskavami odkriti ostanki rimskih stavb in plasti.74 V osrednjem delu Ceste 5. maja je poznano večje poslopje ali sklop stavb (sl. 2: 14). Najstarejši zidovi so bili verjetno zgrajeni v 1. ali 2. st., kasnejše dozidave in prezidave so potekale vse do 5. st. Množina odpadne kovaške žlindre (25 kg) v eni od zgodnjih plasti nakazuje na kovaško dejavnost nekje v bližini.75 Na Gregorčičevi ulici, pred cerkvijo Ja­neza Krstnika (sl. 2: 13) sta bila izkopana zidova iz druge polovice 2. in 3. st. in del stavbe z ognjiščem iz 4. st.76 Na južnem robu ledine Mirce so bili pri gradnji železniške postaje (sl. 2: 16), v 19. stoletju, odkriti rim-ski zidovi.77 Dve jami z zgodnjerimskimi najdbami sta bili raziskani pri gradnji avtobusne postaje (sl. 2: 15).78 Tudi v notranjosti poznorimske utrdbe so bile prepoznane posamezne zgodnjerimske stavbe, ki so dru­gače usmerjene in drugače grajene kakor poznorimska poslopja.79 Drobne najdbe v utrdbi segajo od začetka 1. st. dalje.80 V južnem delu utrdbe je bila v najnižji plasti odkrita stavba s temelji iz prodnikov in malte81 (sl. 2: 9; 73 Müllner 1879, 133; Pick, Schmid 1916; 1922-1924, 302, sl. 148. Reprodukcija karte Svoljšak et al. 2013, 26. 74 Starejše najdbe: Petru 1975b; FMRSl IV, 49–51, št. 10/1–2; FMRSl V, 26, št. 7. 75 Tratnik, Žerjal 2017, 252; Kramar et al. 2015; Tratnik 2007. Prim. zidove najdene pri Osmuk 2005b–c; plasti z naj­dbami in veliko količine žlindre Osmuk 1977, 198–200. 76 Tratnik, Žerjal 2017, 254-256: velika količina novcev iz tridesetih let 4. st. datira gradnjo stavbe, živela pa je še ko­nec 4. in začetku 5. st. 77 Petru 1975b; Plesničar 1998; Štekar 1960-1961. 78 Osmuk 1993c; 1999d; 2005b. Nekaj najdb je bilo naj­denih v ornici. Svoljšak et al. 2013, 85: z lokacijo in drugačno interpretacijo, da gre pri Mircah za zaselek ali vilo rustiko. 79 Osmuk, Svoljšak, Žbona Trkman 1994, 45–50; Osmuk 1997, 129; Svoljšak et al. 2013, 41–42. 80 Vidrih Perko, Žbona Trkman 2003-2004: najdbe od druge pol. 1. st. do poznorimskega časa. Enako v: Vidrih Per-ko, Žbona Trkman 2005; Fabec 2018, 30: najdbe od konca 1. st. pr. n. št. ali začetkov 1. st. n. št. do 5./6. st. 81 Ostaline zgodnje rimske stavbe v južnem delu utrdbe area shown in Fig. 2: 14).69 Nothing is known of entrances to the fortress on other sides of the fortification wall.70 It is generally believed that the beds of the Lokavšček and Hubelj were corrected during the construction of the fort.71 It is not known whether the fluvius Frigidus was used as a navigable way, but the river must have been im­portant, as it was adopted for the name of the settlement, depicted on Tabula Peutingeriana and also mentioned in ancient texts.72 SETTLEMENT EARLY ROMAN SETTLEMENT – FLUVIO FRIGIDO Müllner was the first to describe the Roman ruins at the Mirce site. He considered that the toponym likely originated from the Latin word murus and wrote of the re­mains of ancient walls, brick and pottery extending across the meadows and fields just behind the southwest forti­fication wall and south of the present-day Gregorčičeva ulica. These remains covered a 20 ha large square surface, with one side measuring 450 m. He ascribed the ruins to the civil settlement at the fortress and the Early Roman road station (statio/mansio) Fluvio Frigido.73 The map that Schmid published later also shows Mirce southwest of the fortification wall. Recent excavations at Mirce unearthed the remains of Roman buildings and layers.74 The central part of the street of Cesta 5. maja even revealed a large building or a building complex (Fig. 2: 14). The earliest buildings were probably erected in the 1st or the 2nd century, with later additions or adaptations following through to the 5th century. One of the early layers yielded a great quantity of smithing slag (25 kg) that indicates the activity of a blacksmith.75 At Gregorčičevi ulica, in front of the church of John the Baptist (Fig. 2: 13), parts of a 4th-century 69 Tratnik, Žerjal 2017, 250, 254. 70 It was long held that the gates in the east fortification walls that led to the bridge towards Šturje were Roman, but they turned out to be medieval in date – Petru 1965, 138; Osmuk 1997, 122, and elsewhere. 71 Osmuk 1997, 121–122. 72 Vedaldi Iasbez 1994, 444-446. Vidrih Perko, Žbona Trkman 2004: with a comprehensive discussion of the hy­pothesis on river transport of raw material and products from the Vipava Valley to Aquileia. 73 Müllner 1879, 133; Pick, Schmid 1916; 1922-1924, 302, Fig. 148. Reproduction of the map in Svoljšak et al. 2013, 26. 74 Earlier finds: Petru 1975b. FMRSl IV, 49–51, No. 10/1– 2; FMRSl V, 26, No. 7. 75 Tratnik, Žerjal 2017, 252; Kramar et al. 2015; Tratnik 2007. Cf. the wall described in Osmuk 2005b-c; for layers with small finds and large quantities of slag, see Osmuk 1977, 198–200. 6 20 5 7 4 1 8 17 2 3 5 9 4 3 12 2 6 Lavri~ev trg 10 7 13 18 811 9 1 11 ŠTURJE 10 Cesta 5. maja 14 14 13 12 MIRCE Goriška cesta 19 15 16 0 100 m zgodnjerimska naselbina / Early Roman period settlement anti~ne najdbe / Roman period ynds zid / wall zgodnjesrednjeveški ostanki / Early Middle Age remnants grobiš~e / cemetery Sl. 2: Ajdovščina. Lokacije najpomembnejših raziskav. M. = 1:5000. Fig. 2: Ajdovščina. Plan of the Roman remains. Scale = 1:5000. 1, 2 Šibeniška ulica; 3 Prosenova hiša; 4 Šibeniška ulica, Lavričev trg; 5 Vilharjeva ulica; 6 Lavričev trg (atrijska hiša; območje stavbe III); 7 Gregorčičeva ulica; 8 Trg prve slovenske vlade (kinodvorana; območje stavbe I); 9 Bratinova hiša in tržnica (stavba I, južni del); 10 Bratinova hiša (stavba II); 11 Prešernova ulica (Stara pekarna); 12 Zahodno grobišče; 13 Gregorčičeva ulica; 14 Cesta 5. maja; 15 Mirce (avtobusna postaja); 16 Mirce (železniška postaja); 17 Ob Hublju (Vrtec); 18 Šturje (sv. Jurij); 19 Goriška cesta (stari internat); 20 Krkočev vrt. glej Urek, Kovačič 2020, sl. 1: A). Nad njo so bili odkriti temelji večjega poslopja (stavba I) in kloaka, pokrita s tegulo (sl. 2: 9; glej Urek, Kovačič 2020, sl. 1: I).82 Na (Bratinova hiša in tržnica) so bile raziskane večinoma v le­tih 1984, 1985 in 1987. Starejši zid se je raztezal na območju južnega dela stavbe I, pod poznorimskimi stavbami. Zid je drugače grajen in drugače usmerjen kot poznorimska stavba. Datiran je v 1., 2. in 3. st. oz. do gradnje obzidja v sedemde­setih letih 3. st. – Osmuk 1986a, 158; 1997; Osmuk, Svoljšak, Žbona Trkman 1994, 45–50; Svoljšak et al. 2013, 41–42. 82 Osrednji del stavbe I je bil raziskan ob gradnji nove kinodvorane leta 1965 (Trg prve slovenske vlade). Takrat so bili najdeni zidovi, arhitekturni elementi z napisom, kloaka (Goriški muzej in ZVKDS): Petru, Svoljšak 1965; 1980; Petru 1965, 134, 139-140; 1972, 350. Na osnovi odkritih napisov in tegule z žigom Q. Caecili Flaviani, ki je prekrivala kloako, Petru to stavbo interpretira kot svetišče iz 1. st.: Petru 1965, 134, 139-140; 1972, 350. Nadaljevanje kloake odkrito leta building with a hearth, as well as two other walls from the second half of the 2nd and 3rd centuries were found.76 Roman walls were also found at the south edge of Mirce (Fig. 2: 16), during the construction of the railway station in the 19th century.77 Two pits with Early Roman finds were investigated in advance of bus station construc­tion (Fig. 2: 15).78 Several Early Roman buildings have been identi­fied within the Late Roman fort, with an orientation 76 Tratnik, Žerjal 2017, 254-256: numerous coins from the 330s date the construction of the building, but also its use in the late 4th and early 5th centuries. 77 Petru 1975b; Plesničar 1998; Štekar 1960-1961. 78 Osmuk 1993c; 1999d; 2005b. Several artefacts were found in the topsoil. Svoljšak et al. 2013, 85: with the location of the excavations and a different interpretation, seeing Mirce as the site of either a hamlet or a countryside villa. 20 stavbo I se verjetno navezujejo tudi najdeni kamniti arhitekturni elementi; okrašen venec napušča (sl. 3), del stebra, trije deli napisov, bi lahko kazali na obstoj monumentalne javne stavbe, morda svetišča (Katalog napisov št. 8-10).83 Zgodnjerimski naselbinski ostanki so bili odkriti tudi na območju stavbe III (t. i. atrijska hiša; sl. 2: 6; glej Urek, Kovačič 2020, sl. 1: III).84 Severno od obzidja utrdbe se v literaturi omenja najdba mozaičnega tlaka (sl. 2: 20).85 Obseg zgodnjerimskega naselja – in pripadajočih grobišč – še vedno ni natančno poznan, naš predlog na osnovi predstavljenih podatkov je prikazan na sl. 2 in zajema obsežno površino, ki jo je začrtal že Petru.86 2005: Osmuk 2005a. Južni del stavbe I, stavba II s termami in obzidje s stol­pom 12 so bili raziskani v letih 1984, 1985 in 1987 ter 1992 (ZVKDS Nova Gorica). Manjša dela v okolici kinodvorane so potekala leta 2005 (ZVKDS Nova Gorica). Osmuk 1986a; 1988; 1990a, 169; 1990b; 1997, 123–127; 2005a; Osmuk, Svolj­šak, Žbona Trkman 1994, 41–50. – Stavba I je bila raziskanav letih 1987 in 1992. Osmuk, Svoljšak, Žbona Trkman 1994,41–44. Osmuk 1997, 123–127. Primerjaj: Vidrih Perko, Žbona Trkman 2003-2004, 46-48 (interpretirata kot svetišče); enako Svoljšak et al. 2013, 43. Stavba I je interpretirana različno. Severni del stavbe I so izkopali leta 2017-2018 in jo da­tirajo v 4. st.(glej Urek, Kovačič, 2020, sl. 1: II, II). 83 Petru, Svoljšak 1965a, 197; Petru 1972, 350; Svoljšak, Žbona Trkman 1986, 388–389, s starejšo literaturo. Le del napušča je bil odkrit kasneje, leta 1987 – Osmuk 1988, 233. Gradbeni napisi so datirani v (konec) 1. st. ali začetek 2. st. Pregled: Svoljšak et al. 2013, 63. 84 Izkopavanja Blagovnica Nanos leta 1968 in 1969. Zgod­nje naselbinske plasti in najdbe so datirane v drugo pol. 1. in 2. st. Pregled: Svoljšak et al. 2013, 41. 85 Točna lokacija ni poznana, t. i. “Krkočev vrt”: Petru 1975b; Štekar 1960–1961, 200; Djurić 1976, 560 (Ajdovščina – št. 1.1). 86 Petru 1965, 134; Osmuk 1997; 1999d; 2005b. Raz­prostranjenost lokacij: Svoljšak et al. 2013, 41; Tratnik, Žerjal 2017, sl. 1 in 2. and construction technique different from those of the Late Roman buildings.79 The small finds recovered from the fort date from the early 1st century AD on.80 In the south part of the fort, a building with foundations made of mortar-bound pebbles was found in the lowest layer81 (Fig. 2: 9; see Urek, Kovačič 2020, Fig. 1: A). Above it, the foundations of a large building (Building I) and a sewer covered with tegulae (Fig. 2: 9; see Urek, Kovačič 2020, Fig. 1: I) were found.82 Several architectural elements of stone have been found and ascribed to Building I: decorated cornice (Fig. 3), lower part of a column shaft and three fragments of inscriptions that indicate a monumental, public building, possibly a temple (Catalogue of inscrip­tions Nos. 8-10).83 Early Roman habitation remains also came to light in the area of Building III (so called Atrium 79 Osmuk, Svoljšak, Žbona Trkman 1994, 45–50; Osmuk 1997, 129; Svoljšak et al. 2013, 41–42. 80 Vidrih Perko, Žbona Trkman 2003-2004: small finds from the second half of the 1st century to the Late Roman pe­riod. The same in: Vidrih Perko, Žbona Trkman 2005; Fabec 2018, 30: small finds from the late 1st century BC or early 1st century AD to the 5th/6th century. 81 The remains of the Early Roman building in the south part of the fortress (Bratinova hiša and market place) were mainly investigated in 1984, 1985 and 1987. An earlier wall of different orientation and building technique was excavated in the south part of Building I, under Late Roman buildings; it is dated to the 1st-3rd centuries or up to the construction of the fortification walls in the 270s. Osmuk 1986a, 158; 1997; Osmuk, Svoljšak, Žbona Trkman 1994, 45–50; Svoljšak et al. 2013, 41–42. 82 The central part of Building I was investigated dur­ing the construction of a new cinema in 1965 (the Trg prve slovenske vlade square). The investigations revealed walls, inscribed architectural elements, a cloaca (Regional Museum of Gorica and ZVKDS): Petru, Svoljšak 1965; 198; Petru 1965, 134, 139-140; 1972, 350. The inscriptions and the tegulae with a Q. Caecili Flaviani stamp that covered the cloaca led Petru to interpret the building as a temple from the 1st century AD: Petru 1965, 134, 139-140; 1972, 350. The continuation of the sewer was found in 2005: Osmuk 2005a. South part of Building I, Building II with a bath com­plex and the fortification wall with Tower 12 in the south were investigated in 1984, 1985, 1987 and 1992 (ZVKDS). Small-scale construction work in the vicinity of the cinema took place in 2005 (ZVKDS). Osmuk 1986a; 1988; 1990a,169; 1990b; 1997, 123–127; 2005a; Osmuk, Svoljšak, Žbona Trkman 1994, 41–50. Building I was investigated in 1987 and1992. Osmuk, Svoljšak, Žbona Trkman 1994, 41–44; Osmuk1997, 123–127. Cf.: Vidrih Perko, Žbona Trkman 2003-2004, 46-48: interpretation as a temple; the same in Svoljšak et al. 2013, 43. Building I is interpreted in different ways. Northern part of building I was excavated in 2017-2018 and it dates to the 4th century (cf. Urek, Kovačič 2020, Fig. 1: I, II). 83 Petru, Svoljšak 1965a, 197; Petru 1972, 350; Svoljšak, Žbona Trkman 1986, 388–389, with earlier references. Only part of the cornice was unearthed later, in 1987 – Osmuk 1988, 233. The building inscriptions date to the (late) 1st or early 2nd century. Overview in: Svoljšak et al. 2013, 63. POZNORIMSKA UTRDBA – MUTATIO CASTRA Tloris obzidja poznorimske utrdbe Castra, ki leži ob sotočju Hublja in Lokavščka, je bil v glavnem raziskan v drugi polovici 20. stoletja. Tloris je nepravilne ovalne oblike (186 × 152 m) s 14 stolpi (sl. 4). Obzidje obsega površino 2,6 ha.87 Analiza novčnih najdb v maltni sre­dici obzidja in ruševinski izravnalni plasti pred njegovo gradnjo kaže, da je bilo obzidje zgrajeno v sedemdesetih ali osemdesetih letih 3. st.88 Schmid omenja obrambni jarek ob zahodnem obzidju,89 šele pred kratkim je bil dokumentiran tudi jarek na južni strani obzidja.90 V notranjosti utrdbe sta bila do najnovejših izko­pavanj raziskana dva večja dela naselbine. Stavba III (t. i. atrijska hiša) v vzhodnem delu utrdbe je datirana v 4. st. (sl. 2: 6; glej Urek, Kovačič 2020, sl. 1: III).91 V južnem delu utrdbe je bila odkrita stavba II s termami, ki je dati­rana v 4. st. (sl. 2: 10; glej Urek, Kovačič 2020, sl. 1: II).92 Večja in domnevno starejša stavba I, zahodno od term (sl. 2: 9; glej Urek, Kovačič 2020, sl. 1: I), je morala biti monumentalna javna stavba. V povezavi z najdbo delov napisa in arhitekturnih elementov (Katalog napisov št. 8-10) so jo interpretirali kot svetišče iz 1. oz. 2. st.,93 vendar se Nada Osmuk nagiba k poznejši dataciji in jo povezuje z vojsko ali upravo.94 V letih 2017-2019 je bila odkrita večja poznorim-ska stavba III s portikom in atrijem na Lavričevem trgu, ki se navezuje na že odkrito “atrijsko hišo” (sl. 2: 5–6; glej Urek, Kovačič 2020, sl. 1: III). Odkopan je bil tudi odsek 87 Pregled: Osmuk 1990b; 1997 z navedenimi starejši-mi objavami. Osmuk, Svoljšak, Žbona Trkman 1994, 24–36. Za posamezne kasneje raziskane odseke glej: Osmuk 2005a; 2005d; Prosenova hiša (Fabec, Tratnik 2009; Fabec 2017); Rustjeva hiša – stolp 11 (Draksler 2018). 88 V malti obzidja sta bila najdena novec zadnje emisije cesarja Galijena (kovan 267/268; v maltni sredici stolpa 12) in novec cesarja Klavdija II. (kovan po njegovi smrti leta 270; na robu maltne plasti obzidja), v izravnalni plasti pred gradn­jo pa novci od Septimija Severa do Proba. – Osmuk 1997, 122 (z datacijo v sedemdeseta leta 3. st.); Kos 2012, 285 (za deset let zamakne gradnjo); Kos 2014c, 35. 89 Pick, Schmid 1922–1924, 302, sl. 148: reprodukcija karte v Svoljšak et al. 2013, 25. Osmuk 1997.90 Rozman, Urek, Kovačič 2019; Urek, Kovačič 2020. 91 Izkopavanja Goriškega muzeja v letih 1968 in 1969, izgradnja blagovnice Nanos; Svoljšak 1968; 1968–1969a–b; Osmuk, Svoljšak, Žbona Trkman 1994, 51–53; Svoljšak et al. 2013, 47-49, 68. 92 Stavba II pod Bratinovo hišo (glej opombo 82). 93 Interpretacija stavbe kot svetišče iz konca 1. ali prve tretjine 2. st. (Hadrijanov čas): Petru 1965, 134, 139-140; 1972, 350–351; Svoljšak, Žbona Trkman 1986; ponovno Vi-drih Perko, Žbona Trkman 2003-2004, 49, sl. 11; Svoljšak et al. 2013, 42-46, 63-64. 94 Osmuk 1997, 123 (interpretira stavbo kot skladiščne prostore, za nastanitev vojske ali uporabo za upravo); str. 127 (datira stavbo v čas po izgradnji obzidja). House; Fig. 2: 6; see Urek, Kovačič 2020, Fig. 1: III).84 There is mention in literature of a mosaic floor north of the fortification wall (Fig. 2: 20).85 The full extent of the Early Roman settlement – and the associated cemeteries – remains unknown, but avail­able evidence has allowed us to propose a plan as shown in Fig. 2, covering a large surface already outlined by Petru.86 LATE ROMAN FORTRESS – MUTATIO CASTRA Investigations in the second half of the 20th century have determined the basic outline of the fortification wall of the Late Roman Castra, lying next to the confluence of the Hubelj and Lokavšček. The walls are irregularly oval in plan (186 x 152 m), enclosing a 2.6 ha large surface and comprising 14 towers (Fig. 4).87 The analysis of the coins recovered from the mortar core of the fortification wall and the levelling layer of debris preceding its con­struction has shown that the walls were constructed in the 270s or 280s.88 Schmid mentions a defensive ditch along the west fortification wall.89 Recent investigations have also un­earthed a ditch along the south fortification wall.90 Early investigations of the fortress interior have examined two large parts of the settlement. The Atrium House was found in the eastern part and dated to the 4th century (Fig. 2: 6; see Urek, Kovačič 2020, Fig. 1: III).91 84 The 1968 and 1969 excavations at the Nanos shopping centre. The early habitation layers and small finds date to the second half of the 1st and the 2nd century. Overview in: Svoljšak et al. 2013, 41. 85 Exact location unknown, recorded as ‘Krkočev vrt’: Petru 1975b; Štekar 1960–1961, 200; Djurić 1976, 560, Ajdovščina – No. 1.1. 86 Petru 1965, 134; Osmuk 1997; 1999d; 2005b; distri­bution of sites: Svoljšak et al. 2013, 41; Tratnik, Žerjal 2017, Figs. 1 and 2. 87 Overview: Osmuk 1990b; 1997 with earlier references. Osmuk, Svoljšak, Žbona Trkman 1994, 24–36. For sections investigated later, see: Osmuk 2005a; 2005d; Prosenova hiša: Fabec, Tratnik 2009; Fabec 2017. Rustjeva hiša – Tower 11: Draksler 2018. 88 A coin of the last emission of Gallienus (minted in 267/268; in the mortar core of Tower 12) and a coin of Claudius II (minted posthumously in 270; at the edge of the mortar layer of the fortification walls) were found in the mortar of the fortification walls, coins from Septimius Sever­us to Probus were found in the levelling layer as preparation for construction. Osmuk 1997, 122: with a dating into the 270s. Kos 2012, 285: dating the construction a decade later; Kos 2014c, 35. 89 Pick, Schmid 1922–1924, 302, Fig. 148: map repro­duced in Svoljšak et al. 2013, 25. Osmuk 1997. 90 Rozman, Urek, Kovačič 2019; Urek, Kovačič 2020. 91 Excavations of the Regional Museum of Gorica in 1968 and 1969, in advance of constructing the Nanos shop­ping centre; Svoljšak 1968; 1968–1969a-b; Osmuk, Svoljšak, Žbona Trkman 1994, 51–53; Svoljšak et al. 2013, 47-49, 68. cestišča na južni strani te stavbe, poznorimski objekti vzdolž Prešernove ulice in nadaljevanje stavbe I proti severu (glej Urek, Kovačič 2020, sl. 1).95 Ob številnih drugih manjših raziskavah so bili odkriti posamezni zidovi rimskih stavb. Na načrtu si odsevera proti jugu sledijo: dva odseka zidu na Šibeniški ulici (sl. 2: 1,2),96 zid na Gregorčičevi ulici (sl. 2: 7).97 Na posameznih mestih znotraj obzidja so bile odkopane plasti ruševine z gradbenim materialom, drobnimi najd­bami in kosi barvanega ometa: Šibeniška ulica, Lavričev trg, Vilharjeva in Prešernova ulica (sl. 2: 4,5,11).98 Na notranji strani zahodnega obzidja (sl. 2: 3) je bil razi­ 95 Urek, Kovačič 2020. 96 Osmuk 2006, 8-9. 97 Znotraj stavbe na parc. št. 546 k. o. Ajdovščina: Osmuk 1992, 191-192. 98 Lavričev trg, vzhodno od objektov na parc. št. 556 in 557 (sl. 2: 4); Osmuk Nada, dokumentiranje leta 2001 (neob­javljeno, dokumentacijo hrani arhiv ZVKDS OE Nova Go-rica). Vilharjeva ulica, vzhodno od objekta na parc. št. 596 (sl. 2: 5; Osmuk 1996b) in Prešernova ulica, Stara pekarna na parc. št. 974/1 (sl. 2: 11; Osmuk 1990a, 169; 1996a). Vse parc. številke so v k. o. Ajdovščina. In the southern part, Building II with a bath complex was excavated and also dated to the 4th century (Fig. 2: 10; see Urek, Kovačič 2020, Fig. 1: II).92 The larger and earlier Building I, located west of the bath complex (Fig. 2: 9; see Urek, Kovačič 2020, Fig. 1: I), was most likely a public building. It was associated with fragments of inscriptions and architectural members (Catalogue of inscriptions Nos. 8-10), hence interpreted as a temple, from the 1st or 2nd century.93 Nada Osmuk, however, rather believes the building is later and connected with the army or administration.94 A large Late Roman Building III with a porticus and an atrium was discovered in 2017-2019 at Lavričev trg (Fig. 2: 5–6; see Urek, Kovačič 2020, Fig. 1: III), which is connected with the previously unearthed “Atrium House” (Fig. 2: 6). Also excavated was a section of a road south of this building, as well as Late Roman buildings along the modern street of Prešernova ulica and a northward continuation of Building I.95 Individual walls of Roman buildings also came to light at numerous other interventions. They are marked on the plan from north to south as follows: two sec­tions of a wall at Šibeniška ulica (Fig. 2: 1,2),96 a wall at Gregorčičeva ulica (Fig. 2: 7).97 Debris layers with building material, small finds and pieces of painted wall plaster came to light at several locations in the fortressinterior: Šibeniška ulica, Lavričev trg, Vilharjeva ulica and Prešernova ulica (Fig. 2: 4,5,11).98 A shallow ditch, parallel with the fortification wall and filled in the 4th– 5th centuries, was found on the interior side of the west fortification wall (Fig. 2: 3).99 There are also several known sections of sewers; they do not seem to form a regular sewing system, rather individual sewers were installed so as to run directly under the fortification wall. 92 Building II under Bratinova hiša – see Fn. 82. 93 For the interpretation as a temple from the late 1st or first third of the 2nd century (Hadrianic), see Petru 1965, 134, 139-140; 1972, 350–351; Svoljšak, Žbona Trkman 1986; again in Vidrih Perko, Žbona Trkman 2003-2004, 49, Fig. 11; Svoljšak et al. 2013, 42-46, 63-64. 94 Osmuk 1997, 123: she interprets the building either as a storage facility, soldiers’ quarters or administration office; p. 127: dates the building to a time after the construction of the fortification walls. 95 Urek, Kovačič 2020. 96 Osmuk 2006, 8-9. 97 In the interior of the modern building at Lot No. 546, Ajdovščina cadastral community: Osmuk 1992, 191-192. 98 Lavričev trg, east of the buildings on Lot Nos. 556 and 557 (Fig. 2: 4); Osmuk Nada, recorded in 2001, unpublished. Records held in the archives of the ZVKDS. Vilharjeva ulica, east of the building on Lot No. 596 (Fig. 2: 5; Osmuk 1996b), and Prešernova ulica, Stara pekarna on Lot No. 974/1 (Fig. 2: 11; Osmuk 1990a, 169; 1996a). All lots are in the Ajdovščina cadastral community. 99 Prosenova hiša: Fabec 2017. skan plitek jarek, vzporeden z obzidjem in zasut v 4.–5. st.99 Znanih je tudi nekaj kanalizacijskih odtokov, ki pa ne kažejo enotnega odvodnega sistema, temveč so bili vedno povezani le z odtoki v obzidju. OBMOČJE VZHODNO OD POTOKA HUBELJ Poseljen je bil tudi vzhodni breg Hublja. Ruševinska plast z novci iz 4. st. je bila odkrita ob gradnji Vrtca (sl. 2: 17).100 Ruševine stavbe in številne najdbe novcev, orožja ter orodja od 1. do 4. st. so bile izkopane ob cerkvi sv. Jurija v Šturjah101 (sl. 2: 18), kjer se domneva zaselek ali vila rustika.102 POZNA ANTIKA IN ZGODNJI SREDNJI VEK Poselitev v zgodnjesrednjeveškem obdobju je poznana samo v drobcih. Suhi zidovi in jame za kole domnevno zgodnjesrednjeveških lesenih objektov iz 6. ali 7. st. so bili odkriti v ruševinah rimske atrijske hiše znotraj utrdbe (sl. 2: 6).103 Ob cerkvi sv. Jurija v Šturjah pa je bilo raziskanih 25 zgodnjesrednjeveških grobov, ki so datirani v čas od konca 9. do začetka 11. st. (sl. 2: 18).104 GROBIŠČA Zahodno grobišče (sl. 2: 12) se je raztezalo severno od današnje Gregorčičeve ulice (domnevna trasa rimske ceste) in zahodno od kasnejšega poznorimskega obzidja in obrambnega jarka. Proti severu sega grobišče vsaj do osnovne šole. Mnogi grobovi so bili uničeni pri gradnji cerkve sv. Janeza Krstnika (zgrajena v 17. st. na lokaciji starejše kapelice), župnišča in osnovne šole (leta 1954) ter zaklonišča (leta 1985) brez arheološkega nadzora. Raziskan je bil le manjši del grobišča: 44 grobov – 6 99 Prosenova hiša: Fabec 2017. 100 Vrtec (1975): na parc. 623/18, k. o. Ajdovščina (Osmuk 1977, 200–202). Žbona Trkman 1982 z omembo posamične najdbe novca cesarja Klavdija iz sredine 1. st.; FMRSl I–II, 46–47, št. 15. Toponim “Trnje” pogosto označuje ruševine in ostanke zidovja. 101 Brezigar, Josipović 2015, 134; Svoljšak et al. 2013, 51: odkrit tudi novec, datiran v konec 4. st. O starejših rimskih najdbah iz Šturij glej pri Moser 1891, 35; Petru 1971, 97–99; 1975b, 120; FMRSl I–II, 46, št. 14. 102 Svoljšak et al. 2013, 85. 103 Svoljšak 1968; 1968–1969a, 157; 1968–1969b, 186; 1970–1971, 159-161; Svoljšak, Knific 1976, 11-12, 53-54; Osmuk, Svoljšak, Žbona Trkman 1994, 53; Svoljšak et al. 2013, 49-50. 104 Svoljšak et al. 2013, 51-54. Objava: Brezigar, Josipović 2015, 134. EAST OF THE HUBELJ STREAM The east bank of the Hubelj was also inhabited in the Roman period. A debris layer with coins from the 4th century was unearthed during the construction of a nursery school (Fig. 2: 17).100 Debris of a building and numerous coins, weapons and tools dating from the 1st to the 4th century were excavated at the church of StGeorge in Šturje101 (Fig. 2: 18), presumably representing the remains of a hamlet or countryside villa.102 LATE ANTIQUITY AND EARLY MIDDLE AGES Very little is known on the settlement at Ajdovščina in the presumed Early Medieval period. Drywalls and postholes belonging to early medieval wooden buildings from the 6th and 7th centuries have been found within the ruins of the Atrium House (Fig. 2: 6).103 At the church of St George in Šturje, 25 burials dating from the late 9th to the early 11th century have been investigated (Fig. 2: 18).104 CEMETERIES The west cemetery (Fig. 2: 12) is located north of Gregorčičeva ulica (presumed route of the Roman road), i.e. west of the Late Roman fortification wall and defensive ditch. In the north, the cemetery extends at least to the primary school. Numerous graves were destroyed during the construction of the church of John the Baptist (built in the 17th century in place of an earlier chapel), the parish house, primary school (built in 1954) and underground shelter (1985) without watching briefs. Only a small part of the cemetery was archaeologically investigated, reveal­ing 44 burials – 6 cremations and 38 inhumations dating from the 1st to the 4th century.105 100 Vrtec (1975): on Lot No. 623/18, Ajdovščina cadastral community (Osmuk 1977, 200–202). Žbona Trkman 1982 with the note on a stray find of a coin of Claudius from the mid-1st century; FMRSl I-II, 46–47, No. 15. The toponym Trnje often marks ruins or remains of walls. 101 Brezigar, Josipović 2015, 134; Svoljšak et al. 2013, 51: a coin from the end of the 4th century. For earlier Roman finds from Šturje, see Moser 1891, 35; Petru 1975b, 120; 1971, 97–99; FMRSl I-II, 46, No. 14. 102 Svoljšak et al. 2013, 85. 103 Svoljšak 1968; 1968–1969a, 157; 1968–1969b, 186; 1970–1971, 159-161; Svoljšak, Knific 1976, 11-12, 53-54; Osmuk, Svoljšak, Žbona Trkman 1994, 53; Svoljšak et al. 2013, 49-50. 104 Svoljšak et al. 2013, 51-54. Published in: Brezigar, Josipović 2015, 134. 105 Petru 1975b; Žbona Trkman 1987; Osmuk 1987a-b; 1988; 1990a; 1993b; 1996a; 1997, 127–129; 1999c; 2005c. The cemetery presumably extended across roughly 6000 m2: Svoljšak et al. 2013, 38-40. Brezigar 2018: investigations dur­ing the renovations of the Danilo Lokar Primary School re­ žganih in 38 skeletnih grobov, datiranih v čas od 1. do 4. st.105 O južnem grobišču (sl. 2: 19), v bližini mostu čez Hubelj, poznamo le omembe grobov “pri starem inter-natu in javnem kopališču iz sredine 19. stoletja”. Najdeni naj bi bili tako žgani kakor skeletni pokopi.106 Starejša pričevanja omenjajo, da so v dvajsetih letih odkrili 26 grobov na zahodnem delu ledine Mirce.107 Posamezni rimski nagrobniki so bili vzidani v obzidje poznorimskega kastela in srednjeveške stavbe. Najstarejši je s konca 1. st. pr. n. št.108 DROBNE NAJDBE Objavljen je izbor najdb z izkopavanj na južnem (sl. 2: 9) in na osrednjem delu utrdbe (sl. 2: 2,6)109 ter najdbe z manjših arheoloških raziskav znotraj in zunaj mestnega jedra.110 Poznanih je tudi nekaj najdb iz zbir­ke Stipeta Štekarja,111 opeke z žigi,112 izbor amfor113 in uvoženega posodja z različnih lokacij v Ajdovščini.114 Izbor gradiva je predstavljen v katalogu razstave muzeja v Ajdovščini.115 Objavljeni so zgodnjesrednjeveški gro­bovi iz Šturij (sl. 2: 18).116 Drobno gradivo v Ajdovščini kaže, da je bila le-ta odlično vpeta v trgovsko mrežo Regio X in akvilejskega agra. Raznovrstno blago vseh vrst, uvoženo iz celega 105 Petru 1975b; Žbona Trkman 1987; Osmuk 1987a–b; 1988; 1990a; 1993b; 1996a; 1997, 127–129; 1999c; 2005c. Ob-seg grobišča je ocenjen na območje veliko 6000 m2 (Svoljšak et al. 2013, 38-40). Brezigar 2018: raziskave ob prenovi OŠ Danila Lokarja so razkrile 20 grobov z vsaj 30 skeleti. Omen-ja tudi leta 2017 odkrit žgan grob pri nadzoru gradnje plino­voda za občinsko stavbo. 106 Hitzinger 1855; Petru 1965, 134; Pick, Schmid 1922– 1924, na skici je označena lokacija; Petru 1975b, 120; Svoljšak et al. 2013, 26, 38. 107 Petru 1965, 134; Svoljšak et al. 2013, 26. 108 Svoljšak et al. 2013, 57. Natančneje v poglavju Katalog napisov. 109 Vidrih Perko, Žbona Trkman 2003-2004; 2004; 2005, 17–72; Svoljšak et al. 2013. 110 Fabec, Tratnik 2009, 19–21; Tratnik, Žerjal 2017; Kos 2017. 111 Štekar 1960–1961; glej še Žbona Trkman 1990a, 169– 170. 112 Žbona Trkman 1993; Vidrih Perko, Žbona Trkman 2003–2004, 48 (L. Epidi Theodori, L. Q. Thal, Q. Clodi Ambro­si, Tib. Veiti Aviti, Q. Caecili Flaviani, C. Q. Vir., Sallvst, Eva-risti in Ti. Evros); Zaccaria 2014, 181–182 (tegule z napisom Constantini victoris). 113 Vidrih Perko 1994, 88–96; 1997; 2000, 442. 114 Uvoženo posodje različnih produkcij – Vidrih Per-ko, Žbona Trkman 2003–2004; 2004; 2005, nazadnje Tratnik, Žerjal 2017. Samo za afriško sigilato, afriške oljenke in af­riško kuhinjsko keramiko glej Pröttel 1996. 115 Svoljšak et al. 2013, 59–79. 116 Svoljšak et al. 2013, 52–54, 77–78; Brezigar, Josipovič 2015. Even less is known of the south cemetery (Fig. 2: 19), located near the bridge across the Hubelj. We have notes on graves found at the old boarding school and public baths constructed in the mid-19th century. The burials were cremations, some in urns, and inhumations.106 Petru mentions that 26 graves were found in the 1920s in the west part of Mirce.107 Some Roman tombstones were built into the fortification wall of the Late Roman fortress and into medieval houses; the earliest tombstone dates to the late 1st century BC.108 SMALL FINDS A selection of artefacts excavated in the southern (Fig. 2: 9) and central parts of the fortress (Fig. 2: 2,6),109 as well as the finds recovered during small-scale inves­tigations in the fortress’ interior and exterior have been published.110 In addition, some artefacts from the StipeŠtekar collection have been presented to the public,111 as well as stamped brick,112 select amphorae113 and imported pottery recovered from different sites across Ajdovščina.114 A selection of artefacts has been included in the exhibition catalogue of the museum collection in Ajdovščina.115 The early medieval burials from Šturje have also been published (Fig. 2: 18).116 The small finds show that Ajdovščina was well inte­grated in the trading network in regio X and the territory of Aquileia. The diversity of goods, imported from across the Mediterranean and shipped to the ports of the North vealed 20 graves with at least 30 skeletons. Also mentioned is a cremation discovered during the 2017 watching brief of the gas pipeline construction for the municipal building. 106 Hitzinger 1855; Petru 1965, 134; Pick, Schmid 1922– 1924, the location is marked on the sketch; Petru 1975b, 120; Svoljšak et al. 2013, 26, 38. 107 Petru 1965, 134; Svoljšak et al. 2013, 38. 108 Svoljšak et al. 2013, 57. More details in the Catalogue of inscriptions. 109 Vidrih Perko, Žbona Trkman 2003-2004; 2004; 2005, 17–72; Svoljšak et al. 2013. 110 Fabec, Tratnik 2009, 19–21; Tratnik, Žerjal 2017; Kos 2017. 111 Štekar 1960–1961; cf. Žbona Trkman 1990a, 169–170. 112 Žbona Trkman 1993; Vidrih Perko, Žbona Trkman 2003–2004, 48: L. Epidi Theodori, L. Q. Thal, Q. Clodi Am­brosi, Tib. Veiti Aviti, Q. Caecili Flaviani, C. Q. Vir., Sallvst, Evaristi and Ti. Evros; Zaccaria 2014, 181–182: tegulae with the Constantini victoris inscription. 113 Vidrih Perko 1994, 88–96; 1997; 2000, 442. 114 For imported pottery from different production centres,see Vidrih Perko, Žbona Trkman 2003–2004; 2004; 2005, last in Tratnik, Žerjal 2017. Exclusively for African sigillata ware, Afri­can oil lamps and African coarseware, see Pröttel 1996. 115 Svoljšak et al. 2013, 59–79. 116 Svoljšak et al. 2013, 52–54, 77–78; Brezigar, Josipovič 2015. Sredozemlja, ki je bilo dosegljivo v severno jadranskih pristaniščih, posebej v Akvileji, kaže na dobro organizi­rano dobavo, značilno za zahodno Slovenijo tako v zgo­dnjem kakor poznem rimskem obdobju. Ajdovščina je bila že zgodaj, vsaj od 1. st. n. št., pomembno regionalno trgovsko središče. Velik delež uvoženih dobrin v pozno­rimskem obdobju (4. st. in prva polovica 5. st.) nakazuje pomembno, morda oskrbovalno funkcijo utrdbe Castra 117 za del obrambnega sistema Claustra Alpium Iuliarum. Numizmatika: Najstarejši novci so pozno republikanski, najmlajši so iz časa cesarja Honorija (do leta 423).118 Starejši raz­iskovalci so slučajno najdene novce iz 1. st. povezovali z naselbino – mansio; medtem ko naj bi na zahodnem grobišču naj bi prevladovali novci iz 3. st.119 Novejše analize poudarjajo, da se v času od konca 3. do začetka 5. st. poveča delež novcev v obtoku. To bi lahko odražalo pomen naselbine v sklopu vojaške oskrbe zapornega sistema Claustra Alpium Iuliarum. Ajdovščina – Castra: FMRSl I–II, 34–46, št. 13; FMRSl III, 23–48, št. 12; FMRSl IV, 48, št. 9; FMRSl V, 25–26, št. 6; FMRSl VI, 15, št. 8; Ajdovščina – Mirce: FMRSl IV, 49–51, št. 10/1–2; FMRSl V, 26, št. 7; Ajdovščina – Šturje: FMRSl I–II, 46, št. 14; Ajdovščina – Trnje: FMRSl I–II, 46–47, št. 15. STATUS NASELJA Glede na omembe v itinerarijih je bila v naselju tovorna in poštna cestna postaja. Jeruzalemski itinera­rij jo imenuje mutatio.120 Domneva se, da je bil status naselbine v Ajdovščini vicus – vsekakor pa je bilo to večje naselje, morda administrativni center tega dela akvilejskega agra.121 117 Vidrih Perko, Žbona Trkman 2003–2004; 2004; 2005; Tratnik, Žerjal 2017.118 Petru 1975b; Kos 1986; 1994; 2012; 2014a–c; 2017. 119 Petru 1975b. 120 It. Burd. 560, 2. O lociranju in zadolžitvah cestnih postaj glej Šašel Kos 1997b, 29; Calzolari 2000. O sami Ajdovščini: Šašel 1984, 802; v povezavi z gospodarskimi danostmi območja: Vidrih Perko, Žbona Trkman 2003–2004, 58–65; 2004; 2005; podob-no še Zaccaria 2009b, 260. Adriatic, particularly Aquileia, reveals a highly organised supply of goods that is characteristic of western Slovenia in the Early, but also Late Roman periods. Very early on, at least in the 1st century AD, Roman Ajdovščina was a trading centre of regional importance. The great quantity of imported goods in the Late Roman period (4th and first half of the 5th centuries) suggests an important, pos­sibly supply function of the Castra fortress as part of the Claustra Alpium Iuliarum barrier system.117 Numismatic evidence: The earliest coins date to the Late Republican period, the latest were minted under Honorius (up to 423).118 Earlier researchers connected the stray coin finds from the 1st century with the mansio. The coins from the west cemetery largely date to the 3rd century.119 Recent analyses emphasise that there was an increasing amount of coins in circulation from the late 3rd to the early 5th century, which may reflect the role of the fortress in supplying the army deployed to the Claustra Alpium Iuliarum barrier system. Ajdovščina – Castra: FMRSl I-II, 34–46, No. 13; FMRSl III, 23–48, No. 12; FMRSl IV, 48, No. 9; FMRSl V, 25–26, No. 6; FMRSl VI, 15, No. 8; Ajdovščina – Mirce: FMRSl IV, 49–51, No. 10/1-2; FMRSl V, 26, No. 7 Ajdovščina – Šturje: FMRSl I-II, 46, No. 14 Ajdovščina – Trnje: FMRSl I-II, 46–47, No. 15 STATUS OF THE SETTLEMENT Itineraries reveal the existence of a transport and 120 postal station; the Jerusalem Itinerary calls it mutatio.It is presumed that the settlement in Ajdovščina had the status of a vicus – it was certainly a substantial settlement and possibly the administrative centre of this part of Aquileian territory.121 117 Vidrih Perko, Žbona Trkman 2003–2004; 2004; 2005; Tratnik, Žerjal 2017.118 Petru 1975b; Kos 1986; 1994; 2012; 2014a-c; 2017. 119 Petru 1975b. 120 It. Burd. 560, 2. 121 For the location and function of road stations, see Šašel Kos 1997b, 29; Calzolari 2000; for Ajdovščina, see Šašel 1984, 802; for the economic potential of the area, see Vidrih Perko, Žbona Trkman 2003–2004, 58–65; 2004; 2005; simi­larly in Zaccaria 2009b, 360. PREBIVALCI Posamezni nagrobni spomeniki omenjajo rimske državljane, osvobojence ali predstavnike višjega sloja in staroselske prebivalce, ki so pridobili državljanstvo ter pomembne administrativne funkcije.122 Najstarejši, iz konca 1. st. pr. n. št., je nagrobnik Tita Azinija (Katalog napisov št. 3). Mati Sessidena Maxuma C(ai) f(ilia) je bila rimska državljanka, z gentilnim imenom keltskega ali epihornega substrata. Bila je poročena z rimskim državljanom po imenu T(itus) Asinius, L(ucii) f(ilius). Edikula je bila posvečena tudi nekdanji Azinijevi sužnji, osvobojenki Aziniji Grati (Asinia Grata T(iti) l(iber­ta)). Na tem nagrobniku lahko opazujemo popolno integracijo prišlekov italskega rodu s pripadniki druge generacije romaniziranih staroselcev, ki so že pridobili rimsko državljanstvo.123 Na kasnejšo pridobitev rimskega državljanstva oz. na staroselca brez rimskega državljanstva kaže nagrobna stela Tarokona Briscinija (Taroco Briscinius Tropi filius) (Katalog napisov št. 4). Datirana je v prvo polovico 1. st. n. št. Vinogradnik po poklicu (simbolizira ga upodo­bljena falx vinitoria) je dal postaviti nagrobnik zase in za svojo snaho Kvarto Frejo (Quarta Freia Marci filia), ki je domnevno imela rimsko državljanstvo. Imena, predvsem moških Taroka in Rufusa, imajo naprednejšo staroselsko onomastično formulo.124 Rimsko državljanstvo sta imela osvobojenec Publij Publicij Ursion in njegova žena, ki sta omenjena na nagrobniku iz druge polovice 2. st. (Katalog napisov št. 2). Publius Publicius Ursio je bil upravnik ali nadzornik – saltuarius javnih posesti (dum saltus publicos curo) v lasti akvilejske kolonije. V kraju, kjer je prebival, je postavil spomenik sebi in svoji ženi. Iz onomastične formule in gentilnega imena Publicius je razvidno, da je bil osvobojenec in nekdanji javni suženj. Žena Volti­lia Satunna je pripadala staroselskemu prebivalstvu z imenom severnojadranskega jezikovnega podstata.125 Na zahodnem grobišču je bil odkrit nagrobnik (Katalog napisov št. 1) Antonija Valentina (Antonius Valentinus), poveljnika oz. centuriona (princeps legionis) XIII. legije Gemine, ki mu ga je postavil sin. Iz napisa izhaja, da so centuriona umorili razbojniki na območju Hrušice (in Alpes Iulias). Nagrobnik je datiran v (drugo polovico ali pozno) 2. ali 3. st.126 122 Zaccaria 2009b, 258-261. 123 Zaccaria 2009a, 90; 2009b, 160. 124 Buora 2007, 247–248; Zaccaria 2007b, 324; 2009a, 90, 96–99; 2009b, 260. 125 Šašel 1980, 183-184; Zaccaria 2003, 323; Vidrih Perko, Žbona Trkman 2003–2004, 60–61; Zaccaria 2009b, 259. Do-mneva se, da so se omenjene javne posesti razprostirale v bližini njegovega prebivališča v Ajdovščini, torej na območju Vipavske doline ali v hribovitih pašniških in gozdnatih pre­delih Trnovskega gozda. 126 Katalog napisov št. 1. INHABITANTS The funerary monuments from the Ajdovščina area mention Roman citizens, freedmen and freedwomen, members of upper classes and individuals of indigenous origin who obtained citizenship and held important administrative offices.122 The earliest dates to the late 1st century BC. It is a tombstone from the late 1st century BC, erected for Titus Asinius (Catalogue of inscriptions No. 3). T(itus) Asinius was a Roman citizen and married to Sessidena Maxuma, who was also a Roman citizen, but her gentilicium reveals a Celtic or epichoral origin. The tombstone also marked the burial of Asinia Grata, freedwoman of Asinius. The funerary monument is an example of a complete integration of newcomers from Italy with members of the second generation of Roman-ised indigenous population who had already obtained Roman citizenship.123 A slightly later tombstone is that for Taroco Bris­cinius, son of Tropus (Catalogue of inscriptions No. 4), from the first half of the 1st century AD. He was either a Roman citizen or an indigenous person without Roman citizenship. He was a viticulturist, with the attribute of falx vinitoria carved below the inscription, who had the tombstone erected for himself and his daughter-in-law Quarta Freia, daughter of Marcus, who was presumably a Roman citizen. The names, particularly of the men Taroco and Rufus, reveal an advanced indigenous ono-mastic formula.124 Freedman Publius Publicius Ursio and his wife, mentioned on a funerary monument from the second half of the 2nd century (Catalogue of inscriptions No. 2), certainly had Roman citizenship. Ursio was a saltuarius, curator of public woodlands (dum saltus publicos curo) of the colony of Aquileia. He put up the tombstone for himself and his wife in his place of residence. The ono-mastic formula and the gentilicium Publicius show him to be a former public slave. His wife, Voltilia Satunna, was of indigenous origin with a name of a north Adriatic linguistic substratum.125 The tombstone for Antonius Valentinus, princeps legionis XIII Geminae (Catalogue of inscriptions No. 1) was found in the west cemetery. The inscription reveals it was put up by his son, but also that he was killed by rob­bers in the area of Hrušica (in Alpes Iulias). The tombstone is dated to the (second half or late) 2nd or 3rd century.126 122 Zaccaria 2009b, 258-261. 123 Zaccaria 2009a, 90; 2009b, 160. 124 Buora 2007, 247–248; Zaccaria 2007b, 324; 2009a, 90, 96–99; 2009b, 260. 125 Šašel 1980, 183-184; Zaccaria 2003, 323, Vidrih Perko, Žbona Trkman 2003–2004, 60–61; Zaccaria 2009b, 259. These public estates were presumably located near his place of resi­dence at Ajdovščina, i.e. in the Vipava Valley or the forested hills of Trnovski gozd also suitable for pasturing. 126 Catalogue of inscriptions No. 1. Na odlomku napisa (Katalog napisov št. 10) večje javne stavbe, ki je bil izdelan s skrbno izklesanimi mo-numentalnimi črkami, je morda omenjen sevir.127 Onomastična formula imen izpričanih na nagrob­nikih predstavlja predvsem staroselsko prebivalstvo severnojadranskega jezikovnega podstata.128 Nekaj pa je povsem italskih oz. latinskih imen, ki morda pripadajo italskim prišlekom. RAZVOJ NASELBINE Za naselje Fluvio Frigido je sprejeto mnenje, da gre za postajo ob itinerarski cesti in pomemben lokalni administrativni center v gospodarskem zaledju akvi­lejskega agra. Nastanek rimske naselbine so že starejši raziskoval­ci povezovali s potekom stare jantarne poti in itinerarske ceste proti Okri.129 Šašel je njen nastanek postavil celo v drugo polovico 2. st. pr. n. št. kot emporij, postojanko vojaške kontrole, središče kmetijske dejavnosti in vzreje vlečne živine.130 Ker je najstarejši nagrobnik datiran v konec 1. st. pr. n. št.,131 lahko domnevamo, da je bila naselbina zasnovana že nekoliko pred tem ali tekom 1. st. pr. n. št., z uveljavitvijo rimskega nadzora nad Vipavsko dolino in vzpostavitvijo uprave in administracije obmo-čja. Najstarejše objavljene drobne najdbe so datirane v začetek ali prvo polovico 1. st. n. št.132 Z avgustejsko izgradnjo odcepa proti Hrušici je postajališče in trgovsko središče postalo pomembna zadnja postaja pred strmim vzponom v Alpes Iuliae. Latinsko ime naselbine “Kraj ob Mrzli reki” verjetno odraža novonastalo naselje ob javni cesti in prehodu čez reko.133 Raziskave so pokazale, da rimska poselitev v 1. in 2. st. ni bila gosto strnjena le na območju okrog poznejšega srednjeveškega jedra Ajdovščine, ampak je obsegala še dobršen del okolice (Mirce), vključno z najdbami ob železniški progi. V povezavi z najdbo nagrobnika nadzornika javne posesti P. Publicija Ursiona in večje javne stavbe, kateri pripadajo arhitekturni elementi in gradbeni napis iz konca 1. ali iz 2. st., se v tedanji nasel­ 127 V bližini so najdeni še napisi (Katalog napisov št. 8 in 9) ter odlomek okrašenega venca napušča.128 Alföldy 1978; 1999; Zaccaria 2009b, 243–244 s starej­šo literaturo. 129 Müllner 1879, 133; Stucchi 1946, 29; Petru 1965, 134; 1971; Bosio 1970, 193; Šašel 1980; 1984, 119–121. Nazadnje izčrpno in pregledno Vidrih Perko, Žbona Trkman 2005, 17; Zaccaria 2009b, 260. 130 Šašel 1984, 119–121, natančneje str. 120. 131 Osmuk 1997, 127; Svoljšak et al. 2013, 57. Glej Kata-log napisov št. 3. Vidrih Perko, Žbona Trkman 2005; Svoljšak et al. 2013, 60–62; Tratnik, Žerjal 2017; Fabec 2017: med najsta­rejšimi so odlomki keramike s črnim premazom, značilni za konec 1. st. pr. n. št in prvo četrtino 1. st. n. št. 133 Šašel Kos 1997b, 31. The fragment of a building inscription from a large public building (Catalogue of inscriptions No. 10), per­ 127 haps mentions a sevir. The onomastic name formulas on the tombstones reveal a largely indigenous population of the North Adriatic linguistic substratum,128 as well as some typically Italic or Latin names that presumably belong to Italian immigrants. DEVELOPMENT OF THE SETTLEMENT It is generally accepted that the settlement of Fluvio Frigido was a road station along the via publica from Aquileia to Emona and an important local administrative centre in the economic hinterland of Aquileia. From the early researchers on, the beginnings of the settlement have been associated with the traffic along the ancient Amber Route and public road towards the Ocra Pass.129 Jaroslav Šašel dated its beginning as early as the second half of the 2nd century BC, when it functioned as an emporium, an army control post, a centre for agricul­ture and draught animal rearing.130 The earliest tombstone from Ajdovščina, dated to the late 1st century BC,131 sug­gests that the settlement was founded either some time before this date or more broadly in the 1st century BC, in the period when the Romans established control and administrative oversight of the Vipava Valley. Having said that, the earliest published small finds date to the early or first half of the 1st century AD.132 With the Augustan construction of the road across the Hrušica Pass, this postal and trade centre became the last major stop before the steep ascent to the Alpes Iuliae. The Latin name of the settlement, roughly translated into ‘Place upon the Cold River’, probably designates the newly established settlement along the public road and river crossing.133 Research has shown that the core of Roman settle­ment in the 1st and 2nd centuries covered not only the medieval centre of Ajdovščina, but also wider (Mirce), 127 Other inscriptions (Catalogue of inscriptions Nos. 8 and 9) and a fragment of a decorated cornice were found in proximity. 128 Alföldy 1978; 1999; Zaccaria 2009b, 243–244 with earlier references. 129 Müllner 1879, 133; Stucchi 1946, 29; Petru 1965, 134; 1971; Bosio 1970, 193; Šašel 1980; 1984, 119–121. For recent comprehensive and detailed publications, see Vidrih Perko,Žbona Trkman 2005, 17; Zaccaria 2009b, 260. 130 Šašel 1984, 119–121, particularly p. 120. 131 Osmuk 1997, 127; Svoljšak et al. 2013, 57. See Cata­logue of inscriptions No. 3. 132 Vidrih Perko, Žbona Trkman 2005; Svoljšak et al. 2013, 60–62; Tratnik, Žerjal 2017; Fabec 2017: among the earliest finds are sherds of ceramica a vernice nera character­istic of the late 1st century BC and the first quarter of the 1st century AD. 133 Šašel Kos 1997b, 31. bini domneva lokalni upravni center. V njem bi lahko pobirali mitnino, davke in dajatve iz gospodarskega zaledja Akvileje – rodovitne Vipavske doline in Trno­vskega gozda.134 Gozdovi so bili vir lesa, oglja, lovnih živali; pašniki za vzrejo drobnice, ravnice za pridelavo vina, žita in drugih pridelkov ter za izdelke kakor so npr. opeka in lončenina. V naselju so bili prisotni obrtniki (kovaška žlinda na Cesti 5. maja) in nedvomno tudi trgovci, saj drobne najdbe odražajo zelo bogato dobavo izdelkov iz oddaljenih trgov preko pristanišča v Akvileji. Nagrobnik poveljnika 13. legije Gemine priča, da so že v nevarnih časih druge polovice 2. ali v 3. stoletju javno cesto proti Emoni varovali vojaški oddelki legij, ki so bile stacionirane v oddaljenih taborih.135 V poznem 3. st. je bilo v severovzhodnem delu naselbine, neposredno ob sotočju Hublja in Lokavšč­ka zgrajeno obzidje s stolpi. Pri tem so porušili nekaj starejših stavb. Pri zasnovi so upoštevali lokacijo zaho­dnega grobišča in za obrambo izkoristili oba vodotoka. Izhajajoč iz podatkov arheoloških raziskav in novčnih analiz se izgradnja datira v sredino sedemdesetih ali osemdesetih let 3. st., kar je približno sočasno oziroma 136 desetletje starejše od gradnje trdnjave Ad Pirum. Naselbina izven obzidja je živela dalje. V 4. in 5. st. so še zidali nove stavbe (Gregorčičeva ulica in Cesta 5. maja), živela pa je tudi četrt prek Hublja. Poimenovanje mutatio Castra v Jeruzalemskem itinerariju kaže na to, da je imela naselbina dvojno funkcijo: civilno/trgovsko ter da je bila v konstantinskem obdobju že utrjena. Menimo, da mutatio Castra lahko razumemo kot naselbino s preprežno postajo, utrjenim obzidanim središčem, v katerem je bila nastanjena tudi vojaška posadka. Glede na razmere ob koncu 3. in v 4. st. ter lego naselbine ob ključni prometnici proti Italiji je bila v utrd-bi Castra verjetno stalno navzoča vojaška posadka.137 Petru je menil, da se je Ajdovščina ob izgradnji sistema 134 Katalog napisov št. 2. Šašel 1980, 527; 1984, 119-121; Zaccaria 2003, 323; 2009b, 260; Vidrih Perko, Žbona Trkman 2005. 135 Bratož 2014, 25–26 (op. 58: emonski nagrobniki voja­kov legije XIII Gemine se večinoma datirajo v čas cesarja Galijena [260–268], 38, posebej op. 98). Šašel Kos 2016, 217 (z datacijo v markomanske vojne, tabor legije je bil takrat stacioniran v Daciji: Apulum – Alba Iulia). Glej še: Katalog napisov št. 1. in opombo 127. Kos 2012, 285, 299: avtor v zaključku poudari, da v trdnjavi Ad Pirum ni novcev iz druge polovice 3. st., ki bi jih lahko neposredno povezali z izgradnjo obzidja, povečan dotok denarja je zaznan v obdobju 315-324. Kos 2014b, 127, 130: Licinijev novec (314/315) kot terminus ante quem za iz­gradnjo obzidja. 137 Nekaj orožja in delov vojaških pasnih garnitur pozna-mo z območja izkopavanj atrijske hiše: Svoljšak et al. 2013, 70 (kat. št. 10-21) in 73 (kat. št. 7-18); Pflaum 2000, 89-90; 2004, 263; Ciglenečki, Milavec 2009. Več vojaških predme­tov je bilo odkritih ob izkopavanjih v središču poznorimske trdnjave Castra v letih 2018–2019 (neobjavljeno). including the area along the railway. The tombstone for P. Publicius Ursio, curator of public woodlands, as well as the architectural pieces and building inscriptions of a large public building from the late 1st or 2nd centuries suggest it functioned as a centre of local administration. It may have been charged with collecting toll and taxes from the economic hinterland of Aquileia, i.e. the fertile Vipava Valley and Trnovski gozd.134 In Roman times, forests provided wood, charcoal and game animals, pastures were important for raising livestock, the lowland for vineyards, for growing cereals and other produce, but also to produce goods such as brick and pottery. Artisans were active in the settlement (smithing slag at the Cesta 5. maja site) and doubtlessly also merchants, as the recovered artefacts point to a rich supply of diverse goods even from distant markets, arriving here via the port at Aquileia. The tombstone for the commander of legio XIII Ge-mina shows that, in the tumultuous times of the second half of the 2nd and 3rd centuries, the public road towards Emona was protected by units of legions otherwise sta­tioned in distant forts.135 In the late 3rd century, a fortification wall with towers was constructed in the north-eastern part of the settlement, directly at the confluence of the Hubelj and Lokavšček streams, pulling down several earlier build­ings in the process. The layout of the fortification wall respected the location of the west cemetery and also the position of both streams, which were incorporated as defensive features. The results of archaeological investiga­tions and numismatic evidence date the construction to the 270s or 280s, which is either roughly contemporane­ous or a decade earlier than the construction of the Ad Pirum fort.136 The settlement outside the fortification wall lived on, with new buildings added in the 4th and 5th centuries (the Gregorčičeva ulica and Cesta 5. maja sites), also the part of the settlement beyond the Hubelj. The name mutatio Castra noted in the Jerusalem Itinerary indicates the settlement’s double function: civil-ian/mercantile function, on the one hand, and one of a fortified settlement (already in the times of Constantine), 134 Catalogue of inscriptions No. 2. Šašel 1980, 527; 1984, 119-121; Zaccaria 2003, 323; 2009b, 360; Vidrih Perko, Žbona Trkman 2005. 135 Bratož 2014, 25–26, Fn. 58: the tombstones of the sol­diers of legio XIII Gemina from Emona largely date to the time of the Emperor Gallienus (260–268); 38, particularly Fn. 98. Šašel Kos 2016, 217: dating to the time of the Marco-mannic Wars or later, when the legion was stationed in Dacia (Apulum – Alba Iulia). See Catalogue of inscriptions No. 1. 136 Kos 2012, 285, 299: In conclusion, the author empha-sises that the Ad Pirum fort yielded no coins from the second half of the 3rd century that could be directly related to the construction of the fortification walls, an increased circula­tion is only noted in the 315-324 period. Kos 2014b, 127, 130: coin of Licinius (314/315) as terminus ante quem for the construction of the fortification walls. Claustra Alpium Iuliarum v sredini 4. st. razvila v štabno bazo osrednjega dela zapor.138 Kos povečano aktivnost v trdnjavah Castra in Ad Pirum, o kateri sklepa iz analize novčnih najdb, pove­zuje z upravnimi in vojaškimi reformami Dioklecijana v začetku 4. st. ter s potrebo po okrepljenem varovanju severovzhodne meje Italije.139 Trdnjava Castra je lahko služila tudi kot skladiščni oz. oskrbovalni center za čete stacionirane na bližnjih linijskih zaporah.140 Vojska je gotovo izkoristila dobro strateško lego in trgovsko mrežo, ki je bila v naselbini Fluvio Frigido že prej vzpostavljena, kar odraža tudi zelo bogata dobava izdelkov iz oddaljenih trgov. Pomen ceste čez Hrušico v 4. st. izkazujejo trije, v bližini Ajdovščine odkriti miljniki, s posvetili v čast vladarjem.141 Čeprav numizmatične analize iz zadnjih let pravijo, da je mogoče dotok svežega denarja v trdnjavo zaznati le še v prvih treh desetletjih 5. st.,142 pa najmlajše keramične najdbe kažejo, da se je življenje v utrdbi in njeni okolici vsaj v skromni obliki nadaljevalo tudi v drugi polovici 5. st.143 Ostanki zgodnjesrednjeveških stavb v ruševinah atrijske hiše potrjujejo obljudenost še v 6. in 7. stoletju. 138 Šašel, Petru 1971, 98-99: v opombi 9 navaja še mnen­je Stucchija, da gre za oskrbovalno bazo planinskih čet. Tezo o vojaškem taboru je sprejela tudi N. Osmuk (1998).139 Kos 2014b, 130-131. 140 Že Stucchi 1946, 129. 141 Miljniki iz časa vlade Konstancija in Galerija, Julija­ na Odpadnika ter Valentinijana in Valensa so bili odkriti v Mirnu, Ajdovščini, Sanaboru in pri Colu (Šašel 1975, 83, kat. št. 38-41), več jih je bilo odkritih tudi drugje ob glavni cesti proti Akvileji (Tiussi 2010). O posvetilni funkciji miljnikov: Witschel 2002, 361, 364-366. 142 Kos 2012, 296, 300. Primerjaj še Ciglenečki, Milavec 2009; Ciglenečki 2015, 391, 394–395,402–403; Pflaum 2004, 147; Kos 2014a. on the other. We believe that mutatio Castra can be seen as a larger settlement with a road station and a fortified centre, the latter also housing an army garrison. Given the general situation towards the end of the 3rd and the 4th century, as well as its location along the key access route to Italy, the Castra fortress probably perma­nently hosted an army garrison.137 Peter Petru was of the opinion that the settlement became the headquarters of the central part of the Claustra Alpium Iuliarum barrier system during its construction in the mid-4th century.138 The analysis of the coin finds from the Castra and Ad Pirum forts has led Peter Kos to interpret the increased activity at both sites as a reflection of the administrative and military reforms that Diocletian passed in the begin­ning of the 4th century and also of the need for added protection of Italy’s north-eastern border.139 The Castra fortress may have been used as a storage and supply centre for the troops stationed at the nearby linear defences.140 The army merely took advantage of the favourable strategic position and trading network already present at Fluvio Frigido, mirrored in the rich supply of goods from distant markets. Three milestones honouring contemporary em­perors and found in proximity to Ajdovščina reveal the significance of the road across the Hrušica Pass in the 4th century.141 The results of the numismatic analyses in recent years have shown fresh influx of coins into the fortress traceable up to the first three decades of the 5th century.142 The pottery finds, on the other hand, reveal that life in the fort and its vicinity continued, at least in a diminished capacity, into the second half of the 5th century.143 More­over, the remains of early medieval buildings constructed in the ruins of the Atrium House show that the settlement continued to be inhabited in the 6th and 7th centuries. Translation: Andreja Maver 137 The excavations at the Atrium House yielded several pieces of weaponry and military belts: Svoljšak et al. 2013, 70 (Cat. Nos. 10-21) and 73 (Cat. Nos. 7-18); Pflaum 2000, 89-90; 2004, 263; Ciglenečki, Milavec 2009. More military items came to light during the 2018–2019 excavations in the centre of the Late Roman Castra fortress (unpublished). 138 Šašel, Petru 1971, 98-99: giving the opinion of Stuc-chi in Fn. 9 that it functioned as a supply base for mountain army troops. Nada Osmuk (1998) also accepted the hypoth­esis on an army fort. 139 Kos 2014b, 130-131. 140 Already Stucchi 1946, 129. 141 Milestones dating to the time of Constantius and Ga-lerius, Julian the Apostate, as well as Valentinian and Valens were found at Miren, Ajdovščina, Sanabor and near Col (Ša­šel 1975, 83, Cat. Nos. 38-41), more at other locations along the main road towards Aquileia (Tiussi 2010). For a dedica­tory function of milestones, see Witschel 2002, 361, 364-366. 142 Kos 2012, 296, 300. 143 Also cf. Ciglenečki, Milavec 2009; Ciglenečki 2015, 391, 394–395,402–403; Pflaum 2004, 147; Kos 2014a. KATALOG NAPISOV / CATALOGUE OF INSCRIPTIONS Vesna Tratnik, Tina Žerjal, Marjeta Šašel Kos, Andreja Maver Okrajšave / Abbreviations GMK = Goriški muzej Kromberk – Nova Gorica, Muzejska zbirka Ajdovščina / Ajdovščina Museum Collection MPGA = Musei Provinciali di Gorizia, palača Attems / Palazzo Attems 1. Nagrobnik Antonija Valentina Nagrobna stela s poglobljenim napisnim poljem, iz nabrežinskega apnenca. Mere: ohranjene 124 cm viš., 60 cm šir. in 15 cm deb. Stela je bila preuporabljena že v rimskem času, v ta namen je bila stanjšana in izdelana je bila luknja za spojko. Stranski ploskvi sta verjetno originalni, grobo obde­lani, zelo grobo v spodnjem delu. Stela je bila mogoče tudi skrajšana zgoraj in spodaj. Najdišče: Ajdovščina. Odkrito 13. ali 14. aprila leta 1881 med izkopom javnega kanala v bližini stolpa 1 na jugozahodnem vogalu poznorimskega obzidja. Verjetno pred hišo Godina. Hkrati sta bila odkrita še spomenika kat. št. 3 in 4. Hrani: MPGA. Pridobljen leta 1899. h. height w. width th. thickness 1. Tombstone for Antonius Valentinus Tombstone with a sunken inscription field, of Aurisina limestone. Size: 124 cm in surviving h., 60 cm in w. and 15 cm in surviving th. The slab was reworked and fitted with a bracket clamp hole in the lower part for reuse at a later time in the Roman period. The side surfaces appear original, coarsely dressed, the bottom part of the tombstone is left rough. The top and bottom ends may have been removed. Find context: Ajdovščina. Unearthed on April 13 or 14, 1881, during earthworks near Tower 1, at the SW corner of the Late Roman fortification wall (probably in front of the Godina house). Unearthed together with the monuments under Cat. Nos. 3 and 4. Kept in: MPGA. Came to the museum in 1899. Besedilo: Antonio Va[len]­tino princi[pi] leg(ionis) XIII Gem(inae) int[er]­fecto a latro[ni]­bus in Alpes Iul[ias] loco quod appella­tur Scelerata Antonius Valen­tinus filius patr*[i] Prevod: Sin Antonij Valentin [je dal postaviti] (nagrobni spomenik) očetu Antoniju Valentinu, poveljniku XIII. legije Gemine. Ubili so ga razbojniki v Julijskih Alpah, na mestu/kraju, ki se imenuje Prekleto (= z zločinom oskrunjeno). Oddelki XIII. legije Gemine (“dvojne”: legija, ki je sestavljena iz ostankov dveh legij) so najbrž nadzorovali ta nevarni odsek glavne prometnice Akvileja–Emona–Podonavje (oz. Sirmij) okoli po­stojanke Ad Pirum (Hrušica) v okviru posebnega vojaškega območja (praetentura Italiae et Alpium), ustanovljenega za neposredno obrambo Italije v času markomanskih vojn. Datacija: verjetno druga polovica 2. st. ali 3. st. Glavna literatura: Inscr. It. X, 4, 339 = Suppl. It. 10, str. 235 = EDR145642 = HD021084 = lupa16409. Objave: L’Indipendente (tržaški časopis), 15. junij 1881, št. 1456; Benussi 1882, 170; = Suppl. It. 58 = Pais 1110; Maionica 1885, 7; Gregorutti 1892, 50, št. 85; Sticotti 1906, 172–175, št. 2; = ILS 2646; Degrassi 1954, 147– 148: datira v markomanske vojne ali začetek 3. st.; = ILJug 451; Šašel 1975–1976, 614 (=1992, 445): datira v 2. st.; Šašel 1975, 95, 121: do sredine 3. st.; Pavan 1979, 500—501: 3. st./ni starejše od severskega obdobja; = Inscr. Aq. 2785; Zaccaria 1992, 235: verjetno druga polovica 2. st., princeps označuje primus princeps prior; Petolescu 1996, št. 146; Šašel Kos 1997a, 194: imenska formula brez praenomen kaže na kasnejšo datacijo spomenika; prisotnost legije povezana z območjem praetentura Italiae; Cerva 1998, 10, št. 2; = AE 1998, 0546; Šašel Kos 1999b, 198: datira v čas markomanskih vojn ali v poznejši čas; Hope 2001, 139; Lafer 2003, 81–83, št. 25; = AE 2003, 128; Svoljšak et al. 2013, 57 s fotografijo: datacija v 2. st.; Bratož 2014, 25–26, op. 58 in str. 38, posebej op. 98; Šašel Kos 2016, 217; EDR145642 (Cigaina, zadnji popravki 11. 4. 2019): da­tira med letoma 171 in 250 (na osnovi palaeographia; nomina; historia, antiquitates); Krašna 2019, 57–58, št. 7. Text: Antonio Va[len]­tino princi[pi] leg(ionis) XIII Gem(inae) int[er]­fecto a latro[ni]­bus in Alpes Iul[ias] loco quod appella­tur Scelerata Antonius Valen­tinus filius patr*[i] Translation: Son Antonius Valentinus (had this monument set up) for his father Antonius Valentinus, princeps of the Thirteenth Legion Gemina, killed by robbers at a place called Accursed in the Julian Alps. Units of legio XIII Gemina (Gemina meaning double, i.e. legion composed of the remains of two legions) were probably stationed in the area of the Ad Pirum (Hrušica) fort to control the dangerous section of the main road leading from Aquileia to Emona and further on to the Danube Basin (Sirmium) as part of the military zone (praetentura Italiae et Alpium) established to directly control Italy’s borders during the Marcomannic Wars. Dating: probably second half of the 2nd or the 3rd c. Principal publication: Inscr. It. X, 4, 339 = Suppl. It. 10, p. 235 = EDR145642 = HD021084 = lupa 16409. Bibliography: L’Indipendente (Triestine newspaper), 15. 06. 1881, No. 1456; Benussi 1882, 170; = Suppl. It. 58 = Pais 1110; Maionica 1885, 7; Gregorutti 1892, 50, No. 85; Sticotti 1906, 172–175, No. 2; = ILS 2646; Degrassi 1954, 147–148: Marcomannic Wars or early 3rd c.; = ILJug 451; Šašel 1975–1976, 614 (= 1992, 445): 2nd c.; Šašel 1975, 95, 121: up to the mid-3rd c.; Pavan 1979, 500—501: 3rd c./not before the Severan period; = Inscr. Aq. 2785; Zaccaria 1992, 235: probably second half of the 2nd c., princeps stands for primus princeps prior; Petolescu 1996, No. 146; Šašel Kos 1997a, 194: ono-mastic formula without the praenomen indicates a later date; presence of the legion connected with praetentura Italiae; Cerva 1998, 10, No. 2; = AE 1998, 0546; Šašel Kos 1999b, 198: time of the Marcomannic Wars or later; Hope 2001, 139; Lafer 2003, 81–83, No. 25; = AE 2003, 128; Svoljšak et al. 2013, 57 with a photograph: 2nd c.; Bratož 2014, 25–26, Fn. 58 and p. 38, in particular Fn.98; Šašel Kos 2016, 217; EDR145642 (Cigaina, last cor­rections on 11. 4. 2019): between 171 and 250 (based on palaeographia; nomina; historia, antiquitates); Krašna 2019, 57–58, No. 7. 2. Nagrobnik Publija Publicija Ursiona Nagrobna stela ali oltar iz nabrežinskega apnenca, original-no ohranjena je samo sprednja stranica. Mere: ob odkritju 122 cm viš. × 44 cm šir., do danes je ohranjen odlomek velikosti 41 cm viš. × 30 cm šir. × 16 cm deb. Najdišče: Ajdovščina. Ob odkritju leta 1818 je bil vzidan v podboj hiše nekega strojarja (kjer ga je Hattinger prerisal), konec 19. stoletja ga je Gregorutti pregle­dal vzidanega v notranjosti hiše št. 70. Kasneje so ga vzidali v mestno hišo in leta 1949 razbili. Hrani: GMK (inv. št. AG 3981). Besedilo (nekdanje): P(ublius) Public(ius) Ursio v(ivus) s(ibi) f(ecit) et coniugi kariss(imae) Voltiliae Satunn[ae?] Dum saltus pu­blicos curo d[e]­cidi hoc in pri­vato agello. Prevod: Publij Publicij Ursion je dal za življenja postaviti (nagrobnik) sebi in najdražji soprogi Voltiliji Satuni. Medtem ko sem (prej) skrbel za javne gorske pašnike, (zdaj) počivam na tej majhni zasebni parceli. 2. Funerary monument for Publius Publicius Ursio Fragment of a tombstone or funerary altar, of Aurisina lime­stone, broken off on all sides but the front. Size: 122 cm in surviving h. and 44 cm in surviving w. upon discovery, but only a 41 × 30 × 16 cm large fragment survives today. Find context: Ajdovščina. Found in 1818 built into the door frame of the house of a tanner (where Hattinger drew it). In the late 19th century, Gregorutti examined it when it was built into the interior wall of the house No. 70. It was later moved again to the town hall and in 1949 broken up. Kept in: GMK (Inv. No. AG 3981). Text (upon discovery): P(ublius) Public(ius) Ursio v(ivus) s(ibi) f(ecit) et coniugi kariss(imae) Voltiliae Satunn[ae?] Dum saltus pu­blicos curo d[e]­cidi hoc in pri­vato agello. Translation: Publius Publicius Ursio had (this tombstone) put up in his lifetime for himself and his very dear wife Voltilia Satunna. While I (once) cared for public woodlands, I (now) rest on this small private plot. Domneva se, da so se omenjene javne posesti razprostirale v bližini njegovega prebivališča v Ajdovščini, torej na območju Vipavske doline ali v hribovitih pašniških in gozdnatih predelih Trnovskega gozda. Javni gorskipašniki, očitno v Vipavski dolini (Šašel domneva območje na črti Otlica–Predmeja–Col), so bili bodisi last mesta Akvileje (Degrassi, Panciera) bodisi rimskedržave (Šašel). Datacija: verjetno druga polovica 2. st. Glavna literatura: Inscr. It. X, 4, 340 = CIL V, 715 = Suppl. It. 10, str. 235–236 = EDCS-04200807. Objave: Müllner 1879, 248, št. 98; = Pais 1107 = ILS 6682; Maionica 1885b, 7; Gregorutti 1892, 48, št. 83; Sti­cotti 1906, 169, op. 1; Sticotti 1908, 285; Degrassi 1954, 25; Panciera 1957, 32; Petru 1975b; Zaccaria 1979, 216; Panciera 1979, 404: 2. ali 3. stoletje; Šašel1980, 183–184; Šašel 1984, 119–121; Svoljšak, Žbona Trkman 1986, 389-390, št. 4, 393, T. 1; Zaccaria 1992, 235–236; Zaccaria 2003, 323; Vidrih Perko, Žbona Trk-man 2003–2004, 60–61, sl. 3; Zaccaria 2009b, 259–260; Svoljšak et al. 2013, 58; Krašna 2019, 54, št. 6. 3. Nagrobik Tita Azinija Odlomek nagrobne stele iz nabrežinskega apnenca. Mere: 24 cm viš. × 107 cm šir. × 33 cm deb. Sprednja, zadnja in stranski ploskvi so verjetno originalni. Stranski ploskvi sta grobo obdelani. Stela s portretno nišo ima na arhitravu napis, v levem kotu niše je v reliefu viden del kapitela levega pilastra. Najdišče: Ajdovščina. Odkrito leta 1881 hkrati s spomeni­koma kat. št. 1 in 4 med izkopom javnega kanala v bližini stolpa 1 na jugozahodni strani poznorimskega obzidja. Hrani: MPGA. Pridobljen leta 1899. It is presumed that the public woodlands mentioned in the text were located close to his place of residence at Ajdovščina, i.e. in the Vipava Valley or the hilly pastures and woodlands of Trnovski gozd. The publicwoodlands (Šašel presumes the area along the Otlica– Predmeja–Col line) were owned either by the town of Aquileia (Degrassi, Panciera) or by the Roman state(Šašel). Dating: probably second half of the 2nd c. Principal publication: Inscr. It. X, 4, 340 = CIL V, 715 = Suppl. It. 10, pp. 235–236 = EDCS-04200807. Bibliography: Müllner 1879, 248, No. 98; = Pais 1107 = ILS 6682; Maionica 1885b, 7; Gregorutti 1892, 48, No. 83; Sticotti 1906, 169, Fn. 1; Sticotti 1908, 285; Degrassi 1954, 25; Panciera 1957, 32; Petru 1975b; Zaccaria 1979, 216; Panciera 1979, 404: 2nd or 3rd c.; Šašel 1980, 183–184; Šašel 1984, 119–121; Svoljšak, Žbona Trkman 1986, 389-390, No. 4, 393, Pl. 1; Zaccaria 1992, 235–236; Zaccaria 2003, 323; Vidrih Perko, Žbona Trkman 2003–2004, 60–61, Fig. 3; Zaccaria 2009b, 259–260; Svoljšak et al. 2013, 58; Krašna 2019, 54, No. 6. 3. Tombstone for Titus Asinius Fragment of a tombstone, of Aurisina limestone. Size: 24 cm in surviving h., 107 cm in w. and 33 cm in th. The front, back and side surfaces appear original, the sides coarsely dressed. It is a fragment of a stela / Porträtstele, with inscriptions on the architrave and presumable remains of the left pilaster capital in relief bordering the panel with portraits, also in relief, possibly one under each of the three inscribed names. Find context: Ajdovščina. Found in 1881 together with the monuments under Cat. Nos. 1 and 4 during earthworks near Tower 1, at the SW corner of the Late Roman fortification wall. Kept in: MPGA. Came to the museum in 1899. Besedilo: T(itus) Asinius Sessidena C(ai) f(ilia) Asinia T(iti) l(iberta) L(ucii) f(ilius) pater Maxuma mater Grata Vr 2: UM in TE v ligaturi Prevod: Oče Tit Azinij, Lucijev sin. Mati Sesidena Maksuma, Gajeva hčerka. Azinija Grata, Titova osvobojenka. Datacija: konec 1. st. pr. n. št. Gentilicij Sessidena razen na tem mestu v imperiju ni izpričan. Glavna literatura: Inscr. It. X, 4, 341 = Suppl. It. 10, str. 236 = EDCS-04600153 (napačno pod Tergeste). Objave: = Pais, Suppl. It. 59 = Pais 1111; Maionica 1885b, 8; Gregorutti 1892, 51-52, št. 86; Sticotti 1906, 182, št. 7; Petru 1975b, 121; Zaccaria 1992, 236, št. 341: zadnja četrtina 1. st. pr. n. št.; Zaccaria 2009a, 90; Zaccaria 2009b, 160; Svoljšak et al. 2013, 57 s fotografijo; Krašna 2019, 93, št. 30. 4. Nagrobnik Tarokona Briscinija Nagrobna stela iz nabrežinskega apnenca. Mere: 121 cm viš. × 55 cm šir. × 20 cm deb. Sprednja ploskev je gladka, stranski sta grobo obdelani. Stela je bila preuporabljena že v rimskem času, odstranjen je bil zgornji del, v spodnjem delu so bile izdelane dve luknji za spojki in dve za čep. Ena od teh je v poglobljenem kvadratu, druga ima vlivni kanal. Napisno polje je brez okvirja, nad njim je pro-filiran zatrep z zaklinki. V zatrepu je glava Meduze (gorgoneion), v vsakem zaklinku je izklesan delfin, ki gleda navzdol. Pod napisom je upodobljen fovč (falx arboraria ali falx vinitoria), za katerega Sticotti meni, da nima simboličnega pomena, ampak izpoveduje lastnikov poklic. Najdišče: Ajdovščina. Odkrito leta 1881 hkrati s spome­nikoma kat. št. 1 in 4 med izkopom javnega kanala v bližini stolpa 1 na jugozahodni strani poznorimskega obzidja. Hrani: MPGA. Pridobljen leta 1899. Besedilo: Taroco Briscinius Tropi f(ilius) iusit sibi ponere titulum et inperavit arbitratu Rufi, L(ucii) L(ucii) filiorum Quartae Freiae M(arci) f(iliae) nuru suae inpesa sua. Vr 4: pravilno imperavit Text: T(itus) Asinius Sessidena C(ai) f(ilia) Asinia T(iti) l(iberta) L(ucii) f(ilius) pater Maxuma mater Grata Line 2: UM ligatured to TE Translation: Father Titus Asinius, son of Lucius. Mother Sessidena Maksuma, daughter of Gaius. Asinia Grata, freedwoman of Titus. Dating: late 1st c. BC. The inscription provides the only example of the gentilicium Sessidena in the Roman Empire. Principal publication: Inscr. It. X, 4, 341 = Suppl. It. 10, p. 236 = EDCS-04600153 (erroneously under Tergeste). Bibliography: = Pais, Suppl. It. 59 = Pais 1111; Maionica 1885b, 8; Gregorutti 1892, 51-52, No. 86; Sticotti 1906, 182, No. 7; Petru 1975b, 121; Zaccaria 1992, 236, No. 341: last quarter of the 1st c. BC; Zaccaria 2009a, 90; Zaccaria 2009b, 160; Svoljšak et al. 2013, 57 with a photograph; Krašna 2019, 93, No. 30. 4. Tombstone for Taroco Briscinius Tombstone, of Aurisina limestone. Size: 121 cm in surviving h., 55 cm in w. and 20 cm in th. The front is smoothed, the sides are coarsely dressed. The up­per end was removed and the slab fitted with two bracket clamp and two dowel holes for reuse at a later time in the Roman period; one of the dowel holes lies within a slightly sunken square and the other has a pour channel. The unframed inscription panel is topped by a framed pediment with spandrels. The pediment holds a gor­goneion, while an outward facing dolphin is carved in each of the spandrels. A falx arboraria or falx vinitoria is shown below the inscription, which Sticotti sees as the attribute of the deceased man’s profession rather than an object od symbolic significance. Find context: Ajdovščina. Unearthed in 1881 together with the monuments under Cat. Nos. 1 and 4 during earthworks near Tower 1, at the SW corner of the Late Roman fortification wall. Kept in: MPGA. Came to the museum in 1899. Text: Taroco Briscinius Tropi f(ilius) iusit sibi ponere titulum et inperavit arbitratu Rufi, L(ucii) L(ucii) filiorum Quartae Freiae M(arci) f(iliae) nuru suae inpesa sua. Line 4: correctly imperavit Prevod: Tarokon Briscinij (morda: Taron Kobriscinij), Tropov sin, je ukazal in zapovedal sebi postaviti nagrobnik po razsoji Lucijevih sinov Rufa in Lucija. Svoji snahi Kvarti Freji, Markovi hčerki, (je dal vklesati napis) na svoje stroške. Taroco in Rufus sta domači/staroselski osebni imeni (prae­nomina). Za družinsko ime Briscinij prim. Brissinius s posvetila Eji iz Nezakcija (Inscr. It. X 1, 659); očitno gre za domače severnojadransko ime. Možno čitanje ime­na v obliki Kobriscinius (ni namreč ločilnih znamenj) ima prav tako paralelo v venetskem imenoslovju. Datacija: prva polovica 1. st. n. št. Glavna literatura: Inscr. It. X, 4, 342 = Suppl. It. 10, str. 236 = EDCS-04600154 (napačno pod Tergeste). Objave: = Pais, Suppl. It. 60 = Pais 1112; Maionica 1885, 8; Gregorutti 1892, 52, št. 87; Sticotti 1906, 175–177, št. 3; Zaccaria 1992, 236, št. 342: prva polovica 1. st.; Vidrih Perko 2003, 49–56; Zaccaria 2007b, 324; Buora 2007, 247–248; Ventura 2008, 79–80, sl. 7; Zaccaria 2009a, 90; Zaccaria 2009b, 260; Svoljšak et al. 2013, 59 s fotografijo; Krašna 2019, 76–77, št. 19. Translation: Taroco Briscinius (or Taro Cobriscinius), son of Tropus, ordered that the monument be put up for him­self at the discretion of Rufus and Lucius, sons of Lucius. (He had the inscription carved) to his daughter-in-law Quarta Freia, daughter of Marcus, from his own funds. Taroco and Rufus are local/indigenous praenomina. The family name of Briscinius is comparable to that of Brissinius, from the dedication to Eia at Nesactium (Inscr. It. X 1, 659), suggesting a local, north Adriatic name. The possible reading as Cobriscinius, allowed by the absence of punctuation, also has a parallel in the Venetic onomastics. Dating: first half of the 1st c. AD. Principal publication: Inscr. It. X, 4, 342 = Suppl. It. 10, p. 236 = EDCS-04600154 (erroneously under Tergeste). Bibliography: = Pais, Suppl. It. 60 = Pais 1112; Maionica 1885, 8; Gregorutti 1892, 52, No. 87; Sticotti 1906, 175–177, No. 3; Zaccaria 1992, 236, No. 342: first half of the 1st c.; Vidrih Perko 2003, 49–56; Zaccaria 2007b, 324; Buora 2007, 247–248; Ventura 2008, 79–80, Fig. 7; Zaccaria 2009a, 90; Zaccaria 2009b, 260; Svoljšak et al. 2013, 59 with a photograph; Krašna 2019, 76–77, No. 19. 5. Odlomek nagrobnika Nagrobnik, odlomek z napisom. Najdišče: najden leta 1871 na dvorišču graščine v pozno­ rimski utrdbi. Hrani: izgubljen. Besedilo: C. Mutil[ius? ---] Varus pater [--- Prevod: Gaj Mutilij (?) Var, oče ... Objave: Inscr. It. X, 4, 343 = Suppl. It. 10, str. 236. Sticotti 1908, 288 (po Kandlerju); Zaccaria 1992, 236. 6. Odlomek nagrobnika Nagrobnik iz apnenca, odlomek. Mere: 37 × 39 cm. Zdi se, da je napis v spodnjem desnem kotu ohranjen v celoti. Najdišče: Ajdovščina. Vzidan v tlak pred hišo Angelo Ca-sagrande (št. 2) (konec 19. stoletja po Gregoruttiju in Sticottiju; hiša Križaj po Petru 1975b). Hrani: izgubljen. Besedilo: [---]tri [---] matr(i). Vr 2: RT v ligaturi Prevod: ... očetu (?), ...materi. Objave: Inscr. It. X 4, 344 = CIL V 716 = Pais 1108 = EDCS-04200808. Maionica 1885, 7; Müllner 1879, 249, št. 99; Gregorutti 1892, 49, št. 84; Sticotti 1908, 287. 7. Odlomek nagrobnika Odlomek nagrobnika iz sivega marmorja. Poglobljeno na­pisno polje obdaja profiliran okvir. Odbit, originalno je ohranjena le sprednja ploskev. Mere: odlomek: 20 cm viš. × 18 cm šir. × 4,3 cm deb. Najdišče: med raziskavami zahodnega obzidja leta 1990, v zasutju nad severozahodnim vogalom temelja stolpa 3. Hrani: GMK, inv. št. AG 10925. Besedilo: -----­+[---] /ohr. črka je lahko E ali L./ DEL[---] Objave: Osmuk 1990a, 167, 169, sl. 34; Osmuk 1997, 120. 5. Funerary inscription Fragment of a funerary inscription. Find context: Found in 1871 in the courtyard of a mansion in the interior of the Late Roman fortress. Kept in: now lost. Text: C. Mutil[ius? ---] Varus pater [--­ Translation: Gaius Mutilius (?) Varus, father ... Bibliography: Inscr. It. X, 4, 343 = Suppl. It. 10, p. 236. Sticotti 1908, 288 (after Kandler); Zaccaria 1992, 236. 6. Funerary inscription Fragment of a funerary inscription, of limestone. Size: 37 × 39 cm. The inscription in the bottom right corner appears complete. Find context: Ajdovščina. Built into the floor in front of the Angelo Casagrande house (No. 2) (in the late 19th c. according to Gregorutti and Sticotti; the Križaj house according to Petru 1975b). Kept in: lost. Text: [---]tri [---] matr(i). Line 2: ligatured RT Translation: ... father (?), ...mother. Bibliography: Inscr. It. X 4, 344 = CIL V 716 = Pais 1108 = EDCS-04200808. Maionica 1885, 7; Müllner 1879, 249, No. 99; Gregorutti 1892, 49, No. 84; Sticotti 1908, 287. 7. Funerary inscription Fragment of a funerary inscription with a sunken inscription field within a moulded frame, of greyish marble. The slab was reworked, only front side appears original. Size: 20 × 18 × 4.3 cm. Find context: Ajdovščina. Unearthed during the 1990 inves­tigations at the west fortification wall, in the fill at the NW corner of Tower 3. Kept in: GMK, Inv. No. AG 10925. Text: +[---] /surviving letter either E or L/ DEL[---] Bibliography: Osmuk 1990a, 167, 169, Fig. 34; Osmuk 1997, 120. 8. Napis Napisna plošča iz nabrežinskega apnenca. Mere: 34,5 × 24 × 20,5 cm, višina črk 20 cm. Odbit, ohranjeni sta le sprednja in zadnja ploskev, debelina je verjetno originalna. Najdišče: Ajdovščina. Leta 1934 opisal Vito Loser v Brati­novi hiši. Leta 1965 ponovno najden med izkopavanji pod tedaj porušeno staro kinodvorano. Hkrati so bili odkriti tudi drugi deli gradbenih napisov, ki sodijo k isti ali različnim monumentalnim stavbam (glej še kat. št. 9 in 10). Hrani: GMK (inv. št. AG 85). Besedilo: ---]INIA[---/morda tudi ---]VINI[--- Datacija: Konec 1. st. ali začetek 2. st. Objave: Inscr. It. X 4, 346 = Suppl. It. 10, str. 236. Petru, Svoljšak 1965a; Petru, Svoljšak 1965b; Svoljšak,Žbona Trkman 1986, 385–386, št. 1, sl. 1; Zaccaria 1992, 236; Svoljšak et al. 2013, 63, št. 4, o svetišču str. 43: datacija v 1. st.; Krašna 2019, 127, št. 57. 9. Napis Dva odlomka plošče kompozitne arhitekture z napisom, iz nabrežinskega apnenca. Napis v profiliranem okvirju je enovrstični. Mere: manjši levi odlomek meri 42 × 25 × 10 cm, večji desni 74 × 60 × 11,5–17 cm, skupaj pa 74 cm viš. × 63 cm šir. × 10–17 cm deb.; črke so visoke 22,5 cm, okvir 29,5 cm. Plošča je odlomljena na levi strani in spodaj. Napis se je nadaljeval na levi in desni, desna plošča ni ohranjena. Luknja na nosilni ploskvi kaže, da sta bili plošči poveza­ni s spojko v obliki črke U. Sprednja ploskev je gladko obdelana v napisnem polju, bolj grobo nad in pod njim. 8 cm pod okvirjem na spodnji strani je vrezana vodo­ravna črta. Nosilna in desna stranska ploskev sta grobo obdelani. Sprednji rob desne stranice je gladko obdelan 8. Inscription Fragment of an inscription slab, of Aurisina limestone. Size: 34.5 × 24 × 20.5 cm, the letters measure 20 cm in h. The fragment is broken on all sides except the front and back; the latter surface and consequently slab th. ap­pear original. Find context: Ajdovščina. In 1934, Vito Loser described it when kept in the Bratina house. In 1965, it was rediscovered during the rescue excavations at the old cinema that had been pulled down. Unearthed together with other fragments belonging to one or several large buildings (Cat. Nos. 9, 10). Kept in: GMK (Inv. No. AG 85). Text: ---]INIA[---/ or ---]VINI[--­ Dating: 1st c. or late 1st/early 2nd c. Bibliography: Inscr. It. X 4, 346 = Suppl. It. 10, p. 236. Petru, Svoljšak 1965a; Petru, Svoljšak 1965b; Svoljšak,Žbona Trkman 1986, 385–386, No. 1, Fig. 1; Zaccaria 1992, 236; Svoljšak et al. 2013, 63, No. 4, on a temple p. 43: 1st c.; Krašna 2019, 127, No. 57. 9. Inscription Two fragments of an inscribed slab of composite architecture, of Aurisina limestone. The inscription runs in a single line within a moulded frame. Size: smaller left fragment measures 42 × 25 × 10 cm, larger right one 74 × 60 × 11.5–17 cm, together 74 × 63 × 10–17 cm; the letters are 22.5 cm and the inscription field 29.5 cm high. The slab is broken off on the left and bottom sides. The inscription continued on both sides, to the right on the missing adjacent slab; the surviving clamp hole on the bedding surface shows that the two slabs were held together by a bracket clamp. The front is smoothly dressed on the inscription field, coarsely above and below the moulded frame with a horizontal line carved s ploščatim dletom. Zadnja (notranja) ploskev je grobo obdelana v zgornjih 45 cm in zelo grobo pod tem pa-som, vmes je vklesana vodoravna črta. Odsotnost fascij na sprednji stranici, debelina odlomkov in zelo groba obdelava na spodnjem delu zadnje ploskve govorita proti interpretaciji odlomkov kot delov arhitrava oz. preklade. Odlomka sta bila del kompozitne arhitekture, glede na velikost črk postavljena na precejšnji višini v delu, kjer je bila zadnja ploskev popolnoma skrita. Najdišče: Ajdovščina, odkrit leta 1965 med izkopavanji kinodvorane hkrati z napisoma kat. št. 8 in 10 in odlomkom okrašenega napuščnega venca. Hrani: GMK (inv. št. AG 84). Besedilo: ---]RAM[--- Datacija: 1. st. ali konec 1. st. ali začetek 2. st. Objave: ILJug 1222a-b; Petru, Svoljšak 1965a, 197; Petru, Svoljšak 1965b, 125; Petru 1965, 139; Petru 1975b, 120; Svoljšak, Žbona Trkman 1986, 386-388, št. 2, sl. 2; Svoljšak et al. 2013, 63, št. 2, o svetišču str. 43: datacija v 1. st.; Krašna 2019, 128, št. 58. 10. Napis Odlomek plošče z gradbenim napisom iz nabrežinskega apnenca. Mere: 36 cm viš. × 54 cm šir. × 11,5 cm deb.; višina črk 22,5 cm. Zgoraj je profiliran okvir, pod njim so tri črke. Nad prvi-ma dvema je vklesana vodoravna črta, ki označuje številko šest (6). Napis se je začel na manjkajoči plošči, ki je bila na levi strani. Luknja na nosilni ploskvi kaže, da sta bili obe plošči povezani s spojko. Nosilna in zadnja ploskev sta grobo obdelani. Višina črk, profi­lacija okvirja napisnega polja in debelina odlomka so enaki kot pri odlomkih št. 9. 8 cm below the frame. The bedding and right side surfaces are coarsely dressed. The right surface has the front edge smoothly dressed with a flat chisel. The back is coarsely dressed in the upper 45 cm and left rough below; the two parts are separated by a carved horizontal line. The absence of fasciae on the front, the thickness of the fragments and the surface finish on the back refute the possibility of the slab representing an architrave. It did form part of composite architecture, located at a considerable height given the size of the letters and positioned so that the back was not visible. Find context: Ajdovščina. Found during the 1965 rescue exca­vations at the old cinema together with other fragments with inscriptions (Cat. Nos. 8, 10) and a fragment of a 30 cm thick decorated cornice. Kept in: GMK (Inv. No. AG 84). Text: ---]RAM[--­ Dating: 1st or late 1st/early 2nd c. Bibliography: ILJug 1222a-b; Petru, Svoljšak 1965a, 197; Petru, Svoljšak 1965b, 125; Petru 1965, 139; Petru 1975b, 120;Svoljšak, Žbona Trkman 1986, 386-388, No. 2, Fig. 2; Svoljšak et al. 2013, 63, No. 2, on a temple p. 43: 1st c.; Krašna 2019, 128, No. 58. 10. Inscription Fragment of an inscribed slab, of Aurisina limestone. Size: 36 × 54 × 11.5 cm; letters are 22.5 cm high. The inscription is carved within a moulded frame. A hori­zontal line is carved above the first two surviving letters of the inscription, marking the number six. It began on the missing adjacent slab to the left. The surviving clamp hole on the bedding surface shows that the two slabs were held together by a bracket clamp. The bed­ding and back surfaces are coarsely dressed. The height Najdišče: Ajdovščina, odkrit leta 1965 med izkopavanji ki­nodvorane hkrati z napisoma kat. št. 8, 9 in odlomkom okrašenega napuščnega venca. Hrani: GMK (inv. št. AG 4149). Besedilo: ---]· VVI V[--- Datacija: 1. st. ali konec 1. oz. začetek 2. st. Objave: ILJug 1222c; Petru, Svoljšak 1965a, 197; Petru, Svoljšak 1965b, 125; Petru 1965, 139; Petru 1969, 17; Petru 1972, 350; Petru 1975b, 120; Svoljšak, Žbona Trkman 1986, 388-389, št. 3, sl. 3; Svoljšak et al. 2013, 63, št. 3, o svetišču str. 43: datacija v 1. st.; Krašna 2019, 128, št. 58. 11. Miljnik Miljnik iz sivega apnenca. Spodaj odlomljen. Mere: viš. odlomka 54,5 cm, s prem. 39,5 cm. Na zgornji ploskvi ima okroglo vdolbino za pritrditev do-datnega kosa. Površina kamna je precej poškodovana, ponekod so vidni ostanki malte. Najdišče: Ajdovščina. Leta 1908 je bil najden med kopan­jem za mestni vodovod, v globini 1 m, ob glavni cesti (tedanja Goriška ulica, današnja Gregorčičeva). Pris-lonjen je bil na vogal zgradbe Štefana Štekarja (hiša št. 50). Ostanki malte kažejo, da miljnik ni bil najden in situ. Vzidan morda še v poznorimskem obdobju v obzidje utrdbe. Hrani: Narodni muzej Slovenije (inv. št. L 138). Besedilo: Dd(ominis) nn(ostris) Fl(avio) Val(erio) Constantio et Galerio Valerio Maxi miano invictis Augg(ustis) et Valeri(i)s Severo et Maxi mi[no nobil]issimis [Ceas(aribus) --­of the letters, the inscription frame moulding and the thickness of the fragment are the same as those of the slab No. 9. Find context: Ajdovščina. Found during the 1965 rescue excavations at the old cinema together with other frag­ments with inscriptions (Cat. Nos. 8, 9). Kept in: GMK (Inv. No. AG 4149). Text: ---]· VVIV[--­ Dating: 1st c. or late 1st/early 2nd c. Bibliography: ILJug 1222c; Petru, Svoljšak 1965a, 197; Petru, Svoljšak 1965b, 125; Petru 1965, 139; Petru 1969, 17; Petru 1972, 350; Petru 1975b, 120; Svoljšak, Žbona Trkman 1986, 388-389, No. 3, Fig. 3; Svoljšak et al. 2013, 63, No. 3, on a temple p. 43: 1st c.; Krašna 2019, 128, No. 58. 11. Milestone Milestone, of grey limestone, the upper part survives. Size: 54.5 cm in surviving height, 39.5 cm in diameter. There is a circular hollow on top for fixing an additional piece. The surface of the milestone is quite damaged, with traces of mortar visible in places. Find context: Ajdovščina. Found in 1908 during the earth­works for the municipal water supply system, at a depth of a metre along the main street (then Goriška ulica, now Gregorčičeva ulica). It was leaning againstthe corner of the building of Štefan Štekar (house No. 50), given the traces of mortar presumably not in situ. It may have been built into the fortification wall already during the Late Roman period. Kept in: Narodni muzej Slovenije (Inv. No. L 138). Prevod: Našima gospodoma Flaviju Valeriju Konstanciju in Galeriju Valeriju Maksimianu, nepremagljivima avgustoma, in Valerijanoma Severu in Maksiminu, najplemenitejšima cezarjema ... Datacija: Miljnik iz časa tetrarhije vladarjev avgustov Konstancija Klora in Galerija ter cezarjev Valerija Severa in Valerija Maksimina Dačana je datiran v leto 305 ali 306. Glavna literatura: Inscr. It. X, 4, 379 = Šašel 1975, 83, št. 39 = Suppl. It. 10, 238 = RINMS 176 = EDR007205 = EDCS-04600172 = lupa, št. 23481. Objave: Sticotti 1908, 283-284; Petru 1975b, 120; Zaccaria 1992, 238; Šašel Kos 1997a, 469–471, RINMS 176; Šašel Kos 2004, št. 176 in spremna beseda o miljni­kih; Witschel 2002, 386, št. 14, o posvetilni funkciji miljnikov str. 361, 364-366; Svoljšak et al. 2013, 12. Text: Dd(ominis) nn(ostris) Fl(avio) Val(erio) Constantio et Galerio Valerio Maxi miano invictis Augg(ustis) et Valeri(i)s Severo et Maxi mi[no nobil]issimis [Ceas(aribus) --­ Translation: To our lords Flavius Valerius Constantius, and Galerius Valerius Maximianus, invincible Augusti, and to the Valerii: Severus and Maximinus, the most noble Caesars ... Dating: AD 305 or 306, during the tetrarchy of Constantius Chlorus and Galerius as Augusti, Valerius Severus and Maximinus Daia as Caesars. Principal publication: Inscr. It. X, 4, 379 = Šašel 1975, 83, No. 39 = Suppl. It. 10, 238 = RINMS 176 = EDR007205 = EDCS-04600172 = lupa, No. 23481. Bibliography: Sticotti 1908, 283-284; Petru 1975b, 120; Zaccaria 1992, 238; Šašel Kos 1997a, 469–471, RINMS 176; Šašel Kos 2004, No. 176 with an introduction on milestones; Witschel 2002, 386, No. 14, on the dedicatory function of milestones pp. 361, 364-366; Svoljšak et al. 2013, 12. Okrajšave / Abbreviations AE = L‘Année épigraphique. CIL = Corpus inscriptionum Latinarum. EDCS =Epigraphik-Datenbank Clauss / Slaby [http://db.edcs. eu/epigr/epi.php?s_sprache=en]. EDH = Epigraphische Datenbank Heidelberg [https://edh-www. adw.uni-heidelberg.de/home?&lang=de]. EDR = Epigraphic Database Rome [http://www.edr-edr.it]. FMRSl I = P. Kos, Die Fundmünzen der römischen Zeit in Slowenien I, Berlin 1988. FMRSl II = P. Kos, Die Fundmünzen der römischen Zeit in Slowenien II. – Berlin 1988. FMRSl III = P. Kos. A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien III, Berlin 1995. FMRSl IV = A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien IV. – Berlin 1998. FMRSl V = A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien V. – Mainz 2005. FMRSl VI = A. Šemrov, in P. Kos, Die Fundmünzen der rö­mischen Zeit in Slowenien VI (Collection Moneta 110). –Wetteren, Ljubljana 2010. ILJug = A. et J. Šašel, Inscriptiones Latinae quae in Jugoslavia inter annos MCMXL et MCMLX repertae et editae sunt (Situla 5), Ljubljana 1963; iidem, Inscriptiones Latinae quae in Jugoslavia inter annos MCMLX et MCMLXX repertae et editae sunt (Situla 19), 1978; iidem, Inscriptiones Latinae quae in Jugoslavia inter annos MCMII et MCMXL repertae et editae sunt (Situla 25), 1986. ILS = Inscriptiones Latinae selectae, ed. H. Dessau, Berlin 1892–1916. Inscr. It. X, 4 = Inscriptiones Italiae, vol. X – regio X, fasciculus IV – Tergeste, curavit Petrus Sticotti, Roma 1951. Inscr. Aquil. = J. B. Brusin, Inscriptiones Aquileiae, I–III, Udine 1991–1993. lupa = UBI ERAT LUPA, F. und O. Harl, www.ubi-erat-lupa. org (Bilddatenbank zu antiken Steindenkmalern). Opera selecta 1992 = J. Šašel, Opera selecta (Situla 30), Ljub­ljana 1992. Pais = E. Pais, Corporis inscriptionum Latinarum supplementa Italica, Rom 1884. RINMS = M. Šašel Kos, The Roman Inscriptions in the National Museum of Slovenia / Lapidarij Narodnega muzeja Slovenije (Situla 35), Ljubljana 1997. Suppl. It. = Supplementa Italica, Rome 1981–. ALFÖLDY, G. 1978, Ein ‘nordadriatischer’ Gentilname und seine Beziehungen. – Zeitschrift für Papyrologie und Epi­graphik 30, 123–136. ALFÖLDY, G. 1999, Städte, Eliten und Gesellschaft in der Gal-lia Cisalpina. Epigraphisch-historische Untersuchungen. – Heidelberger Althistorische Beiträge und Epigraphische Studien 30, 21–33. BENUSSI, B. 1882, L‘Istria sino ad Augusto [parte I]. – Arche­ografo Triestino, ser. II/8 (1981–1982), 167–258. BOSIO, L. 1973, La Venetia 0rientale nella descrizione della Tabula Peutingeriana. – Aquileia Nostra 44, 37–83. BOSIO, L. 1991, Le strade romane della Venetia e dell'Histria. - Padova. BRATINA, P. 2010, Nova arheološka odkritja v Vipavski do-lini. – Goriški letnik 33-34/1, 155-180. BRATINA, P. 2013, Od najstarejših sledi do naselitve Slovanov. – V / In: J. Pavšič (ur. / ed.), Vipavska dolina: neživi svet, rastlinstvo, živalstvo, zgodovina, umetnostna zgodovina, gmotna kultura, gospodarstvo, naravovarstvo, Slovenske pokrajine 1 237–250, Ljubljana. BRATOŽ, R. 2000, Soča in prehodi čez reko v antiki. – Goriški letnik 27, 27–50. BRATOŽ, R. 2007, Rimska zgodovina 1. Od začetkov do nastopa cesarja Dioklecijana. – Zbirka Zgodovinskega časopisa 33, Knjižna zbirka Scripta. BRATOŽ, R. 2014, Med Italijo in Ilirikom. Slovenski prostor in njegovo sosedstvo v pozni antiki. – Zbirka Zgodovinskega časopisa 46, Dela I. razreda SAZU 39. BRATOŽ 2018, La battaglia del Frigidus (394 d.C.) nelle ricer-che degli ultimi vent‘anni. - V / In: S. Cavazza (ur. / ed.), Studi e ricerche per il LXXXIX convegno della Deputazione di storia patria per il Friuli, 9-60. BREZIGAR, B. 2017, Ajdovščina – arheološko najdišče Castra. – Varstvo spomenikov. Poročila 52, 11. BREZIGAR, B. 2018, Ajdovščina – arheološko najdišče Castra. – Varstvo spomenikov. Poročila 53, 11. BREZIGAR, B., D. JOSIPOVIČ 2015, Staroslovansko pokopali-šče v Šturjah (Ajdovščina). – Goriški letnik 37-38, 127-149. BUORA, M. 2007, Falces vinitoriae dell‘Italia nordorientale. – Aquileia nostra 78, 241–264. CALZOLARI, M. 2000, Gli itinerari della tarda antichita e il nodo stradale di Aquileia. – V / In: S. Blason Scarel (ur. / ed.), Cammina, cammina: dalla via dell'ambra alla via della fede, 18–42, Udine. CERVA, M. 1998, Sul brigantaggio nella X Regio. – Atti e Me-morie della Societa Istriana di Archeologia e Storia Patria n. s. 46, 7–28. CIGLENEČKI, S. 1997, Strukturiranost poznorimske poselitve Slovenije (Strukturierung spätantiker Besiedlung Sloweni-ens). – Arheološki vestnik 48, 191–202. CIGLENEČKI, S. 1999a, Results and Problems in the Ar­chaeology of the Late Antiquity in Slovenia / Izsledki in problemi poznoantične arheologije v Sloveniji. – Arheološki vestnik 50, 287–309. CIGLENEČKI, S. 1999b, Zidani braniki Italije. – V / In: B. Aubelj (ur. / ed.), Zakladi tisočletij, Zgodovina Slovenije od neandertalcev do Slovanov, 317–319, Ljubljana. CIGLENEČKI, S. 2015, Late Roman army, Claustra Alpium Iuliarum and the fortifications in the south-eastern Alps / Poznorimska vojska, Claustra Alpium Iuliarum in utrjena krajina v jugovzhodnih Alpah. – V / In: J. Istenič, B. La-harnar, J. Horvat (ur. / eds.), Evidence of the Roman Army in Slovenia / Sledovi rimske vojske na Slovenskem, Katalogi in monografije 41, 385–430. CIGLENEČKI, S., T. MILAVEC 2009, The defence of the north-eastern Italy in the first decennia of the 5th century. – Forum Iulii 33, 176-187. CUNTZ, O. 1902, Die römische Strasse Aquileia-Emona, ihre Stationen und Befestigungen. – Jahreshefte des Österreichis­chen Archäologisches Institutes in Wien 5, Beiblatt, 140–160. DEGRASSI, A. 1954, Il confine nord-orientale dell‘Italia romana. Ricerche storico topografiche. – Dissertationes Bernenses I/6. DJURIĆ, B. 1976, Antični mozaiki na ozemlju SR Slovenije (Les mosaiques antiques sur le territoire de la R. S. de Slovénie). – Arheološki vestnik 27, 537-625. DRAKSLER, M. 2018, Arheološke raziskave v Rustjevi hiši, Ajdovščina. – V / In: P. Stipančič, B. Djurić (ur. / eds.), Arheologija v letu 2017, dediščina za javnost. Zbornik povzetkov. Strokovno srečanje Slovenskega arheološkega društva, Ljubljana, Narodni muzej Slovenije – Metelkova, 7.–8. marec 2018, 16, Ljubljana. FABČIČ, V. 1997, Poizkus modularne rekonstrukcije po­znorimske utrdbe Castra (Versuch einer modularen Rekonstruktion der spätrömischen Befestigung Castra). – Arheološki vestnik 48, 131-142. FABEC, T. 2017, Ajdovščina. Mestno jedro in arheološko naj­dišče Castra. – Monografije CPA 4. FABEC, T., V. TRATNIK 2009, Ajdovščina. – Varstvo spome­nikov. Poročila 46/2009 (2010), 19–21. GASPARI, A. 2017, Deblak s konca 2. stoletja pr. n. št. iz Ljub­ljanice na Vrhniki / The late 2nd century BC logboat from the Ljubljanica river at Vrhnika. – Ljubljana. GREGORUTTI, C. 1892, L‘antico Timavo e le vie Gemina e Postumia. – Estratto da Archeografo Triestino N. S., 16–18. HITZINGER, P. 1861, Pläne römischer Orte in Krain. – Mittheilungen des historischen Vereines für Krain 16, 46–47. HOPE, V. M. 2001, Constructing identity. The Roman funerary monuments of Aquileia, Mainz and Nimes. – BAR. Inter­national Series 960. HORVAT, J. 1999, Roman Provincial Archaeology in Slovenia Following the year 1965: Settlement and Small Finds / Rimska provincialna arheologija v Sloveniji po letu 1965: poselitvena slika in drobna materialna kultura. – Arheološki vestnik 50, 215–257. HORVAT, J., A. BAVDEK 2009, Okra. Vrata med Sredozemljem in srednjo Evropo / Ocra. The gateway between the Mediter­ranean and Central Europe. – Opera Instituti Archaeologici Sloveniae 17. KASTELIC, J. 1987, Ajdovščina. Arheologija. – V/ In: Enciklo­pedija Slovenije 1, 28-29, Ljubljana. KOS, P. 1986, The Monetary Circulation in the Southeastern Alpine Region ca. 300 B. C. – A. D. 1000 (Denarni obtok na prostoru jugovzhodnih Alp 300 pr. n. št. – 1000).– Situla 24. KOS, P. 1997, Interpretacija (antičnih) novčnih najdb. Metodolo­gija - njene možnosti in pasti (Interpretation of (Roman) coin finds. Methodology: its potentials and pitfalls). – Arheološki vestnik 48, 97–115. KOS, P. 2012, The construction and abandonment of the Claustra Alpium Iuliarum defence system in light of the numismatic material / Gradnja ali opustitev obrambnega sistema Claustra Alpium Iuliarum v luči numizmatičnega gradiva. – Arheološki vestnik 63, 265-291. KOS, P. 2014a, Barriers in Julian Alps and Notitia Dignitatum / Zapore v Julijskih Alpah in Notitia Dignitatum.– Arheološki vestnik 65, 409-422. KOS, P. 2014b, Izgradnja zapornega sistema Claustra Alpium Iuliarum. Antični, arheološki in numizmatični viri / Con­struction of the Claustra Alpium Iuliarum fortifications. Historical, archaeological and numismatic sources. – V / In: J. Kusetič, P. Kos, A. Breznik, M. Stokin (ur. / eds.), Claustra Alpium Iuliarum – med raziskovanjem in upravl­janjem / Claustra Alpium Iuliarum – Between Research and Management, 112–132, Ljubljana. KOS, P. 2014c, Ad Pirum (Hrušica): Claustra Alpium Iulia-rum. – Zavod za varstvo kulturne dediščine Slovenije. Vestnik 26/1. KOS, P. 2017, Ajdovščina (Castra) – novčne najdbe iz naselbin­skih plasti zunaj obzidja. Pomen novcev za interpretacijo najdišča / Ajdovščina (Castra) – coin finds from the extra muros settlement. The significance of the coins for the interpretation of the site. – Arheološki vestnik 68, 295–323. KRAMAR et al. 2015 = S. KRAMAR, V. TRATNIK, I. M. HRO­VATIN, A. MLADENOVIĆ, H. PRISTACZ, N. ROGAN ŠMUC 2015, Mineralogical and chemical characterization of Roman slag from the archeological site of Castra (Ajdo­vščina, Slovenia). – Archaeometry 57/4, 704–719. KRAŠNA, A. 2019, Arhitekturni in sepulkralni izdelki iz nabre­žinskega apnenca v Sloveniji. – Diplomsko delo / BA thesis, Oddelek za arheologijo Filozofske fakultete Univerze v Ljubljani (neobjavljeno / unpublished). LAFER, R. 2003, Epigraphische Zeugnisse von latrones in der regio X und in Dalmatien. – V / In: K. Strobel, R. Lafer (ur. / eds.), Der Alpen-Adria-Raum in Antike und Spätantike. Die Geschichte eines historisch-geographischen Raumes im Spiegel der epigraphischen, literarischen, numismatischen und archäologischen Quellen. Akten der IV. Intern. Table Ronde zur Geschichte der Alpen-Adria-Region in der Antike, Altertumswissenschafliche Studien Klagenfurt 1, 75–92, Klagenfurt. MAIONICA, E. 1885, Programma del Ginnasio di Gorizia 1858. – Gorizia. MARCHESETTI, C. 1903, I castellieri preistorici di Trieste e della regione Giulia. – Atti del Museo civico di storia naturale 10/4, Trieste. MIKL CURK, I., S. CIGLENEČKI, D. VUGA 1993, Po poteh rimskih vojakov v Sloveniji / In the Footsteps of Roman Soldiers in Slovenia. – Ljubljana. MLINAR, M. 1999, Vipavska dolina v prazgodovini. – Diplom­sko delo / BA thesis, Oddelek za arheologijo Filozofske fakultete Univerze v Ljubljani (neobjavljeno / unpublished). MOSER, K. L. 1891, Funde in Wippachthale. - Mittheilungen der Anthropologischen Gesellschaft in Wien 35, 33. MÜLLNER, A. 1879, Emona. Archäologische Studien aus Krain. – Ljubljana. OSMUK, N. 1977, Ajdovščina. – Varstvo spomenikov 21, 199–202. OSMUK, N. 1979, Ajdovščina. – Varstvo spomenikov 22, 278. OSMUK, N. 1986a, Ajdovščina. – Varstvo spomenikov 28, 258-260. OSMUK, N. 1986b, Ajdovščina. – Varstvo spomenikov 28, 295. OSMUK, N. 1987a, Ajdovščina. – Arheološki pregled 28 (1989), 125-127. OSMUK, N. 1987b, Ajdovščina. – Varstvo spomenikov 29 (1989), 252. OSMUK, N. 1988, Ajdovščina. – Varstvo spomenikov 30, 233-235. OSMUK, N. 1990a, Ajdovščina. – Varstvo spomenikov 32, 167–169. OSMUK, N. 1990b, Obzidje rimske utrdbe Castra v Ajdovščini. – Arheološki vestnik 41, 183-198. OSMUK, N. 1990c, Ajdovščina. – Varstvo spomenikov 32, 198. OSMUK, N. 1991, Obnovljena antika v Ajdovščini – tri dese­tletja kasneje. – Varstvo spomenikov 33, 115–127. OSMUK, N. 1992, Ajdovščina – Castra. – Varstvo spomenikov. Poročila 34 191-192. OSMUK, N. 1993a, Ajdovščina, obzidje. – Varstvo spomenikov. Poročila 35/1993 (1995), 83. OSMUK, N. 1993b, Ajdovščina, zahodno grobišče. – Varstvo spomenikov. Poročila 35/1993 (1995), 83-84. OSMUK, N. 1993c, Ajdovščina, Mirce. – Varstvo spomenikov. Poročila 35/1993 (1995), 84. OSMUK, N. 1996a, Ajdovščina. Castra. – Varstvo spomenikov. Poročila 37/1996 (1998), 7. OSMUK, N. 1996b, Ajdovščina. Castra. – Varstvo spomenikov. Poročila 37/1996 (1998), 7. OSMUK, N. 1997, Ajdovščina - Castra. Stanje arheoloških raziskav (1994) (Ajdovščina - Castra. Forschungstand (1994)). – Arheološki vestnik 48, 119–130. OSMUK, N. 1999a, Ajdovščina. Castra – extra muros. – Var-stvo spomenikov. Poročila 38/1999 (2001), 5. OSMUK, N. 1999b, Ajdovščina. Castra. – Varstvo spomenikov. Poročila 38/1999 (2001), 5. OSMUK, N. 1999c, Ajdovščina. Castra – zah. grobišče. – Var-stvo spomenikov. Poročila 38/1999 (2001), 6. OSMUK, N. 1999d, Ajdovščina. Castra / Mirce. – Varstvo spomenikov. Poročila 38/1999 (2001), 6–7. OSMUK, N. 1999e, Ajdovščina. Mirce in Na gmajni. – Varstvo spomenikov. Poročila 38/1999 (2001), 7. OSMUK, N. 2003, Ajdovščina. - V / In: B. Djurić et al., Zemlja pod vašimi nogami. Arheologija na avtocestah Slovenije. Vodnik po najdiščih, 92, Ljubljana (= The Earth Beneath Your Feet. Archaeology on the Motorways in Slovenia. Guide to Sites, Ljubljana 2004). OSMUK, N. 2005a, Ajdovščina. Arheološko najdišče Castra. – Varstvo spomenikov. Poročila 42/2005 (2006), 7. OSMUK, N. 2005b, Ajdovščina. Arheološko najdišče Castra, extra muros – jugozahodno. – Varstvo spomenikov. Poročila 42/2005 (2006), 7-8. OSMUK, N. 2005c, Ajdovščina. Castra, infrastruktura – extra muros – zahod. – Varstvo spomenikov. Poročila 42/2005 (2006), 8. OSMUK, N. 2005d, Ajdovščina – Varstvo spomenikov. Poročila 42/2005 (2006), 8. OSMUK, N., D. SVOLJŠAK, B. ŽBONA TRKMAN 1994, Ajdovščina – Castra. – Kulturni in naravni spomeniki Slovenije, Zbirka vodnikov 183, Ljubljana. PANCIERA, S. 1957, Vita economica di Aquileia in eta roma­na. – Aquileia. PANCIERA, S. 1979, Il territorio di Aquileia e l‘epigrafia. – Antichita Altoadriatiche 15, 383–411. PAVAN, M. 1979, Presenze di militari nel territorio di Aquileia. – Antichita Altoadriatiche 15, 461.513. PETOLESCU, C. C. 1996, Inscriptiones Daciae Romanae. Inscriptiones extra fines Daciae repertae Graece et Latinae (saec. I.II.III.), vol. I: Italia et Pars Occidentis. – Bucarest. PETRU, P. 1960-1961a, Ajdovščina. – Varstvo spomenikov 8, 199–200. PETRU, P. 1960-1961b, Ajdovščina. – Varstvo spomenikov 8, 236–237. PETRU, P. 1965, Obnovljena antika v Ajdovščini. – Varstvo spomenikov 10, 131-144. PETRU, P. 1969, Najnovija istraživanja Julijskih Alpa. – Osječki zbornik 12, 5–24. PETRU, P. 1971, Ajdovščina. – V / In: Šašel, Petru (ur. / eds.) Claustra Alpium Iuliarum 1. Fontes, Katalogi in Momo­grafije 5, 97-99. PETRU, P. 1972, Novejše arheološke raziskave Claustra Al-pium Iuliarum in kasnoantičnih utrdb v Sloveniji (Recenti ricerche archeologice delle Claustra Alpium Iuliarum e delle fortificazioni tardo antiche in Slovenia). – Arheološki vestnik 23, 343-366. PETRU, P. 1974, Ajdovščina: neue Ausgrabungsresultate. – V / In: Birley, E., Dobson, B., Jarret, M. (ur. / ed.), Roman Frontier Studies 1969. Eighth International Congress of Limesforschung, 178–184, Cardiff. PETRU, P. 1975a, Kasnoantične zapore. – V / In: Arheološka najdišča Slovenije,105, Ljubljana. PETRU, P. 1975b, Ajdovščina. – V / In: Arheološka najdišča Slovenije, 120–121, Ljubljana. PETRU, P. 1976, Ricerche recenti sulle fortificazioni nelle Alpi orientali. – Antichita Altoadriatiche 9, 229–236. PETRU, P., D. SVOLJŠAK 1965a, Ajdovščina. – Varstvo spo­menikov 10, 197. PETRU, P., D. SVOLJŠAK 1965b, Ajdovščina - Arheološki pregled 7, 125. PETRU, P., D. SVOLJŠAK 1980, Ajdovščina. – Rešena arhe­ološka dediščina Slovenije 1945-1980 (katalog razstave), 45, Ljubljana. PFLAUM, V. 2000, Claustra Alpium Iuliarum in barbari. Najdbe poznorimske vojaške opreme in orožja ter sočasne zgodnje barbarske najdbe na ozemlju današnje Slovenije. – Magistrsko delo / MA thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). PFLAUM, V. 2004, Poznorimski obrambni in vojaški sledovi 5. stoletja na ozemlju sedanje Slovenije. – Doktorska disertacija / Ph. D. thesis, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). PICK, K., W. SCHMID 1916, Beitrage zur Altertumskunde des Wippacher Tales [1. Kastell und Poststation am Frigidus. 2. Die Schlacht am Frigidus]. – Laibacher Zeitung, 25. in 27. Sept. 1916. PICK, K., W. SCHMID 1922-1924, Frühgeschichtliche Befesti­gungsanlagen im Bereiche der Isonzofront. – Jahreshefte des Österreichischen Archäologischen Institutes in Wien 21-22, Beiblatt, 277-308. PLESNIČAR, P. 1998, Ajdovščina: pogled v njeno preteklost: ko­mentirana objava zgodovine Ajdovščine in Šturij (pripravila I. Uršič). – Nova Gorica. PREMERSTEIN, A., S. RUTAR 1899, Romische Strassen und Befestigungen in Krain. –Wien. PRÖTTEL, H. M. 1996, Mediterrane Feinkeramikimporte des 2. bis 7. Jahrhunderts n. Chr. im oberen Adriaraum und in Slowenien. – Kölner Studien zur Archäologie der römischen Provinzen 2. PUSCHI, A. 1901, Limes italicus orientalis o i valli romani delle Giulie. Relazione preliminare. – Atti e memorie della Societa istriana di archeologia e storia patria 17, 376-401. PUSCHI, A. 1902, I valli romani delle Alpi Giulie. - Archeo­grafo triestino n.s. 24 (supplemento), 119-150. PUSCHI, A. 1903, La strada romana da Aquileia ad Emona. – Archeografo triestino n.s. 1 (29), 109-125. ROZMAN, L., M. UREK, A. KOVAČIČ 2019, Ajdovščina (arheološko najdišče Castra) – nove raziskave. – V / In: P. Stipančič, A. Gaspari (ur. / eds.), Arheologija v letu 2018, dediščina za javnost. Zbornik povzetkov (Strokovno srečanje Slovenskega arheološkega društva, Ljubljana, Narodni mu-zej Slovenije – Metelkova, 3.–4. marec 2019), Ljubljana, 50. SCHMID, W. 1922-1924, Metulum und Fluvius Frigidus. - Jahreshefte des Österreichischen Archaeologischen Institutes in Wien 21-22, 495-508. SCHMID, W. 1923-1924, Römische Forschungen in Öster­reich. – Bericht der Römisch-Germanischen Kommission XV, 183-189. STICOTTI, P. 1906, Le lapidi del museo di Gorizia. - Archeo­grafo Triestino n.s. 3 (31), 169–188. STICOTTI, P. 1908, Scoperte d‘antichita a Trieste e nel suo agro. – Archeografo Triestino n.s. 4 (32), 283–285. STUCCHI, A. 1946, Aidussina romana. – Ce fastu? XXI, 29–37. STUCCHI, A. 1948, Il tracciato della strada romana da Aqui­leia a Lubiana nella valle di Vipacco. – Ce fastu? 24, 19–23. SVOLJŠAK, D. 1965, Ajdovščina. – Varstvo spomenikov 10, 193–194. SVOLJŠAK, D. 1967, Gradišče nad Ajdovščino. – Varstvo spomenikov 12, 82–83. SVOLJŠAK, D. 1968, Ajdovščina, blagovnica “Nanos” – rimsko naselje i tvrdženje, seoba naroda. – Arheološki pregled 10, 147-150. SVOLJŠAK, D. 1968–1969a, Ajdovščina. – Varstvo spomenikov 13–14, 155-157. SVOLJŠAK, D. 1968–1969b, Ajdovščina. – Varstvo spomenikov 13–14, 185–186. SVOLJŠAK, D. 1970–1971, Posočje v zgodnjem srednjem veku. – Arheološki vestnik 21–22, 153-162. SVOLJŠAK, D. 1988–1989, Posočje v bronasti dobi. – Arheo­loški vestnik 39–40, 367-386. SVOLJŠAK, D. 2000, Dalla via dell‘ambra alla strada Hun-garorum. La rete stradale in Slovenia dalla preistoria fino al primo Medioevo. – V / In: S. Blason Scarel (ur. / ed.), Cammina, cammina: dalla via dell‘ambra alla via della fede, 102–104, Udine. SVOLJŠAK, D., T. KNIFIC 1976, Vipavska dolina, zgodnje­srednjeveška najdišča (Early-medieval sites in the Vipavska dolina (Vipava Valley)). – Situla 17. SVOLJŠAK, D., B. ŽBONA TRKMAN 1986, Novi napisi v Posočju. – Arheološki vestnik 37, 385–398. SVOLJŠAK et al. 2013 = D. SVOLJŠAK, B. ŽBONA TRKMAN, N. OSMUK, B. BREZIGAR 2013, Fluvio Frigido, Castra, Flovius, Ajdovščina. – Nova Gorica. ŠAŠEL, J. 1970, Castra. – V / In: Paulys Realencyclopädie der Classischen Altertumswissenschaft Suppl. XII, 138–139. ŠAŠEL, J. 1971, Antični viri. – V / In: Šašel, Petru (ur. / eds.) Claustra Alpium Iuliarum 1. Fontes, Katalogi in Momo­grafije 5, 17–45. ŠAŠEL, J. 1973, Via Gemina. – Arheološki vestnik 24, 901–902 (= Opera selecta 1992, 592–593). ŠAŠEL, J. 1974, Okra. – Kronika 22, 9–17. ŠAŠEL, J. 1975, Rimske ceste v Sloveniji. – V/ In: Arheološka najdišča Slovenije, 74–87, Ljubljana. ŠAŠEL, J. 1975–1976, Iuliae Alpes. – V / In: Atti. Centro Studi e Documentazione sull'Italia Romana) VII, 601–618. (= Opera selecta 1992, 432–449). ŠAŠEL, J. 1980. Pastorizia e transumanza. – Rivista storica dell'Antichita 10, 179–185 (= Opera selecta 1992, 522–528). ŠAŠEL, J. 1984, Sistemi di difesa della ‚Porta Illirico-italica‘ nel tardo antico. – V/ In: Il crinale d'Europa. L'area illirico danubiana nei suoi rapporti con il mondo classico, 113–123, Roma (= Opera selecta 1992, 795-805). ŠAŠEL, J., P. PETRU (ur. / eds.) 1971, Claustra Alpium Iuliarum I. Fontes. – Katalogi in monografije 5. ŠAŠEL KOS, M. 1990, Nauportus: antični literarni in epigrafski viri / Nauportus: Literary and Epigraphical Sources in Nau­portus (Vrhnika). – V / In: J. Horvat, Nauportus (Vrhnika), Dela 1. razr. SAZU 33, 17-33, 143-159. ŠAŠEL KOS, M. 1997a, The Roman Inscriptions in the National Museum of Slovenia / Lapidarij Narodnega muzeja Slove­nije. – Situla 35. ŠAŠEL KOS, M. 1997b, Pošta v antiki. - V / In: Pošta na slo­venskih tleh, 18-43, Ljubljana. ŠAŠEL KOS, M. 1999a, Markomanske vojne. – V / In: B. Aubelj (ed. / ur.), Zakladi tisočletij, Zgodovina Slovenije od neandertalcev do Slovanov, 196–198, Ljubljana. ŠAŠEL KOS, M. 1999b, Poštna služba v antiki – hitra in počasna. – V / In: B. Aubelj (ed. / ur.), Zakladi tisočletij, Zgodovina Slovenije od neandertalcev do Slovanov, 260–262, Ljubljana. ŠAŠEL KOS, M. 2002a, The boundary stone between Aquileia and Emona (Mejnik med Akvilejo in Emono). - Arheološki vestnik 53, 373-382. ŠAŠEL KOS, M. 2002b, Il confine nord-orientale dell’Italia romana. Riesame del problema alla luce di un nuovo do-cumento epigrafico. – Aquileia Nostra 73, 245-260. ŠAŠEL KOS, M. 2004, Lapidarij narodnega muzeja Slovenije. Rimski spomeniki. Vodnik. – Ljubljana. ŠAŠEL KOS, M. 2016, Cenotaphs and unusual war-time deaths in the southeastern Alps and Pannonia. – V / In: J. Horvat (ur. / ed.), The Roman army between the Alps and the Adriatic. Opera Instituti Archaeologici Sloveniae 31 / Studia Alpium et Adriae I, 213–224. ŠTEKAR S. 1960-1961, Ajdovščina. – Varstvo spomenikov 8, 200–201. TIUSSI, C. 2010, Un ritrovamento di miliari nel greto del fiume Torre a Villesse (Gorizia) e la via Aquileia - Iulia Emona. – Aquileia Nostra 81, 277-360. TRATNIK, V. 2007, Ajdovščina. – Varstvo spomenikov. Poročila 44/2007 (2008), 15–17. TRATNIK, V., T. ŽERJAL 2017, Ajdovščina (Castra) – pose-litev izven obzidja (Ajdovščina (Castra) – the extra muros settlement). – Arheološki vestnik 68, 245-294. UREK, M., A. KOVAČIČ 2020, Fluvio Frigido, Castra – Ajdo­vščina. – V / In: J. Horvat, I. Lazar, A. Gaspari (ur. / eds.), Manjša rimska naselja na slovenskem prostoru / Minor Ro­man settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 47–59. VALVASOR, J. W. 1689, Die Ehre des Herzogthums Crain. – Laibach. VEDALDI IASBEZ, V. 1994, La Venetia Orientale e l'Histria. Le fonti letterarie greche e latine fino alla caduta dell'Im­pero Romano d 'Occidente. – Ricerche e studi sulla Gallia Cisalpina 5. VEDALDI IASBEZ, V. 2000, Aquileia e le fonti letterarie dell‘eta imperiale e tardoantica. – Antichita Altoadriatiche 47, 297-312. VENTURA, P. 2008, Recenti acquisizioni e riletture dal ter­ritorio nord-orientale di Aquileia. – V / In: F. Slavazzi, P. Maggi, La scultura romana dell‘Italia Settentrionale. Qua-rant‘anni dopo la mostra di Bologna. Flos Italiae, Documenti di archaeologia della Cisalpina Romana 8, 73–87, Firenze. VIDRIH PERKO, V. 1994, Poznoantične amfore v Sloveniji. - Disertacija / PhD thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). VIDRIH PERKO, V. 1997, Some Late Roman Ceramic finds from the Slovenia Karst Region. – Acta Rei Cretariae Ro­manae Fautorum 35, 249-258. VIDRIH PERKO, V. 2000, Poznoantične amfore v Sloveniji. – Annales. Ser. hist. sociol. 10/2 (22), 421–455. VIDRIH PERKO, V., B. ŽBONA TRKMAN 2003–2004, Trgo­vina in gospodarstvo v Vipavski dolini in Goriških brdih v rimski dobi. Interpretacija na podlagi najdišč Loke, Neblo, Bilje in Ajdovščina. – Goriški letnik 30–31 (2005), 17–72. VIDRIH PERKO, V., B. ŽBONA TRKMAN 2004, Aspetti ambientali e risorse naturali nell'indagine archeologica: il caso della valle del Vipacco e i suoi rapporti con l'economia aquileiese. – Antichita altoadriatiche 58, 23–42. VIDRIH PERKO, V., B. ŽBONA TRKMAN 2005, Ceramic finds from Ajdovščina – Fluvio Frigido, an Early Roman road station and Late Roman fortress Castra. – Acta Rei Cretariae Romanae Fautorum 39, 277– 286. VUGA, D. 1984, Vipavska dolina 1: arheološki vodnik. – Ma-ribor. VUGA, D. 1993a, Ajdovščina. – V / In: Po poteh rimskih vojakov v Sloveniji / In the Footsteps of Roman Soldiers in Slovenia, 22–23, Ljubljana. VUGA, D. 1993b, Bojišče na reki Frigidus. – V / In: Po poteh rimskih vojakov v Sloveniji / In the Footsteps of Roman Soldiers in Slovenia, 24–25, Ljubljana. WITSCHEL, C. 2002, Meilensteine als historische quelle? Das Beispiel Aquileia. – Chiron 32, 325-393. ZACCARIA, C. 1992, Regio X. Venetia et Histria. Tergeste – Ager Tergestinus et Tergesti adtributus. – Supplementa Italica, n. s. 10, 139–283. ZACCARIA, C. 2003, Amministrazione e vita politica ad Aquileia dalle origini al III sec. d.C. – Antichita Altoadria­tiche 54, 293-338. ZACCARIA, C. 2007a, Tra Natisone e Isonzo. Aspetti ammi­nistrativi in eta romana. – V / In: Le Valli del Natisone e dell'Isonzo tra Centro Europa e Adriatico. Atti del Convegno internazionale di studi (S. Pietro al Natisone, 15-16 sett. 2006), Studi e ricerche sulla Galia Cisalpina 20, 129-144. ZACCARIA, C. 2007b, Epigrafia dell‘ arco alpino orientale: novita, riletture, progetti. – V / In: E. Migliario, A. Baroni (ur. / eds.), Epigrafia delle Alpi. Bilanci e prospettive. Atti del Convegno internazionale di Studi (Trento 3-5 novembre 2005), Labirinti 107, 315-350. ZACCARIA, C. 2009a, Romani e non Romani nell’Italia nordorientale: la mediazione epigrafica. – Antichita Alto-adriatiche 68, 71–108. ZACCARIA, C. 2009b, Forme e luoghi della “mediazione” nell’Italia nordorientale romana. - V / In: F. Crevatin (ur. / ed.), I luoghi della Mediazione. Confini, scambi, saperi. Atti della Giornata di Studio (Trieste, 18 dicembre 2007), Fonti e studi per la storia della Venezia Giulia, ser. II/18, 241–261, Trieste. ZACCARIA, C. 2014, Constantino ad Aquileia: tra epigrafia e retorica. – V / In: G. Cuscito (ur. / ed.), Constantino il Grande a 1700 anni dall' Editto di Milano, Antichita Alto-adriatiche 78, 179–192. ŽBONA TRKMAN, B. 1982, Ajdovščina. – Varstvo spome­nikov 24, 165. ŽBONA TRKMAN, B. 1986, Ajdovščina. – Varstvo spome­nikov 28, 260. ŽBONA TRKMAN, B. 1987, Memento. Ajdovščina-Castra--Fluvio Frigido. – Goriški letnik 12–14 (1985–1987), 315–319. ŽBONA TRKMAN, B. 1990a, Ajdovščina. – Varstvo spome­nikov 32, 169–170. ŽBONA TRKMAN, B. 1990b, Zbirka Simona Kovačiča iz Šempetra. – Varstvo spomenikov 32, 198–199. ŽBONA TRKMAN, B. 1993, I bolli laterizi dell‘Isontino: stato delle ricerche. – V / In: C. Zaccaria (ur. / ed.), I laterizi di eta romana nell'area nordadriatica, Cataloghi e monografie archeologiche di Civici musei di Udine 3, 187–196. Tina Žerjal Arhej d. o. o. Drožanjska cesta 23 SI-8290 Sevnica tina.zerjal@guest.arnes.si Vesna Tratnik Narodni muzej Slovenije Prešernova cesta 20 SI-1000 Ljubljana vesna.tratnik@nms.si Marjeta Šašel Kos Znanstvenoraziskovalni center SAZU Inštitut za arheologijo Novi trg 2 SI-1000 Ljubljana mkos@zrc-sazu.si Andreja Maver Strokovno prevajanje, Andreja Maver s.p. Prešernova cesta 4 SI-1000 Ljubljana andreja_maver@yahoo.com Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 47–59 FLUVIO FRIGIDO, CASTRA – AJDOVŠČINA Raziskave / Investigations 2017–2019 Maruša UREK, Ana KOVAČIČ Izvleček Predstavljeni so predhodni rezultati arheoloških raziskav v starem mestnem jedru Ajdovščine (Fluvio Frigido, Castra). V letih 2017–2019 so arheološka izkopavanja potekala znotraj in delno zunaj obzidja poznorimske utrdbe Castra. Jugo­vzhodno od stolpa 12 so bili raziskani del obzidja, obrambni jarek in pohodna površina (berma). Znotraj obzidja so bile odkrite poznorimske stavbe, kanalizacija in cesta. V osrednjem delu utrdbe je bil delno raziskan velik stavbni kompleks z razčlenjeno večfazno zasnovo, ki kaže na osrednjo vlogo znotraj utrdbe. Najdbe v ruševinski plasti objekta potrjujejo vojaški značaj naselbine. Ključne besede: Italija (10. regija), Ajdovščina, Fluvio Frigido, Castra, rimska doba, naselje, obzidje, obrambni jarek, cesta, centralna stavba, kanalizacija, militaria Abstract The contribution presents the preliminary results of the archaeological investigations taking place in 2017–2019 in the old town centre of Ajdovščina (Fluvio Frigido, Castra). The excavations were conducted intra and in a limited part also extra muros of the Late Roman fortress called Castra. A section of the fortification wall, defensive ditch and ancient ground surface (berm) was investigated southeast of Tower 12. Excavations intra muros unearthed Late Roman buildings, sewage canals and a road. The extensive and multi-phase building complex occupying the centre of the fortress interior presumably represented the focal point of the fortress. The small finds in the debris layer of this complex confirm a military character of the settlement. Keywords: Italy (Regio X), Ajdovščina, Fluvio Frigido, Castra, Roman period, settlement, fortification wall, defensive ditch, road, central building, sewerage system, militaria https://doi.org/10.3986/9789610502586_02 V sklopu celovite obnove starega mestnega jedra v Ajdovščini od novembra 2017 potekajo arheološke raziskave pod vodstvom Luke Rozmana. V letih 2017 in 2018 so raziskave zajemale območje Prešernove ulice pred južnim obzidjem utrdbe Castra, odsek Prešernove ulice v notranjosti utrdbe in Lavričev trg. Raziskave je opravila ekipa arheologov Magelan skupine d.o.o. Od decembra 2018 potekajo arheološka izkopavanja pod okriljem podjetja Avgusta d.o.o. PRAZGODOVINA Med arheološkimi raziskavami na Prešernovi ulici in na večjem delu Lavričevega trga so bili v aluvialnih plasteh najdeni posamezni manjši odlomki prazgodo­vinske lončenine. Na jugozahodnem delu Lavričevega trga (pod rimsko cesto) je bil odkrit del paleostruge, na robu katere je bila večja koncentracija bronastodobne lončenine “in situ”. Odkritje razkriva, da so prebivalci utrjene prazgodovinske naselbine (kaštelirja) na Gradiš-ču ali Školju (232 m) koristili ravnico ob sotočju Hublja in Lokavščka. Lončenino lahko preliminarno primer-jamo z najdbami z najdišča Zemono, ki je datirana v srednjo bronasto dobo.1 ZGODNJERIMSKO OBDOBJE Na Prešernovi ulici in na območju poznorimske stavbe IV (sl. 1) so bile ugotovljene ožgane hodne po­vršine z jamami ter nasutja kovaške žlindre iz zgodnje­rimskega obdobja. V čas pred izgradnjo poznorimske utrdbe sodi temelj, ki je bil presekan s poznorimsko stavbo II (sl. 1: A; 3). Usmerjen je bil od severozahoda proti jugovzhodu ter grajen iz prodnikov in malte. Del istega poslopja je bil odkrit tudi pod stavbo I v letih 1984, 1985 in 1987.2 Na Lavričevem trgu ni bilo arhitekturnih ostankov zgodnje rimske faze, v nasutjih pod osrednjo poznorimsko stavbo III pa je bilo odkritih več drobnih najdb iz obdobja pred izgradnjo utrdbe Castra. POZNORIMSKO OBDOBJE – UTRDBA MUTATIO CASTRA Tloris obzidja poznorimske utrdbe, ki leži ob soto-čju Hublja in Lokavščka, je nepravilne oblike. Obzidje ima 14 stolpov. Analize novcev so pokazale, da je bilo obzidje zgrajeno v sedemdesetih ali osemdesetih letih 3. stoletja.3 Do obravnavanih raziskav sta bila v notra­njosti raziskana dva večja dela naselbine, t. i. atrijska 1 Bratina 2014, 566. 2 Osmuk 1986, 158; Svoljšak et al. 2013, 47–49, 68; po­drobneje Žerjal, Tratnik 2020, v tej knjigi. 3 Kos 2012, 285; 2014, 35. Extensive archaeological investigations, led by Luka Rozman, have been taking place in Ajdovščina since November 2017 as part of a complete renovation of the old town centre. In 2017 and 2018, archaeologists (Magelan Skupina) examined the area between the street of Prešernova ulica just outside the southern fortification wall, a section of the same street in the fortress interior and the square of Lavričev trg. From December 2018 onwards, excavations are conducted by the Avgusta company. PREHISTORY The investigations at Prešernova ulica and Lavričev trg reached down to the alluvial layers that yielded in­dividual small sherds of prehistoric pottery. A section of the palaeochannel was found in the southwest part of Lavričev trg (under the Roman road), with an in situ concentration of Bronze Age pottery at the edge of the palaeochannel. This find indicates that the inhabitants ofthe prehistoric hillfort located on either Gradišče or Školj (232 m asl) were also active in the flatland below, at the confluence of the Hubelj and Lokavšček streams. Accord­ing to preliminary observations, the sherds are compa­rable to the Middle Bronze Age pottery from Zemono.1 EARLY ROMAN PERIOD Burnt ground surfaces with pits, as well as deposits of smithing slag from the Early Roman period were un­earthed at Prešernova ulica and in the area of Late Roman Building IV. Also predating the construction of the Late Roman fortress is the foundations subsequently cut by Late Roman Building II (Figs. 1: A; 3). The foundations were constructed of mortar-bound cobbles and ran in a northwest-southeasterly direction. Forming part of the same building are the remains unearthed in 1984, 1985 and 1987 under Building I.2 Lavričev trg yielded no Early Roman building remains, while several small finds pre­dating the construction of the Castra fortress were found in the deposits under central Late Roman Building III. LATE ROMAN PERIOD – MUTATIO CASTRA The Late Roman fortress located at the confluence of the Hubelj and Lokavšček streams has an irregular layout and a fortification wall with fourteen towers. Numismatic analyses have shown that the fortification wall was constructed in the 270s or 280s.3 Prior to 2017, 1 Bratina 2014, 566. 2 Osmuk 1986, 158; Svoljšak et al. 2013, 47–49, 68; for more details, see Žerjal, Tratnik 2020, in this book. 3 Kos 2012, 285; 2014, 35. kanal / sewer 1 8 10 11 Sl. 1: Ajdovščina. Stavbe na območju utrdbe. M. = 1:100. Fig. 1: Ajdovščina. Buildings in the fortress interior. Scale 1:100. hiša (vzhodni del stavbe III)4 ter stavba I5 in stavba II s termami ob južnem obzidju.6 Pred južnim obzidjem (v južnem delu Prešernove ulice) so bile rimskodobne ostaline v večji meri uničene 4 Svoljšak 1968-1969; Svoljšak et al. 2013, 47–49, 68. 5 Osmuk 1986, 258–260; Svoljšak et al. 2013, 43. Več o različnih interpretacijah z navedbami literature: Žerjal, Trat­nik 2020. 6 Osmuk 1986, 258–260; Svoljšak et al. 2013, 42-46, 63-64; Žerjal, Tratnik 2020. two large parts of the fortress interior were investigated: the so-called Atrium House (east part of Building III),4 as well as Buildings I5 and II, the latter with baths in the vicinity of the southern fortification wall.6 4 Svoljšak 1968-1969; Svoljšak et al. 2013, 47–49, 68. 5 Osmuk 1986, 258–260; Svoljšak et al. 2013, 43. For more on interpretations and references Žerjal, Tratnik 2020. 6 Osmuk 1986, 258–260; Svoljšak et al. 2013, 42–46, 63–64; Žerjal, Tratnik 2020. 49 Faza / Phase 1 zid, faza 1 / wall, Phase 1 zid, faza 2a-c / wall, Phases 2a-c prizidek / annex prizidek / annex Sl. 2: Ajdovščina. Stavba III, razvoj po fazah. M. = 1:100. Fig. 2: Ajdovščina. Building III, phases of development. Scale 1:100. Sl. 3: Ajdovščina. Temelj zgodnjerimske stavbe. Pogled proti severozahodu. Fig. 3: Ajdovščina. Foundations of an Early Roman building (view towards the northwest). (Foto / Photo: L. Rozman; arhiv / archives: Magelan skupina d.o.o.) Sl. 5: Ajdovščina. Pogled na del obzidja ob stolpu 12, ki je bil po­škodovan z novodobno kanalizacijo. Pogled proti severovzhodu. Fig. 5: Ajdovščina. Section of the fortification wall at Tower 12 damaged by the modern sewage system (view to the northeast). (Foto / Photo: L. Rozman; arhiv / archives: Magelan skupina d.o.o.) Sl. 4: Ajdovščina. Pogled na obrambni jarek in del pohodne površine (berme) jugovzhodno od stolpa 12. Pogled proti jugu. Fig. 4: Ajdovščina. Defensive ditch and ground surface (berm) southeast of Tower 12 (view to the south). (Foto / Photo: L. Rozman; arhiv / archives: Magelan skupina d.o.o.) z recentnimi gradbenimi posegi. Deloma se je ohranil le predel tik ob obzidju. Na tem mestu je bil najden 2 m globok in 6 m širok obrambni jarek, ki je potekal vzpo­redno z obzidjem in je bil od njega oddaljen 8 m (sl. 1; 4).7 Od obzidja se je ohranil 3,4 m širok temelj iz večjih lomljencev (sl. 1; 5), ki je bil vezan s pravokotnim teme­ljem stolpa št. 12 (sl. 1; 6). Obzidje in obrambni jarek je povezovala tlakovana pohodna površina (berma; sl. 1; 4). STAVBE Znotraj obzidja je bil odkrit vzhodni del stavbe II s termami (44 × 19 m), ki so bile deloma raziskane v osemdesetih letih 20. stoletja.8 Celotna stavba meri 44 × 29 m. Pri zadnjih izkopavanjih so bili odkriti temelji najmanj devetih prostorov. V severnem predelu poslopja je ležal večji prostor (8) z osrednjim nizom stebrov (sl. 1; 7). Temelji (širine med 0,60 in 0,90 m) so bili grajeni iz pretežno lomljencev flišnega peščenjaka in apnenčevih prodnikov lokalnega izvora.9 Stavba je imela vsaj dve 7 Obrambni jarek ob zahodnem obzidju je omenjal že Schmid. Za navedbo literature glej Žerjal, Tratnik 2020. 8 Osmuk 1986, 258–260. V terenski dokumentaciji razis­kav 2017–2019 je bila stavba imenovana objekt 1. 9 Verbič 2018. Sl. 6: Ajdovščina. Pravokotni temelj stolpa 12. Fig. 6: Ajdovščina. Rectangular foundations of Tower 12. (Foto / Photo: L. Rozman; arhiv / archives: Magelan skupina d.o.o.) The Roman-period remains just outside the south­ern fortification wall (south part of Prešernova ulica) were largely destroyed during modern-period construction work. Only the part immediately next to the fortification wall survived in part, revealing a 2 m deep and 6 m wide defensive ditch running parallel with the fortification wall at a distance of 8 m (Figs. 1; 4).7 Foundations constructed of large pieces of rubble and measuring 3.4m in width (Figs. 1; 5) survived of the fortification wall, connected with the rectangular foundations of round Tower 12 (Figs. 1; 6). The area between the fortification wall and the defensive ditch revealed a paved ground surface (berm; Figs. 1; 4). BUILDINGS The remains of five buildings have thus far been investigated in the fortress interior. Of Building II that measures 44 × 29 m in total, the east part with baths (44 × 19 m) was already investigated in the 1980s,8 while recent excavations unearthed the foundations of at least nine rooms. Large Room 8 with a central line of columns was investigated in the northeast corner (Figs. 1; 7). Its 0.6–0.9 m wide foundations were predominantly built of flysch sandstone rubble and locally available limestone cobbles.9 The building had at least two construction phases. Lime mortar floors of poorer quality made up the ground surfaces of the early phase, while more solid mortar floors were laid in the later phase. In spite of be­ing poorly preserved, the central part of the building can be interpreted as living quarters. Surviving in the east is part of a waterproof floor that suggests the existence of a pool. Also found in the east is Apsed Room 10–11 (Figs. 1; 8), possibly a small bath complex. The recorded partial renovation of the apsed room confirms the existence of two construction phases. Unfortunately, recent interven­tions destroyed the part between this and the main part of the building under Prešernova ulica to the west and their relationship remains unknown. 7 Schmid mentions a defensive ditch along the western fortification wall; for references, see Žerjal, Tratnik 2020. 8 Osmuk 1986, 258–260. Referred to as Building 1 in the field records of the 2017–2019 excavations. 9 Verbič 2018. Sl. 7: Ajdovščina. Severni del stavbe II z ohranjenimi bazami stebrov oz. plintami. S puščico je označen del mlajše stavbe. Pogled proti jugu. Fig. 7: Ajdovščina. North part of Building II with surviving column plinths. The arrow marks part of a later building (view to the south). (Foto / Photo: L. Rozman; arhiv / archives: Magelan skupina d.o.o.) Two metres north of Building II is Building IV (38 × 14 m; Figs. 1; 9).10 It comprised at least five rooms with hearths. Room 1 yielded a large number of coins (more than 300). This building also shows two construction phases. Occupying the central part of the fortress was a large building of a complex and multi-phase layout, i.e. Build­ing III (89 × 39 m; Figs. 1; 2; 10). Its east part was already investigated in 1968–1969.11 Excavations recorded 23 rooms constructed in several phases. In Phase 1, only the east part was constructed, comprising four large rooms of an L-shaped layout with rare surviving patches of the floor (Fig. 2); it continued beyond the excavation area to the north. Phase 2a involved the south wing of the original building to be partly torn down, renovated and extended further to the west, transforming it into a narrow building with a porticus along the main road through the fortress and a monumental entrance that led onto a large, also por­ticated courtyard. The east wing was turned into a single large space with two rows of columns dividing it into three equally wide aisles (Figs. 1; 2). The walls of Phase 2a stood on solid foundations measuring 1–1.4 m in width and up to 1 m in depth, constructed of flysch sandstone 10 Referred to as Building 2 in the field records of the 2017–2019 excavations. 11 Referred to as Building 3 in the field records of the 2017–2019 excavations. The east part of the building inves­tigated in 1968–1969 was initially named the Atrium House; Svoljšak 1968-1969; Svoljšak et al. 2013, 47–49, 68. Sl. 8: Ajdovščina. Stavba II. Prostora 10 in 11 z apsido. Vogal stavbe je poškodovala mlajša rimskodobna kanalizacija. Pogled proti jugu. Fig. 8: Ajdovščina. Building II. Apsed Room 10–11. Its corner was damaged by the later Roman sewerage (view to the south). (Foto / Photo: M. Kot; arhiv / archives: Magelan skupina d.o.o.) gradbeni fazi. Hodne površine starejše faze so večinoma predstavljali manj kvalitetni apneni estrihi, v mlajši fazi stavbe pa je bil dokumentiran trden estrih. Osrednjemu delu poslopja lahko kljub slabi ohranjenosti pripišemo bivalni značaj. V vzhodnem delu objekta se je ohranil del vodotesnega estriha, ki nakazuje na obstoj bazena. Na manjši kompleks term v vzhodnem krilu stavbe kaže tudi prostor 10–11 z apsido (sl. 1; 8). Da je imel tudi ta del objekta dve fazi, govori delna prezidava prostoraz apsido. Žal je bil stik z osrednjim delom stavbe pod Prešernovo ulico uničen z recentnim posegom. Dva metra severno od stavbe II leži stavba IV (38 × 14 m; sl. 1; 9).10 V njej je bilo raziskanih najmanj pet prostorov z ognjišči. V prostoru 1 je bila odkrita večja količina novcev (čez 300). Tudi v tej stavbi sta bili dokumentirani najmanj dve gradbeni fazi. Na osrednjem območju utrdbe je bil odkrit velik stavbni kompleks z razčlenjeno večfazno zasnovo, stav­ba III (89 × 39 m; sl. 1; 2; 10), katere vzhodni del je bil raziskan že v letih 1968–1969.11 Dokumentiranih je bilo triindvajset prostorov, grajenih v več fazah. V fazi 1 je bila stavba omejena na vzhodni del poznejšega obsega in je imela štiri velike prostore z redkimi ohranjenimi hodnimi površinami (sl. 2). Nadaljevala se je proti se­veru, izven območja izkopavanj. V fazi 2a so delno porušili in predelali zahodni del prvotnega objekta ter zgradili velik stavbni kompleks. Južno krilo s stebriščem je ležalo ob glavni cesti skozi utrdbo. Monumentalni vhod z nadstreškom je skozi južno krilo vodil na veliko notranje dvorišče, tudi ob-dano s stebriščem. Vzhodno krilo je predstavljal velik prostor z dvema vrstama stebrov (sl. 1; 2). Zidovi faze 2a so stali na trdnih temeljih širine 1–1,4 m in globine do 1 m, grajeni so bili iz lomljencev flišnega peščenjaka.12 V fazi 2b so z zidovi zaprli vhod na notranje dvorišče in s pregradnimi zidovi razdelili južno krilo na več manjših prostorov s pečmi (sl. 2). Za omenjeno fazo je značilna gradnja zidov slabše kvalitete, ki so bili večinoma brez temeljev. Sestavljali so jih pretežno z malto vezani večji prodniki iz brežine reke Hublja.13 Do zadnjih raziskav je veljalo mnenje, da je bil način gradnje z rečnimi prodniki omejen na zgodnjerimsko obdobje.14 Danes vemo, da so tak način ponovno uporabili v zadnjih fazah izgradnje utrdbe Castra. Na prostorsko stisko v zadnjih fazah utrdbe nakazujejo tudi prezidave v fazi 2c. Na notranji strani osrednjega dvorišča in na območju vzhodnega krila so na obstoječe zidove prislonili manjše prostore. Nadstreški notranjega dvorišča so tako delno izgubili svojo funkcijo (sl. 2). Suhozidni zidovi iz lomljencev in 10 V terenski dokumentaciji objekt 2. 11 V terenski dokumentaciji objekt 3. Vzhodni del stavbe, raziskan v letih 1968–1969, je bil najprej poimenovan “atrijs­ka hiša”; Svoljšak 1968-1969; Svoljšak et al. 2013, 47–49, 68. 12 Verbič 2018. 13 Verbič 2018. 14 Svoljšak et al. 2013, 41. rubble.12 In Phase 2b, the entrance to the courtyard was walled up and partition walls constructed along the length of the south wing to divide it into several smaller rooms with furnaces (Fig. 2). The walls of this phase were built less well, of predominantly mortar-bound large cobbles from the banks of the Hubelj, and mostly without foun­dations.13 Until the recent investigations, the construction technique using river cobbles was believed to have been limited to the Early Roman period;14 today it is clear that they were reused in the last construction phases of Castra. Spatial constraints in the last phases are also apparent from the adaptations in Phase 2c. Constructed along the interior façade of the courtyard and inside the east wing were smaller rooms that in part reused the walls of ear­lier phases; the courtyard portici thus lost their function (Fig. 2). The new rooms had drystone foundations that presumably supported a wooden superstructure.15 Also attributable to one of the last phases is a solid lime mortar floor excavated in Room 3 (Fig. 10). Building 12 Verbič 2018. 13 Verbič 2018. 14 Svoljšak et al. 2013, 41 15 Until recently, drystone walls were attributed to early medieval buildings (Svoljšak 1968-1969; Svoljšak et al. 2013, 47–49, 68); the recent investigations yielded no early medieval portable finds that would corroborate such an attribution. prodnikov so bili grajeni v eni legi in so najverjetneje podpirali lesene konstrukcije.15 V eno mlajših faz je sodil tudi trdni apneni estrih v prostoru 3 (sl. 10: prostor vzhodno od vhoda). Stavba se je nadaljevala proti zahodu in severu pod obstoječe stanovanjske stavbe. Glede na osrednjo lego znotraj utrdbe in ob glavni cesti, obliko in velikost ter na številne kovinske najdbe vojaškega značaja lahko stavbo interpretiramo kot cen­tralni objekt znotraj vojaške utrdbe. Južno od stavbe III je bil odkrit severozahodni vogal stavbe I (raziskano območje 13 × 3 m).16 Ohranila sta se dva temelja v smeri sever – jug in vzhod – zahod ter ostanek zidu. Po usmeritvi in širini temeljev (1,4– 1,53 m) ju lahko povežemo z že prej delno raziskano mogočno stavbo I (sl. 1).17 Če severozahodni vogal stav-be povežemo z južnim delom, je stavba merila približno 47 × 24 m. Stavba je s temeljem sekala cesto iz faze 1, ki je bila, sicer v slabem stanju, prav tako dokumentirana na tem območju. Po legi nad najstarejšo cesto ter usmeritvi zidov, ki se ujema z ostalimi stavbami, bi verjetno lahko stavbo I umestili v čas poznorimske utrdbe oz. v fazo 2. 15 Do nedavnega je veljalo, da so suhozidni zidovi pri­padali zgodnjesrednjeveškim stavbam (Svoljšak 1968-1969; Svoljšak et al. 2013, 47–49, 68), vendar med obravnavanimi raziskavami ni bilo dokumentiranih najdb, ki bi to potrjevale. 16 V terenski dokumentaciji objekt 4. 17 Osmuk 1986, 258–260; Svoljšak 1965; Svoljšak et al. 2013, 43; Žerjal, Tratnik 2020. III continued both west and north, under the present-day buildings. Its central position within the fortress and along the main thoroughfare, its shape and size, as well as the numerous metal finds of a military character indicate that Building III represented a complex of primary importance within the fortress. South of Building III, excavations unearthed the northwest corner of Building I (the investigated area measured 13 × 3 m).16 The foundations of two adjoining walls were unearthed, running north-south and east-west, as well as part of the wall above them. The orientation and width of the foundations (1.4–1.53 m) suggest they belong to the large building already excavated in its south part during earlier campaigns (also called Building I).17 Con­necting the two parts reveals a roughly 47 × 24 m large building. Its foundations in the northwest corner cut by were constructed on top of the road from Phase 1, which was found in a poor state. The stratigraphic position with regards to the earlier road and the orientation of the walls that corresponds with that of other buildings indicate that Building I may be dated to the time of the Late Roman fortress, i.e. Phase 2. The western edge of the excavation area revealed two sections of the foundations of Building V (9 × 8 m) and 16 Referred to as Building 4 in the field records of the 2017–2019 excavations. 17 Osmuk 1986, 258–260. Svoljšak 1965; Svoljšak et al. 2013, 43; Žerjal, Tratnik 2020. Na zahodnem delu raziskanega območja so bili odkriti temelji stavbe V (sl. 1; raziskano območje 9 × 8 m) in kamnito tlakovanje, ki jih, glede na najdbe in 18 položaj ob cesti, lahko umestimo v čas utrdbe. KANALIZACIJA Dokumentiranih je bilo osem odsekov rimskodob­ne kanalizacije, grajenih v dveh fazah (sl. 1). V fazi 1 je bil kanal vkopan neposredno v geološko osnovo oz. ni imel podlage. Potekal je od severozahodnega dela Lavričevega trga (sl. 1; 11) pod notranjim dvoriščem stavbe III najprej proti jugovzhodu, zavil proti vzhodu, del pa se je odcepil proti jugovzhodu in se je iztekal v drenažni jarek, ki je obkrožal stavbo III v fazi 1. Iz faze 2, ko so imeli kanali podlago iz kamnitih plošč oz. tegul, sta bila ohranjena dva odseka, krajši na severnem delu Lavričevega trga in daljši na zahodnem delu trga, kjer je potekal od severa proti jugu. Kanala iz faze 2 sta bila odkrita tudi ob vzhodni strani termalnega dela stavbe II (sl. 1). CESTA Rimska cesta je potekala v smeri vzhod– zahod (z rahlim odklonom proti jugu) med stavbo III ter stavba-mi I, IV in V. Vsaj trikrat je bila obnovljena s prodnatimi tlakovanji ter z vmesnimi peščenimi, gruščnatimi in prodnatimi utrditvami (sl. 12), kar nakazuje dolgotraj­nejšo uporabo. Potek ceste se v mlajših fazah ni veliko spremenil, spremenila pa se je struktura in širina ceste. Verjetno je potekala od Lokarjevega parka,19 delno je bila raziskana na južni strani Lavričevega trga (sl. 1). Najvzhodneje je bila dokumentirana severno od stolpa 10, kjer je utrdbo domnevno tudi zapustila. Na podlagi poteka ceste in večjega razmaka med stolpi 1 in 2 ter 9 in 10 lahko predvidevamo, da je imela utrdba še dva dodatna, do sedaj neugotovljena stolpa.20 Del ceste se je že v fazi 1 odcepil proti severu (ni vidno na sl. 2), vendar trasa zaradi velikega uničenja v severnem delu Lavričevega trga ni povsem jasna. Ob veliki preureditvi stavbe III v fazi 2 se je spremenil tudi potek ceste proti severu. Takrat je potekala skozi vhod na notranje dvo- Zaradi majhne površine izkopnega polja trenutno ni mogoče rekonstruirati tlorisa stavbe. 19 Med arheološkimi raziskavami vzhodno od obzidja utrdbe v letih 2006–2007 so bile v Lokarjevem parku in na severnem odseku Ceste 5. maja dokumentirane utrjene (ho-dne) površine iz prodnikov in peska ter zbite plasti zloženih manjših kamnov, ki bi lahko nakazovale na potek ceste ne­koliko južneje, kot se je sprva predvidevalo (Tratnik, Žerjal 2017, 250, 253, sl. 4, 5, 10; sonda 1, 16C). 20 16 stolpov omenjajo že nekateri avtorji. Natančneje o samih prispevkih: Svoljšak et al. 2013; Žerjal, Tratnik 2020. Sl. 11: Ajdovščina, severozahodni del Lavričevega trga. Po-krovne kamnite plošče kanalizacijskega kanala, ki je potekal vzporedno s cesto faze 1. Fig. 11: Ajdovščina. Sewer covered with stone slabs in the northwest part of Lavričev trg and running parallel with the Phase 1 road. (Foto / Photo: M. Kot; arhiv / archives: Magelan skupina d.o.o.) a stone floor that can also be attributed to the time of the fortress based on the associated small finds and position along the road.18 SEWAGE SYSTEM Excavations recorded eight sections of the Roman-period sewage system constructed in two phases (Fig. 1). The sewers of Phase 1 were dug into the underlying geology without a specially made floor. They have been recorded running from the northwestern part of Lavričev trg eastwards (Figs. 1; 11) under the courtyard of Building III, with a section forking southwards and running into the drainage canal around Building III of Phase 1. The sewers of Phase 2 had a floor of stone slabs or tegulae. Two sections have been unearthed: a shorter one in the north 18 The unearthed section of the building is too small to allow for its size and plan to be reconstructed. rišče. Odcep proti severu je izgubil funkcijo z zaprtjem vhoda na dvorišče v fazi 2b. NAJDBE Keramično gradivo je časovno opredeljeno v obdobje od sredine 1. do 5. st., s poudarkom v drugi polovici 3. in v 4. st. Večina gradiva kaže na uvoz iz Italije, Jadrana, vzhodnih in afriških provinc. Gradivo, ki bi ga lahko umestili v čas zgodnjega srednjega veka in srednjega veka, ni bilo odkrito.21 Zgodnjerimsko gradivo so večinoma sestavljale keramične in steklene najdbe iz 1. in 2. st., ki so bile do-kumentirane v sekundarni legi in v ruševinskih nasutjih za izravnavo terena pred gradnjo stavbe III. Posamezne kovinske najdbe so imele vojaški značaj. Med najdbami so bili tudi ostanki starejših stavb (mozaične kocke, 21 Do izkopavanj v letu 2019 so bile tudi novoveške najd-be redke. part of Lavričev trg and a longer one in the west part, the latter leading from north to south. Two other sections of Phase 2 sewers were unearthed along the east side of the baths of Building II (Fig. 1). ROAD The main Roman thoroughfare led from west to east (with a slight southward deviation) and was flanked by Building III to the north and Buildings I, IV and V to the south. It was renovated on at least three occasions by adding a gravel road surface and beddings of sand, rubble or gravel, which indicates a long-term use (Fig. 12). While the orientation of the road remained roughly the same throughout, its structure and width changed in later phases. It presumably entered the walled settlement at Lokarjev park,19 it was partially investigated in the south part of Lavričev trg (Fig. 1) and also recorded north of Tower 10, which is its easternmost known section, where it presumably left Castra to lead further east. The loca­tion of the road, as well as the greater distance between Towers 1 and 2 in the west and Tower 9 and 10 in the east in relation to that between other towers suggest that the fortification wall had two more, as yet unidentified gate towers.20 Already in Phase 1, a section of the road forked north, but its location is unclear due to a very poor pres­ervation of Roman remains in the north part of Lavričev trg. The large-scale renovation of Building III in Phase 2 also involved the change in course of the northbound road, which now led through the monumental entrance onto the courtyard. The northbound section ceased to be used after the monumental entrance had been walled up in Phase 2b. SMALL FINDS The pottery finds span from the mid-1st to the 5th century with the bulk dating to the second half of the 3rd and the 4th century. Most are imports from Italy, the Adriatic area, as well as the eastern and African prov­inces. None of the recovered artefacts is attributable to early or later medieval periods.21 Most of the Early Roman finds are sherds of pottery or glassware dating to the 1st and 2nd centuries, recorded 19 The archaeological investigations conducted in 2006– 2007 east of the fortification wall in Lokarjev park and in the north section of the street of Cesta 5. maja recorded compact surfaces of cobbles and sand, as well as compact layers of smaller stones, which may indicate that the road ran slightly further to the south than initially thought (Tratnik, Žerjal 2017, 250, 253, Figs. 4, 5, 10; Trench 1, 16C). 20 Several authors mention sixteen towers of the fortifi­cation wall; for more details, see Svoljšak et al. 2013; Žerjal, Tratnik 2020. 21 Prior to the excavations in 2019, even post-medieval finds were rare. votlaki – tubuli in ostali gradbeni material), ki so bile verjetno porušene pred izgradnjo poznorimske utrdbe. Večina poznoantičnega gradiva je izhajala iz ru­ševinskih plasti, ki so bile dokumentirane po celotnem izkopnem polju, in ne iz zaprtih kontekstov. Med drob­nim gradivom je nekaj takšnih, ki jih lahko povezujemo z zgodnjekrščansko tematiko, npr. bronasti ključek, okrašen s križem in del oljenke z grobom sv. Lazarja ali tabernakljem na disku. Oljenke s tem motivom so bile do sedaj najdene le v severni Afriki in po obliki sodijo v tip Atlante X ali Hayes II, ki so jih začeli izdelovati v drugi četrtini 5. st.22 Po fakturi sodeč, oljenka ne sodi v afriško produkcijo. V ruševinskih plasteh (predvsem nad stavbama III in IV) je bilo veliko odlomkov steklenega posodja, predvsem čaš in steklenic, ki jih lahko datiramo od 2. do 5. st. Poleg zelo številnih novcev so bili odkriti številni bronasti in železni predmeti iz 3. in 4. st., ki potrjujejo vojaški značaj naselbine (sl. 13). in secondary position in the levelling layers preceding the construction of Building III. Individual metal finds can be associated with the army. The small finds also include the remains of earlier buildings (mosaic tesserae, ceramic tubuli and so forth) that were presumably torn down before the construction of the Late Roman fortress. Most artefacts from Late Antiquity were found in debris layers across the excavation area and not in closed contexts. They include several items that may be associ­ated with Early Christianity, for example a bronze key bearing a cross and part of an oil lamp with the depiction of the Tomb of St Lazarus or tabernacle on the disc. Oil lamps with this motif have thus far only been found in North Africa and are of either the Atlante X or Hayes II type that began to be produced in the second quarter of the 5th century.22 In its fabric, however, the oil lamp is not of African production. The debris layers (mainly those above Buildings III and IV) yielded numerous sherds of glassware, primarily of beakers and jugs datable from the 2nd to the 5th centuries. Excavations yielded a mass of coins, but also many bronze and iron artefacts from the 3rd and 4th century that confirm a military character of the settlement (Fig. 13). Translation: Andreja Maver 22 Fioriello 2003, 91. 22 Fiorello 2003, 91. BRATINA, P. 2014, Zemono pri Vipavi / Zemono near Vipa­va. – V / In: B. Teržan, M. Črešnar (ur. / eds.), Absolutno datiranje bronaste in železne dobe na Slovenskem / Absolute dating of the Bronze and Iron Ages in Slovenia, Katalogi in monografije 40, 563–568. FIORELLO, C. S. 2003, Le lucerne imperiali e tardoantiche di Egnatia. – Bari. KOS, P. 2012, The construction and abandonment of the Claustra Alpium Iuliarum defence system in light of the numismatic material / Gradnja ali opustitev obrambnega sistema Claustra Alpium Iuliarum v luči numizmatičnega gradiva. – Arheološki vestnik 63, 265-291. KOS, P. 2014, Izgradnja zapornega sistema Claustra Alpium Iuliarum. Antični, arheološki in numizmatični viri / Con­struction of the Claustra Alpium Iuliarum fortifications. Historical, archaeological and numismatic sources. – V / In: J. Kusetič, P. Kos, A. Breznik, M. Stokin (ur / eds.), Claustra Alpium Iuliarum – med raziskovanjem in upravljanjem / Claustra Alpium Iuliarum – Between Research and Mana­gement, 112–132, Ljubljana. OSMUK, N. 1986, Ajdovščina. – Varstvo spomenikov 28, 258-260. SVOLJŠAK, D. 1968–1969, Ajdovščina. – Varstvo spomenikov 13–14 (1970), 155–157.SVOLJŠAK et al. 2013 = D. SVOLJŠAK, B. ŽBONA TRKMAN, N. OSMUK, B. BREZIGAR 2013, Fluvio Frigido, Castra, Flovius, Ajdovščina. – Nova Gorica. TRATNIK, V., T. ŽERJAL 2017, Ajdovščina (Castra) – pose-litev zunaj obzidja (Ajdovščina (Castra) – the extra muros settlement). – Arheološki vestnik 68, 245–294. VERBIČ, T. 2018, Poročilo o geološkem ogledu arheološkega izkopnega polja v Ajdovščini (junij 2018). – (Neobjavljeno poročilo / Unpublished Report; Arhej, d.o.o.). ŽERJAL, T., V. TRATNIK 2020, Fluvio Frigido, Castra – Ajdo­vščina. – V / In: J. Horvat, I. Lazar, A. Gaspari (ur. /eds.), Manjša rimska naselja na slovenskem prostoru / Minor Ro­man settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 9–46. Maruša Urek Maruša Urek, arheološke raziskave s. p.Žirovnica 107 SI-4274 Žirovnica arheoloske.raziskave@gmail.com Ana Kovačič Poslovne storitve, Ana Kovačič, s. p. Mucherjeva 8 SI-1000 Ljubljana radohova@gmail.com Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 61–76 AD PIRUM - HRUŠICA Peter KOS Izvleček Rimski zaselek leži na najvišjem prevalu (867 m) med Vipavsko dolino in Logaško planoto, prek katerega je od 1. st. nova rimska državna cesta povezovala Akvilejo in Emono. Od sredine 3. st. je bil preval utrjen z močnim obzidjem, ki je zaobjelo tudi tedanje stavbe zaselka. Trdnjava, ki jo je prečni zid delil v zgornji (strm in zato neposeljen) in spodnji (poseljen) del, je bila del poznorimskega zapornega sistema Claustra Alpium Iuliarum. V pisnih virih se kraj omenja kot Ad Pirum summas Alpes oziroma (ne povsem gotovo) tudi kot In Alpe Iulia. V raziskanem delu notranjosti trdnjave so bili odkriti razmeroma skromni sledovi poselitve, ki zajema trinajst stavb, dve peči za žganje apna in dve cisterni iz obdobja od 2. do začetka 5. st. O skromnosti poselitve na prevalu vsaj v raziska­nem delu priča dejstvo, da lahko dokažemo le sočasen obstoj treh (2. st.: stavbe 6, 11, 13), sedmih (4. st.: stavbe 1–5, 9, 12) oziroma največ osmih (3. st.: stavbe 1–4, 6, 8, 10, 12) zgradb. Čas gradnje in uničenja posameznih stavb je bilo mogoče opredeliti predvsem na osnovi novčnih najdb (1172 rimskih novcev). Za obstoj poštne ali preprežne postaje na prostoru znotraj kasnejše trdnjave v 1. st. oziroma v prvi polovici 2. st. raziskovanja niso odkrila dokazov. Namembnosti trinajstih stavb na osnovi drobnega arheološkega gradiva ni mogoče ugotoviti. Ključne besede: Italija (10. regija), Hrušica, Ad Pirum, In Alpe Iulia, Claustra Alpium Iuliarum, rimska doba, obcestna postaja, trdnjava, epigrafske najdbe, 2.–5. st. Abstract The article discusses a Roman hamlet located on a saddle (867 m asl) of the Hrušica plateau that represents the highest point on the route from the Vipava valley in the east to the Logatec plateau in the west. It is across this saddle that the Romans constructed a new main road from Aquileia to Emona in the 1st century AD. In the mid-3rd century, a fort was constructed here that included the buildings of the hamlet. The fort was divided with a transverse interior wall into the upper (steep and hence uninhabited) and lower (inhabited) parts. It formed part of the Late Roman barrier system known as Claustra Alpium Iuliarum. In ancient texts, it is referred to as Ad Pirum summas Alpes, possibly also as In Alpe Iulia. The investigated interior of the Ad Pirum fort revealed relatively scarce habitation traces that comprise only thirteen build­ings, two limekilns and two cisterns dating from the 2nd to the early 5th century. Further evidence that the interior was scarcely inhabited is the fact that only three (2nd century), seven (4th century) or at most eight (3rd century) buildings were found to be in use roughly contemporaneously. The time of the construction and destruction of individual buildings was largely established on the basis of the 1172 recorded Roman coins, while their exact function remains unclear. Investigations also yielded no evidence of the posted existence of a post or relay station from the 1st or first half of the 2nd century. Keywords: Italy (Regio X), Hrušica, Ad Pirum, In Alpe Iulia, Claustra Alpium Iuliarum, Roman period, road station, fort, epigraphic evidence, 2nd–5th centuries https://doi.org/10.3986/9789610502586_03 LEGA Manjši zaselek je bil osnovan na najvišjem delu rimske državne ceste, ki je od 1. stoletja povezovala Akvilejo in Emono. Leži na 867 m visokem prevalu “V Hrušici” na z gozdom porasli kraški planoti Hrušica, ki se razprostira med Trnovskim gozdom, Nanosom in Javorniki ter ločuje Vipavsko dolino od Logaške in Ljubljanske kotline. Od druge polovice 3. st. je bil preval utrjen z obzidjem in trdnjava je bila del zapor­nega sistema Claustra Alpium Iuliarum.1 Prečni zid deli trdnjavo v zgornji del, ki zaradi strmega terena ni bil poseljen, in spodnji del južno in severno od ceste skozi trdnjavo (sl. 1).2 Proti severu in jugu se navezu­je na zaporna zidova, ki ju občasno prekinjajo stolpi oziroma vratni stolp (na jugu). Zaporni zid je proti severozahodu dolg 620 m (z dvema dokumentiranima stolpoma), proti jugovzhodu pa 1011 m (s petimi stolpi in z vratnim stolpom).3 ZGODOVINA RAZISKAV Trdnjavo na Hrušici z njenim dobro ohranjenim obzidjem so že v prvi polovici 19. stoletja povezovali s poznorimskim zapornim sistemom Claustra Alpium Iuliarum, vendar so bila prva manjša arheološka razis­kovanja trdnjave izvedena šele leta 1916. Sistematično so večji del notranjosti trdnjave na Hrušici ter posamezne dele zapornega sistema v njeni bližini raziskali šele med letoma 1971 in 1979.4 PISNI VIRI Ad Pirum summas Alpes (It. Burd. 560. 4) Itinerarij iz tretjega desetletja 4. st. omenja zgolj ime kraja, ki je ležal med mutatio Castra (Ajdovščina) in mansio Longaticum (Logatec), brez omembe njego­vega statusa. In Alpe Iulia (Tab. Peut. III 5) Tabula Peutingeriana, ki je najverjetneje temeljila na originalnem zemljevidu iz časa okoli leta 300,5 med krajema Fluvius Frigidus (Ajdovščina) in Longaticum (Logatec) omenja kraj In Alpe Iulia. Prav zaradi umes­titve med Ajdovščino in Logatcem povezujejo to ime s Hrušico. Predvsem Vedaldi Iasbez je argumentirala, da 1 O zapornem sistemu Kos 2013; Kusetič et al. 2014; Kos 2014; Kusetič 2015; Kos 2015, 13–41; Ciglenečki 2016; Mila­vec 2017 (vse z navedbami starejše literature). 2 Za izdelavo načrtov se najlepše zahvaljujem sodelavcu Juretu Kusetiču. 3 Kos 2015, 102–103. 4 Podrobno o zgodovini raziskovanj Kos 2015, 8–13. 5 Talbert 2010, 133–136. LOCATION A Roman hamlet is located at the highest point of the main road connecting Aquileia and Emona from the 1st century AD onwards. It lies on a 867 m high saddle today known as V Hrušici, on a forested karst plateau of Hrušica in western Slovenia, between the Trnovski gozd plateau, Mt. Nanos and the Javorniki hills, and separating the Vipava valley from the Logatec and the Ljubljana basins. A fort was constructed here in the sec­ond half of the 3rd century that functioned as part of the Claustra Alpium Iuliarum barrier system.1 A transverse wall divided its interior in the upper part, which was not inhabited due to the steep terrain, and the lower part. The road across the saddle led through the middle of the lower part (Fig. 1).2 The fort is associated with two well-preserved barrier walls, one at the north end and the other at the south end, both incorporating towers including a gate tower. The barrier wall running towards the northwest is 620 m long and comprises two recorded towers, while the one leading to the southeast is 1011 m long and has a gate tower, as well as five other towers.3 HISTORY OF RESEARCH The fort at Hrušica and its barrier walls have been associated with the Late Roman Claustra Alpium Iuliarum from the first half of the 19th century onwards, though investigations only began in 1916. The early investigations were small in scale, while larger areas of the fort’s interior, as well as part of the barrier system in proximity were excavated between 1971 and 1979.4 LITERARY EVIDENCE Ad Pirum summas Alpes (It. Burd. 560. 4). An itinerary from the 330s mentions the name of a place located between mutatio Castra (Ajdovščina) and mansio Longaticum (Logatec) without giving its status. In Alpe Iulia (Tab. Peut. III 5). Tabula Peutingeriana, a road map presumably based on an original from around 300,5 mentions a place In Alpe Iulia between Fluvius Frigidus (Ajdovščina) and Longaticum (Logatec). Because of its location, the name has been associated with the fort at Hrušica. Es­ 1 On the barrier system see Kos 2013; Kusetič 2014; Kos 2014; Kusetič 2015; Kos 2015, 13–41; Ciglenečki 2016; Mi-lavec 2017 (all with earlier references). 2 I would like to thank my colleague, Jure Kusetič, for preparing the illustrations. 3 Kos 2015, 102–103. 4 For a detailed history of research, see Kos 2015, 8–13. 5 Talbert 2010, 133–136. Sl. 1: Hrušica. Trdnjava in neposredna okolica. Izmera: J. Kusetič, R. Klasinc, 2013. Fig. 1: Hrušica. The fort and its immediate surroundings (measurements by J. Kusetič, R. Klasinc, 2013). se imena krajev v obliki “in + ablativ” vedno nanašajo na pomembne gorske prelaze.6 Zaradi neujemanja razdalje med kraji, ki jih navaja Tabula Peutingeriana, ostaja vprašanje istovetenja imena In Alpe Iulia s Hrušico odprto.7 Nekateri avtorji zato menijo, da moramo kraj In Alpe Iulia iskati na območju Kalc.8 Alpis Iulia (iugum Carnium) (Anon. Ravenn. 4. 37) Jaro Šašel je argumentiral, da moramo omembo kraja, ki se pri anonimnem geografu iz Ravene nanaša na 5. st., enačiti s hrušiškim prelazom.9 EPIGRAFSKA SPOMENIKA 1 – Oltar (sl. 2) Votivni oltar iz apnenca je bil odkrit v sekundar­ni legi pri odstranjevanju ognjišča v lovskem dvorcu Lanthierijev v notranjosti trdnjave.10 Težko berljiv napis na njem omenja oficirja vojaške enote beneficiarijev, ki je v prvi polovici 3. st. posvetil oltar Jupitru.11 I(ovi) o(ptimo) m(aximo) c(o)hortali L(ucius) No= [------] [------] leg(ionis) [II?] Adiutr(icis) b(ene)f(iciarius) consul(aris) l(ibens) l(aetus) m(erito) v(otum) s(olvit).12 Mere: š: 46 cm; v: 70 cm; d: 30 cm. Datacija: 3. st. Literatura: AE 1921, 72 (Sticotti); Inscr. It. X 4, 348; Giesler 1991, 40 (dat.: zgodnje 3. st.); Schallmayer 1990, Nr. 893 (dat.: 3. st.); Zaccaria 1992, 236 ad no. 348 (dat. 4. st.); lupa 16135 (dat.: 300–400). EDCS-04600158; EDR117371 (F. Mainardis) (dat.: na osnovi epigrafskih formul, zgodovine in paleografije v čas 301–400; oznaka najdišča kot Ajdovščina). Hrani: Civico Museo di Storia ed Arte, Trieste, inv. št. 13551. 6 Vedaldi Iasbez 1994, 101. 7 Za podrobnosti glej Kos 2015. 8 Bosio 1991, 209; Frelih 2003, 29. 9 Šašel 1970–1971, 37, op. 7 (= Šašel 1992, 732, op. 7). 10 Sticotti 1920, 101. 11 Sticotti 1920, 101; AE 1921, 72; Inscr. It. X 4, 348; Gies­ ler 1981, 40; Schallmayer 1990, Nr. 893; Zaccaria 1992, 236 ad no. 348. 12 Napis je naveden po EDR. pecially Vedaldi Iasbez has argued that place names in the form of ‘in + ablative’ consistently refer to important mountain passes.6 Given the differences in the distances between individual places as given in the Tabula Peutin­geriana, however, the question of whether or not we can equate In Alpe Iulia with Hrušica must remain open.7 These differences led some authors to suggest that we should rather seek In Alpe Iulia in the area of Kalce.8 Alpis Iulia (iugum Carnium) (Anon. Ravenn. 4. 37). Jaroslav Šašel argued that we should equate the place that Anonymous of Ravenna mentions in the 5th century with the Hrušica Pass.9 EPIGRAPHIC EVIDENCE 1 – Altar (Fig. 2). A limestone altar was found when removing the fireplace in the hunting lodge of the Lanthieri family, in the interior of the Roman fort.10 The poorly legible inscription mentions a beneficiarius consularis who dedicated the altar to Jupiter.11 I(ovi) o(ptimo) m(aximo) c(o)hortali L(ucius) No= [------] [------] leg(ionis) [II?] Adiutr(icis) b(ene)f(iciarius) consul(aris) l(ibens) l(aetus) m(erito) v(otum) s(olvit).12 Size: w: 46 cm; h: 70 cm; l: 30 cm. Dating: 3rd century. Bibliography: AE 1921, 72 (Sticotti). Inscr. It. X 4, 348. Giesler 1991, 40 (dating: early 3rd century). Schall-mayer 1990, No. 893 (dating: 3rd century). Zaccaria 1992, 236 ad no. 348 (dating: 4th century). Lupa No. 16135 (dating: 300–400). EDCS-04600158 (Clauss / Slaby). EDR 117371 (F. Mainardis) (dat­ing: 301–400 based on epigraphic formulas, history and palaeography; site marked as Ajdovščina). Kept in: Civico Museo di Storia ed Arte, Trieste, Inv. No. 13551. 6 Vedaldi Iasbez 1994, 101. 7 For details, see Kos 2015. 8 Bosio 1991, 209. Frelih 2003, 29. 9 Šašel 1970–1971, 37, Fn. 7 (= Šašel 1992, 732, Fn. 7). 10 Sticotti 1920, 101. 11 Sticotti 1920, 101; AE 1921, 72; Inscr. It. X 4, 348; Giesler 1981, 40; Schallmayer 1990, No. 893; Zaccaria 1992, 236 ad no. 348. 12 The version of the inscription is taken from EDR. Sl. 2: Hrušica, oltar posvečen Jupitru (lupa 16135). Fig. 2: Hrušica, altar dedicated to Jupiter (lupa 16135). 2 – Temelj za spomenik (sl. 3) Na Hrušici je bila najdena 1,12 m visoka baza iz apnenca, domnevno za cesarski spomenik. Leta 1854 je stala ob cesti znotraj trdnjave, kasneje je bila prestavljena izven trdnjave, južno nad cesto proti Ajdovščini.13 Leta 1920 je bila prepeljana v muzej v Trstu.14 Na bazi je poznoantični napis. Bono rei p(ublicae) nato Mere: š: 69 cm; v: 112 cm; d: 46 cm. Datacija: druga polovica 4. st. Literatura: Inscr. It. X 4, 349; CIL III 4613 (p. 2328,42, p. 2328,188); CIL III 11313; lupa 16134; EDR15204 (A. Ragolič), z vso literaturo. Hrani: Civico Museo di Storia ed Arte, Trieste, inv. št. 13592. 13 Hitzinger 1854, 84 piše: “neben der Strasse stehenden … ”; Müllner 1893, 167; Sticotti 1920, 101. 14 Kos 2015, 12, sl. 16. Sl. 3: Hrušica, baza za cesarski kip z napisom (lupa 16134). Fig. 3: Hrušica, inscribed pedestal for an imperial statue (lupa 16134). 2 – Base for an imperial statue (Fig. 3). The Hrušica area also yielded a 1.12 m high base for an imperial statue. It is made of limestone and bears a Late Antique inscription. In 1854, the base stood beside the road that led through the fort, later it was transferred outside the fort, to the left above the road towards Ajdovščina.13 Finally in 1920, it was moved to the museum in Trieste.14 Bono rei p(ublicae) nato Size: w: 69 cm; h: 112 cm; l: 46 cm. Dating: second half of the 4th century. Bibliography: Inscr. It. X 4, 349; CIL III 4613 (p. 2328,42, p. 2328,188); CIL III 11313; lupa 16134; EDR 156204 (Ragolič), with references. Kept in: Civico Museo di Storia ed Arte, Trieste, Inv. No. 13592. 13 Hitzinger 1854, 84 writes “neben der Strasse stehen-den …”; Müllner 1893, 167; Sticotti 1920, 101. 14 Kos 2015, 12, Fig. 16. NUMIZMATIČNO GRADIVO Na Hrušici je bilo mogoče dokumentirati 1172 rimskih novcev.15 Novčne najdbe iz sond v notranjosti trdnjave južno od ceste je objavil in odlično ovrednotil M. Mackensen,16 vse novčne najdbe s Hrušice so po­drobno obravnavane še na dveh mestih.17 Od 1172 nov­cev je bilo v sondah znotraj trdnjavskega obzidja južno in severno od današnje (in verjetno tudi antične) ceste odkritih 343 (29,3 %). Od teh so samo za 263 (22,4 % vseh dokumentiranih novcev) znani natančni najdiščni podatki (sonda, kvadrant, planum), za 80 novcev (6,8 %) pa vsaj eden od podatkov manjka. Kljub temu je prav s pomočjo novcev mogoče podrobneje kronološko umestiti gradnjo oziroma uničenje posameznih objek­tov (stavb, cistern, peči), ne nazadnje pa tudi obnovo vzhodnega obrambnega obzidja. STAVBNI OSTANKI Spodnji del trdnjave, skozi katerega je potekala cesta Emona–Aquileia, zavzema površino 5380 m2, vendar je ravne in za pozidavo primerne površine le 2500 m2.18 Medtem ko so arheološka izkopavanja južno od ceste zajela skoraj vso za zidavo primerno površino, so je raziskave severno od ceste zajele manj kot 50 %. Izkopavanja so odkrila trinajst stavb oziroma njihovih gradbenih struktur ter dve cisterni in dve peči za žganje apna. Poštna postaja (?) Walter Schmid je leta 1917 v trdnjavi raziskal objekt, za katerega je domneval, da je poštna postaja.19 Objekt je sodeč po Schmidovem načrtu in navedbah v dolžino meril okoli 14,5 m, v širino okoli 4,5 m (zahodni prostor) oziroma 3,5 m (osrednji in vzhodni prostor) (po navedbah Schmida 6,4 m).20 Stavba je imela tri prostore, zidovi so bili debeli 50–90 cm, prekrita je bila s strešniki. V osrednjem prostoru so bili odkriti tubuli ter ostanki hipokavsta. Tlak v prostorih je bil delno opečnat, delno kamnit, stene osrednjega prostora so bile ometane, rdeče in belo obarvane ter okrašene z zelenimi lističi. V stavbi so odkrili poznorimske puščice, sulične osti in bodalo. V prostoru v vzhodnem delu stavbe je bilo na površini 1 × 1 m ognjišče. Schmid je domneval, da so poštni postaji pripadajoči hlevi in gospodarska poslopja stali na mestu današnje kmetije Bajec. Razen 15 FMRSl I 17/1; FMRSl III 15; FMRSl VI 15. 16 Mackensen 1981. 17 Kos 1986, 198–201. 18 Kos 2015, 44, sl. 99. 19 Pick, Schmid 1922, 300. 20 Schmid 1932, sl. 98. NUMISMATIC EVIDENCE As many as 1172 Roman coins have thus far been recorded as found at Hrušica.15 Mackensen painstak­ingly analysed the coin finds from the archaeological trial trenching in the fort’s interior south of the road,16 while the integral numismatic evidence from the site has been discussed in detail in two other publications.17 Of the total of 1172, 343 coins (29.3%) have been re­covered in the trenches in the fort’s interior, both south and north of the modern and Roman road. Only 263 of these (22.4%) come from precisely known contexts (trench, grid square, layer), while at least some of these data is missing for the remaining 80 coins (6.8%). In spite of this, coins have been key in dating the construc­tion and destruction of individual buildings, cisterns and limekilns, but also the renovation of the eastern fortification wall. ARCHITECTURAL EVIDENCE The lower part of the fort with the main road be­tween Aquileia and Emona covers a 5380 m2 large area, of which only roughly 2500 m2 is suitable for construc­tion purposes.18 Almost all of the suitable space south of the road has been archaeologically investigated, of the part north of the road less than half. Excavations have unearthed the remains of thirteen buildings, as well as two cisterns and two limekilns. Post station (?) In 1917, Walter Schmid investigated a building in the fort’s interior that he presumed to have functioned as a post station.19 His plan shows a three-room house that measures around 14.5 m in length, around 4.5 m (west room) and 3.5 m (central and east rooms) in width (6.4 m according to the 1932 publication).20 Its walls are roughly 50–90 cm thick and its roof covered with roof tiles. The central room revealed tubuli and remains of underfloor heating, as well as wall plaster painted red and white with green foliage. The floors in the building were partly brick and partly stone. The east room held a hearth covering a surface of 1 x 1 m. Schmid suggested that the stables and outhouses associated with the post station were located in the area of the present-day Bajec farmstead. No records of these investigations have sur­vived apart from Schmid’s brief descriptions. 15 FMRSl I 17/1; FMRSl III 15; FMRSl VI 15. 16 Mackensen 1981. 17 Kos 1986, 198–201. 18 Kos 2015, 44, Fig. 99. 19 Pick, Schmid 1922, 300. 20 Schmid 1932, Fig. 98. Schmidove omembe postaje dokumentacija arheoloških raziskovanj ni ohranjena. Raziskovalci notranjosti trdnjave med letoma 1975 in 1979 so bili prepričani, da z objektom, ki ga je objavil Schmid, ni mogoče povezati arhitekturnih ostankov v arheološko raziskanem spodnjem delu trdnjave. Glede na omembe drobnih najdb gre po vsej verjetnosti za poznorimski objekt. Omenjeni objekt se po vsej verjetnosti nahaja severno od ceste na zahodnem delu še neraziskanega zemljišča proti kmetiji Bajec. STAVBNI OSTANKI JUŽNO OD CESTE Objekti 1–3 (sl. 4: 1–3) Na jugovzhodnem delu trdnjave so bili odkriti sledovi vsaj treh stavb, in sicer v obliki ostankov ognjišč, sledov talnega estriha, lapornatih plošč, sledov stojk za lesene stebre ter nekaj zoglenelih tramov oziroma desk. Ostanki kažejo, da so bile prekrite z lesenimi skodlami (množica izkopanih železnih žebljev), tlak je bil iz estriha oziroma lapornatih plošč ter z ognjiščem v vsaki od njih.21 Dimenzij in oblik teh stavb zaradi skromnih najdb ni mogoče ugotoviti.22 Datacija: sledovi stavb 1–3 so bili odkriti v sondah VIII–XIII in predvsem na osnovi novčnih najdb iz sond lahko sklepamo na obstoj teh stavb v poznem 3. in predvsem v 4. st.23 Objekt 4 (sl. 4: 4) Zidana stavba pravokotne oblike s centralnim ogrevanjem (na zunanji strani zahodne stene je bil odkrit prefurnij) ter kvadratnim prizidkom na njenem jugozahodnem delu (sonde XV, XVII–XX). Stene so bile zgrajene iz lomljenih kamnov, po vsej verjetnosti so bile ometane. V ruševinski plasti so bili odkriti ostanki tubulov, strešne kritine ter fragmenti okenskega stekla, kar govori za kvalitetno grajeno stavbo s hipokavstom, strešno opečnato kritino ter zastekljenimi okni. Prva faza stavbe 4 je bila delno zgrajena na ruševinah starejše (glej objekt 6). Velikosti stavbe 4 ni mogoče ugotoviti. 24 Datacija: na osnovi novčnih najdb je mogoče gradnjo objekta 4, prefurnija (terminus ante quem daje Galijenov novec iz let 267–268) in kvadratnega prizidka (terminus ante quem je posmrtno kovan antoninijan Klavdija II. iz obdobja 270–275) postaviti v sredino 21 Kos 2015, 61. 22 Giesler 1981, 55, Abb. 1, 81. 23 Kos 2015. 24 Za podrobnosti glej Kos 2015, 63–66. The archaeologists investigating the fort between 1975 and 1979 found that none of the archaeologically examined architectural remains in the lower part of the fort corresponded with the building that Schmid had excavated. The small finds in its interior (Late Roman arrowheads, spearheads and a dagger) indicate that it was a Late Roman building. The building is thus most likely located north of the road, in the western part of the uninvestigated section towards the Bajec farmstead. ARCHITECTURAL REMAINS SOUTH OF THE ROAD Buildings 1–3 (Fig. 4: 1–3) Traces of at least three buildings were found in the south-eastern part of the fort (Trenches VIII–XIII). They comprise the remains of hearths, mortar floors, marls slabs, postholes and several charred beams or boards. These indicate wooden buildings covered with wooden shingles, as suggested by a multitude of iron nails, with floors of mortar or marl slabs and a hearth in each of them.21 The scarce remains did not allow the size and layout of the buildings to be reconstructed.22 Dating: The coin finds from Trenches VIII–XIII point to the late 3rd and mainly the 4th century.23 Building 4 (Fig. 4: 4) It was a masonry building with underfloor heating (a praefurnium was found on the exterior side of the west wall) and a square extension in the southwest (Trenches XV, XVII–XX). The walls were made of stone rubble and very probably coated with plaster. The building’s debris layer contained fragments of tubuli, roof tiles and glass panes, all of which indicate comfortable housing. The first phase of this building was partially constructed on top of the ruins of earlier Building 6. Size of the Build­ing 4 cannot be determined.24 Dating: Coin finds date the construction of the house (terminus ante quem from the coin of Gallienus minted in 267–268) and its extension (terminus ante quem from the antoninianus of Claudius II minted post­humously in 270–275) to the mid-3rd century,25 its end after the mid-4th century.26 21 Kos 2015, 61. 22 Giesler 1981, 55, Fig. 1, 81. 23 Kos 2015. 24 For details, see Kos 2015, 63–66. 25 Kos 2015, 65. 26 Kos 2015, 66. Sl. 4: Hrušica. Razporeditev sond južno od ceste Aquileia–Emona z ostanki objektov 1–5, 16 (19 = trdnjavsko obzidje; 20 = po­kopališki zid in cerkev). Načrt: J. Kusetič. Fig. 4: Hrušica. Location of trial trenches south of the road through the fort with marked remains of the unearthed structures 1–5, 16 (19 = fortification wall; 20 = medieval churchyard boundary wall and church). Illustration by Jure Kusetič. 3. st.25 Razrušenje stavbe določajo novčne najdbe v čas po sredini 4. st.26 Objekt 5 (sl. 4: 5) Objekt 5 so zgradili na ruševini uničenega objekta 4 in uporabili ostanke zidov. Ogrevalni sistem je bil opuš-čen, prehod v prizidek zazidan, pri čemer so uporabili material iz ruševin starejše stavbe. Ruševine stavbe 4 so bile pred gradnjo stavbe 5 zravnane in prekrite z novim apnenim tlakom. Severni del stavbe 4, prefurnija in objekta 5 je bil v 13. stoletju uničen ob gradnji srednjeveškega pokopa­liškega zidu (sl. 4: 20) in cerkve, zato velikosti stavb 4 in 5 ni mogoče ugotoviti.27 Datacija: objekt 5 je bil zgrajen po porušenju objek­ta 4, kmalu po sredini 4. st. Terminus post quem uničenja objekta 5 določajo ožgani novci, kovani v času 383–388, v zadnje desetletje 4. oziroma v začetna desetletja 5. st.28 Objekt 6 (sl. 4: 6) Zahodno od objektov 4 in 5 so bili odkriti skromni gradbeni ostanki starejše stavbe (suha zidova, stojke, lapor­nate plošče). Tlak je bil iz lapornatih plošč, odkrita stojka kaže na leseno konstrukcijo sten. Odkriti so bili le deloma ohranjeni ostanki zahodnega in južnega zidu stavbe. 29 Datacija: na osnovi novčnih najdb iz 1. in 2. st. je mogoče gradnjo objekta 6 postaviti v 2. st. Čas uničenja (terminus post quem) opredeljuje ožgan in izrabljen sestercij v sredino 3. st.30 Objekt 7 (sl. 4: 7) Neposredno južno od ceste skozi trdnjavo so bili odkriti ostanki dveh zidov, ki ju moramo glede na usmeritev pripisati istemu antičnemu objektu 7, ki je bil v največjem delu uničen ob gradnji srednjeveškega pokopališkega zidu in cerkve (sl. 4: 20). Eden od zidov (usmeritev J–S) je na risbi tlorisa cisterne in cerkve (prim. sl. 4: objekta 16 in 20) v primarni publikaciji na-pačno identificiran kot del pokopališkega zidu, ki sicer poteka približno štiri metre zahodneje.31 Datacija: domnevno antična, zaradi odsotnosti drobnih najdb podrobnejša časovna opredelitev obeh zidov ni mogoča. 25 Kos 2015, 65. 26 Kos 2015, 66. 27 Podrobno Kos 2015, 67–69. 28 Kos 2015, 69. 29 Glej Kos 2015, 62. 30 Kos 2015, 62. 31 Ulbert 1981, 191, sl. 1. Building 5 (Fig. 4: 5) After the destruction of Building 4, another house (Building 5) was constructed on top of its ruins and re­used part of its walls. The heating system was abandoned, the passage into the extension closed with the reused material. The ruins of the previous building were levelled and covered over with a lime mortar floor. The north parts of Building 4, of the praefurnium and of subsequent Building 5 were completely removed in the 13th century, when the medieval church and churchyard boundary wall were constructed (Fig. 4: 20). The original size of the two buildings can therefore not be established.27 Dating: It was constructed soon after the destruc­tion of Building 4 that occurred not long after the mid­4th century. Coins minted in 383–388 and providing the terminus post quem show that Building 5 must have been destroyed in the last decade of the 4th or the initial decades of the 5th century.28 Building 6 (Fig. 4: 6) Scarce remains of an earlier house, consisting of two drystone walls, postholes and marl slabs, were un­earthed west of Buildings 4 and 5 (Trenches XVI, XX). The marl slabs were used as flooring, while the postholes suggest a wooden wall construction. Of the walls, only parts of the west and south walls survived.29 Dating: Coins from the 1st and 2nd centuries suggest that the house was built in the 2nd century. A burnt and worn sestertius provides the terminus post quem for its destruction in the mid-3rd century.30 Building 7 (Fig. 4: 7) The remains of two Roman walls were unearthed just south of the road through the fort. Their orientation suggests that we should ascribe them to a building largely destroyed during the construction of the medieval church and churchyard boundary wall (Fig. 4: 20). On the drawing of Cistern 2 and church from 1981, one of the walls (orientation S–N) is erroneously identified as part of the churchyard boundary wall that actually runs four metres further to the west.31 Dating: The absence of small finds makes it impos­sible to more precisely date the walls. 27 For details, see Kos 2015, 67–69. 28 Kos 2015, 69. 29 See Kos 2015, 62. 30 Kos 2015, 62. 31 Ulbert 1981, 191, Fig. 1. Objekt 16 – cisterna 2 (sl. 4: 16) Znotraj obeh zidov objekta 7 je bila odkrita cister­na; premer krožnega oboda je 3 m. Datacija: domnevno antična, zaradi odsotnosti drobnih najdb podrobnejša časovna opredelitev objekta ni mogoča. Domnevamo, da je bila cisterna v uporabi že antiki in kasneje še v času delovanja srednjeveške cerkve in pokopališča. STAVBNI OSTANKI SEVERNO OD CESTE Objekt 8 (sl. 5: 8) Leta 1979 so v sondi XXXII, ob severnem robu ceste, odkrili 60–65 cm debel zid stavbe 8, usmerjen V–Z. 32 Proti vzhodu se je nadaljeval v predhodno ra­ziskano sondo XX, kjer so njegov vzhodni del leta 1976 že odkrili.33 Leta 1979 so torej odkrili nadaljevanje zidu proti zahodu in stavbo, ki sta jo določala oba dela zidu 34 skupaj, poimenovali objekt E. Ob notranji strani zidu stavbe 8 je tlak iz estriha, nad njim pa kulturna plast. Datacija: terminus post quem izgradnje objekta 8 daje novec iz obdobja 230–275, ki je ležal na estrihu v sondi XXXII, kvadrant 3. Objekt 9 (sl. 5: 9) V sondah XXXVI in XXXIX so odkrili slabo ohra­njene temelje stavbe 9 (ob izkopavanjih poimenovane objekt G), ki so bili zgrajeni iz lomljencev, povezanih z malto. V notranjosti je bila tlakovana z lomljenci, zadelanimi z malto.35 Datacija: predvsem na osnovi novčnih najdb je mogoče objekt 9 datirati v 4. st. Objekt 10 (sl. 5: 10) Deloma na mestu porušenega objekta 11 so v drugi polovici 2. st. zgradili objekt 10 (ob izkopavanjih poi-menovan objekt A), pri katerem so za gradnjo uporabili les, kamen in malto. Ohranjeni so deli tlaka, obarvanega ometa ter nekaj lesenih brun. Na osnovi zoglenelih brun je deloma mogoče rekonstruirati potek severne in vzhodne 32 Kos 2015, 83, sl. 218 33 Vzhodni odsek zidu je potekal je ob južnem robu son-de XX, vendar so ga takrat napačno razumeli – povezovali so ga s severno ležečim, takrat že odkritim objektom 11 (takrat poimenovanim objekt B). 34 Kos 2015, 83. S črkami od A–G so stavbe poimenovali v času terenskih raziskovalnih del. 35 Kos 2015, 84, sl. 220. Building 16 – Cistern 2 (Fig. 4: 16) A cistern measuring 3 m across was found within the two walls of Building 7. Dating: The absence of small finds only allows for speculations that it was already in use in Roman period, but also later, in the time of the medieval church and churchyard. ARCHITECTURAL REMAINS NORTH OF THE ROAD Building 8 (Fig. 5: 8) The 1979 investigations in the fort’s interior re­vealed a 60–65 cm thick wall, running E–W just north of the road in Trench XXXII.32 It continued in Trench XX, where it had already been recorded in 1976.33 The wall unearthed in 1979 is thus a westward continuation of the wall and the building comprising both sections of the wall was named Building E.34 A mortar floor was found along the interior face of the wall and overlain by a cultural layer. Dating: A coin minted in 230–275 and recovered on top of the mortar floor in Trench XXXII, Grid Square 3, provides the terminus post quem for the construction of the building. Building 9 (Fig. 5: 9) Trenches XXXVI and XXXIX revealed poorly surviving foundations of Building 9 (at the time of ex­cavations named Building G) made of mortar-bound stone rubble. The building’s interior had a floor of stone rubble covered with mortar.35 Dating: The building is attributable to the 4th cen­tury mainly on the basis of coin finds. Building 10 (Fig. 5: 10) It was constructed in the second half of the 2nd cen­tury and partly covered the ruins of Building 11. It was initially marked as Building A. Wood, stone and mortar were used in its construction. The surviving remains 32 Kos 2015, 83. Letters A–G were used to mark buildings during fieldwork. 33 The eastern section of the wall ran along the south­ern edge of Trench XX, but was misinterpreted at the time of discovery – associated rather with the already unearthed Building 11 (then termed Building B). 34 Kos 2015, 83, Fig. 218. 35 Kos 2015, 84, Fig. 220. 8-19 objekt / building XX - XXXIX, F sonda / trench XXIV 11 10 m 10 12 18 XXII 15 17 XX 14 XXVIII XXI XXXII 13 XXXVI XXXIX 8 9 Sl. 5: Hrušica. Razporeditev sond severno od ceste Aquileia–Emona z vrisanimi ostanki stavb 8–15, 17, 18 (19 = trdnjavsko obzidje). Načrt: J. Kusetič. Fig. 5: Hrušica. Location of trial trenches north of the road through the fort with marked remains of the unearthed structures 8–15, 17, 18 (19 = fortification wall). Illustration by Jure Kusetič. stene. Vzhodna stena je segala do cisterne (objekt 15), saj je v ruševini cisterne tudi ruševina zidu stavbe 10. Stavba 10 je bila verjetno zgrajena po postavitvi obzidja (objekt 19), saj ruševina stavbe 10 sega nad maltno gmoto (objekt 14), sočasno z gradnjo obzidja (glej dalje). Objekt 10 in cisterna sta propadla istočasno.36 Datacija: terminus post quem uničenja objekta 10 v požaru postavlja skupek novcev v čas po letu 270. Objekt 11 – tlak (sl. 5: 11) V sondah XXII in XXIV so bili nad živo skalo odkriti arhitekturni ostanki (objekt 11).37 Izkopali so apneni tlak in ostanke obarvanega ometa.38 Ob izko­pavanjih je bil označen kot objekt B. 36 Kos 2015, 81–82, sl. 209–210. 37 Kos 2015, 81–83, sl. 210, 211. 38 V času izkopavanj so z objektom B napačno povezova­li tudi zid in tlak iz oblic v sondi XX, nad katerima je apneni estrih (naš objekt 8). V primeru take rekonstrukcije bi bil ob-jekt 11 velik 16 × 16 m. comprise patches of the floor, painted wall plaster and several wooden beams. The charred beams show the location of the north and east walls. The east wall reached to Cistern 1 (Building 15), with the debris of the building mixed with the remains of the cistern. The building was probably constructed after the fortification wall, as its debris reached above the mass of mortar prepared for the construction of the fortification (Object 14). Building 10 and Cistern 1 fell to ruins at the same time.36 Dating: A group of coins provides the terminus post quem for the destruction of the building in a fire after 270. Building 11 (Fig. 5: 11) Architectural remains of this building were found above the bedrock in Trenches XXII and XXIV.37 They include a lime mortar floor.38 Also attributed to the building is the pieces of painted wall plaster. 36 Kos 2015, 81–82, Figs. 209–210. 37 Kos 2015, 81–83, Figs. 210, 211. 38 The excavators associated the building with the wall 71 Datacija: na apnenem tlaku v sondi XXII so odkrili sestercija Antonina Pija (155–156) in Dive Favstine (141–161), ki določata terminus ante quem izgradnje objekta 11. Objekt 12 (sl. 5: 12) Stavba je bila zgrajena na porušenem objektu 10, od nje pa so ohranjeni ostanki kamnitega temelja zidu, tlak iz sphane ilovice ter ognjišče (sonde XXII, XXIV). 39 Ob izkopavanjih je bila poimenovana kot objekt D. Datacija: terminus post quem gradnje objekta 12 je porušitev objekta 10 (po letu 270). Objekt 12 je živel vsaj do druge polovice 4. st. Objekt 13 – tlak (sl. 5: 13) Gre za tlak iz oblic, ki sega od sonde XXXII do vzhodnega trdnjavskega obzidja (sl. 5: 19) v sondi XX­ VIII. Tlaka ni mogoče povezovati z drugimi stavbnimi ostanki, bil pa naj bi najstarejši gradbeni element na Hrušici. 40 Na tlaku (13) stoji zid – že omenjeni vzhodni odsek zida stavbe 8 (sl. 5: 8) – in je torej mlajši od tlaka. Datacija: na tlaku (objekt 13) v sondi XXVIII je bila odkrita močno profilirana bronasta fibula iz druge polovice 2. st., ki predstavlja terminus ante quem po­stavitve tlaka.41 Objekt 14 – maltna gmota (sl. 5: 14) Gre za debelo plast malte, s katero je prekrit večji del pasu tlaka iz oblic (sl. 5: objekt 13) v sondah XX, XXI in XXVIII. 42 Izkopavalci so to gmoto interpretirali kot ostanek gradnje obzidja (sl. 5: 19). Datacija: sredina 3. st. Terminus post quem nastan­ka maltne gmote (objekt 14) in s tem morda tudi gradnje vzhodnega trdnjavskega obzidja določa bronasta fibula iz druge polovice 2. st., ki je najdena na tlaku (objekt 13) iz oblic in pod maltno gmoto (14). Objekt 15 – cisterna 1 (sl. 5: 15) V delu sond XX in XXII je bila odkrita cisterna pra­vokotnega tlorisa, velikosti 4,1 × 3,4 m, globoka 4 m. 43 Datacija: cisterna je bila zgrajena po porušitvi objekta 11 ter pred gradnjo in porušitvijo objekta 10, 39 Kos 2015, 82–83, sl. 209–211, 216. Kos 2015, 74. Ob izkopavanjih so povezovali tlak iz oblic (naš objekt 13) z objektom 11. 41 Kos 2015, sl. 206 in 232. 42 Kos 2015, sl. 204–206. 43 Kos 2015, 84–85. Dating: Two sestertii of Antoninus Pius (155–156) and Diva Faustina (141–161) were found on top of the lime mortar floor in Trench XXII, which offer the ter­minus ante quem for the construction of the building. Building 12 (Fig. 5: 12) It was constructed on top of the ruins of Building 10 and survives as the remains of stone foundations, a floor of beaten loam and a hearth (Trenches XXII, XXIV). It was initially named Building D.39 Dating: The destruction of Building 10 in the second half of the 3rd century is the terminus post quem for the construction of Building 12, which was in use at least to the second half of the 4th century. Building 13 – cobbled floor (Fig. 5: 13) Marked as Building 13 is a cobbled floor reach­ing from Trench XXXII to the east fortification wall in Trench XXVIII (Fig. 5: 19). It cannot be associated with any of the buildings and is believed to be the earliest construction at Hrušica.40 Standing on top of the floor is the above-mentioned east section of the wall of the later Building 8 (Fig. 5: 8). Dating: A strongly profiled brooch from the second half of the 2nd century was found on top of the cobbled floor in Trench XXVIII that serves as the terminus ante quem for the floor.41 Structure 14 – mass of mortar (Fig. 5: 14) A thick and vast mass of mortar was found on top of the cobbled floor of Building 13 (Fig. 5: 13) in Trenches XX, XXI and XXVIII.42 The mass has been interpreted as the remains of the mortar prepared during the con­struction of the fort’s defensive wall (Fig. 5: 19). Dating: The strongly profiled bronze brooch from the second half of the 2nd century, found on top of the cobbled floor (Building 13) and under the mass of mor­tar (14), serves as the terminus post quem for the mass of mortar and dating the construction of the eastern stretch of the fortification wall to the mid-3rd century (14). and cobbled floor covered by the lime mortar floor in Trench XX; this would suggest a 16 x 16 m large building.39 Kos 2015, 82–83, Figs. 209–211, 216. 40 Kos 2015, 74. During excavation, the floor was as­ cribed to Building 11.41 Kos 2015, Figs. 206 and 232. 42 Kos 2015, Figs. 204–206. katerega ruševine segajo preko cisterne. Njeno gradnjo in uporabo moramo tako postaviti v čas med koncem 2. in drugo polovico 3. st. Objekt 17 – peč I (sl. 5: 17) V vrh zasutja cisterne (15) je bila po njenem uničenju postavljena peč za žganje apna s premerom 2 m, ki je zavzemala skoraj celotno površino cisterne. Kronološko oprijemljivih najdb v peči niso odkrili.44 Datacija: peč je bila postavljena v času po koncu 3. st., po opustitvi cisterne. Objekt 18 – peč II (sl. 5: 18) Večji del stavb 10 in 12 je uničila velika peč za žganje apna (peč II) s premerom 5,4 m.45 Datacija: peč je bila postavljena po porušitvi objek­ta 12. Morda bi jo lahko povezali z obnovo trdnjavskega obzidja v prvih desetletjih druge polovice 4. st.46 SKLEP V raziskanem delu notranjosti trdnjave Ad Pirum so bili odkriti razmeroma skromni sledovi poselitve, ki zajema zgolj trinajst stavb iz obdobja od 2. do začetka 5. st. (sl. 6). Odkriti sta bili tudi dve peči za žganje apna in dve cisterni. Kljub temu, da je že v začetku 1. st. čez Hrušico potekala nova državna cesta, ki je pot med Ak­vilejo in Emono skrajšala za en dan, na raziskanem delu notranjosti trdnjave ni bilo odkritih gradbenih ostankov 1. st., pa tudi izjemno redke drobne arheološke najdbe iz 1. st. so bile najverjetneje v uporabi še v začetku 2. st.47 Z omembe vredno poselitvijo na hrušiškem prevalu lahko zato računamo šele v prvi polovici 2. st., poseli­tev je nato dokumentirana do prvih desetletij 5. st. O skromnosti poselitve na prevalu vsaj v raziskanem delu notranjosti trdnjave priča dejstvo, da lahko dokažemo sočasen obstoj le treh (2. st.: stavbe 6, 11, 13), sedmih (4. st.: stavbe 1–5, 9, 12), oziroma največ osmih (3. st.: stavbe 1–4, 6, 8, 10, 12) zgradb. Za obstoj poštne ali pre­prežne postaje na prostoru znotraj kasnejše trdnjave v 1. oziroma v prvi polovici 2. st. raziskovanja niso odkrila nikakršnih dokazov.48 Drobno gradivo zgodnjega 2. st. kaže na civilenznačaj zaselka lokalnega prebivalstva. Šele sredi 2. st. je mogoče opaziti povečanje števila, raznolikosti in 44 Kos 2015, 86. 45 Kos 2015, 83, sl. 202, 209–211, 216 Terminus post quem obnove vzhodnega trdnjavskega obzidja postavlja novec iz let 352–354; Kos 2015, 50, sl. 119. 47 Giesler 1981, 103. 48 Giesler 1981, 105. Building 15 – Cistern 1 (Fig. 5: 15) A rectangular cistern measuring 4.1 m in length, 3.4 m in width and 4 m in depth was unearthed in Trenches XX and XXII.43 Dating: It was constructed after the destruction of Building 11 (B) and before the construction of Building 10 (A), with the debris of the latter covering the cistern. The construction and use of the cistern should therefore be dated between the end of the 2nd and the second half of the 3rd century. Building 17 – Limekiln I (Fig. 5: 17) The round limekiln measuring 2 m across was set up at the top of the fill of Cistern 1 (15) and taking up almost all of its surface.44 Dating: It contained no chronologically diagnostic finds, but it is believed to have been used after the end of the 3rd century. Building 18 – Limekiln II (Fig. 5: 18) The round kiln measured 5.4 m in diameter and destroyed large parts of Buildings 10 and 12.45 Dating: The kiln was set up after the destruction of Building 12. It may be associated with the renovation of the fortification wall in the third quarter of the 4th century.46 CONCLUSION The investigated interior of the Ad Pirum fort re­vealed relatively scarce traces of habitation that comprise thirteen buildings, two limekilns and two cisterns dating from the 2nd to the early 5th century (Fig. 6). We know that the Romans constructed a new main road at the beginning of the 1st century that led across the Hrušica Pass and shortened the journey from Aquileia to Emona for a whole day. That said, no architectural traces from the 1st century have been found in the interior of the Late Roman fort, while the very rare small finds from the 1st century could also have been used in the 2nd century.47 Available evidence thus offers no support for the exist­ence of a post or relay station from the 1st or first half of the 2nd century in the area later forming the interior of 43 Kos 2015, 84–85. 44 Kos 2015, 86. 45 Kos 2015, 83, Figs. 202, 209–211, 216. 46 A coin minted in 352–354 serves as the terminus post quem for the renovation of the eastern fortification wall; Kos 2015, 50, Fig. 119. 47 Giesler 1981, 103. Stavba, struktura / Building, structure Datacija / Dating Gradnja / Construction Terminus ante quem (taq) Terminus post quem (tpq) Uničenje / Destruction (taq / tpq) 11 2. st. / 2nd c. 141–161 (taq) - 13 2. st. / 2nd c. druga polovica 2. st. (taq) - 14 - maltna gmota / mass of mortar sredina 3. st. / mid-3rd c. druga polovica 2. st. (tpq) - 15 - cisterna 1 / Cistern 1 od 2. st. do prve polovice 3. st. / 2nd c. – first half of the 3rd c. 141–161 (taq) 268–270 (taq) 6 od 2. st. do sredine 3. st. / 2nd c. – mid-3rd c. - 243 tpq 10 prva polovica 3. st. / first half of the 3rd c. 141–161 (tpq) 268–270 (tpq) 12 od druge polovice 3. st. do druge polovice 4. st. / second half of the 3rd c. – second half of the 4th c. 268–270 (tpq) - 4 od sredine 3. do druge polovice 4. st. / mid-3rd c. – second half of the 4th c. 267–268; 270–275 (taq) 335–337; 341–348 (tpq) 8 3. st. / 3rd c. 270–275 (taq) - 1–3 pozno 3. in 4. st. / late 3rd and 4th c. - - 9 4. st. / 4th c. - - 5 od druge polovice 4. st. do prve polovice 5. st. / second half of the 4th – first half of the 5th c. 341–348 (tpq) 383–388 (tpq) 17 - peč za žganje apna I / Limekiln I konec 3. st. / end of the 3rd c. 18 - peč za žganje apna II / Limekiln II druga polovica 4. st. / second half of the 4th c. 7 neopredeljeno / undeterminable 16 - cisterna 2 / Cistern 2 neopredeljeno / undeterminable Sl. 6: Hrušica. Časovna umestitev objektov in drugih struktur. Fig. 6: Hrušica. Chronology of the buildings and other structures. kvalitete odkritih predmetov. Poleg lokalne keramike se pojavijo keramični izdelki iz Afrike. Ob koncu 2. in v 3. st. je zaznati vse več uvoženih predmetov iz provinc severno od Alp.49 Naselbina je bila tako od sredine 2. st. vpeta v rimsko trgovsko mrežo. Gradbeni material (opeke) so dobavljali iz okolice Akvileje.50 Konec 3. st. so prvič dokumentirani vojaški elementi, kar lahko in-terpretiramo s stalno navzočnostjo vojaške posadke od 3. st.51 Najdbe, ki jih lahko povežemo z vojsko, segajo vse do začetka 5. st.52 Na osnovi drobnega gradiva, poveza­nega z žensko nošnjo,53 je sicer mogoče domnevati, da je bilo v 4. stoletju civilnega prebivalstva v glavnem več kot vojaštva. Odkritega je bilo tudi precej raznovrstnega 49 Kos 2015, 97–102. 50 Giesler 1981, 105. 51 Giesler 1981, 65. Kos 2015, 98. 52 Giesler 1981, 68. Pflaum 2002, 265, 275. 53 Giesler 1981, 69–73. the fort.48 It shows that the spot only became occupied on a more permanent basis in the first half of the 2nd century and remained inhabited to the initial decades of the 5th century. It does not seem to have been densely inhabited; we can only prove a roughly simultaneous use of three (2nd century: Buildings 6, 11, 13), seven (4th century: Buildings 1–5, 9, 12) or a maximum of eight buildings (3rd century: Buildings 1–4, 6, 8, 10, 12). The small finds from the 2nd century suggest a ham­let inhabited by local civilian population. An increase in the quantity, variety and quality of small finds is only observable in the mid-2nd century when ceramic imports from Africa appear alongside local pottery. Towards the end of the 2nd and in the 3rd century, there is also an increased use of products imported from provinces north of the Alps;49 the settlement was included into the Roman trading network from the mid-2nd century onwards. The construction material (roof tiles) was de­ 48 Giesler 1981, 105. 49 Kos 2015, 97–102. orodja, vendar v raziskanem delu trdnjave niso zasledili ostankov obrtnih delavnic.54 Namembnosti trinajstih objektov na osnovi drobnih arheoloških najdb ni mogoče ugotavljati. Časovno umestitev posameznih objektov oziroma struktur (po kronološkem zaporedju) kaže preglednica (sl. 6). 54 Kos 2015, 100. Kratice / Abbreviations CIL = Corpus inscriptionum Latinarum. EDCS = Epigraphik-Datenbank Clauss / Slaby (skrbnik / Ser­vice provider: Manfred Clauss) [http://db.edcs. eu/epigr/ epi_de.php]. EDR = Epigraphic Database Roma (skrbnik / Service provider: DigiLab Centro interdipartimentale di ricerca e servizi, Sapienza Universita di Roma) [http://www.edr-edr.it]. FMRSl I = P. Kos, Die Fundmünzen der römischen Zeit in Slowenien I, Berlin 1988. FMRSl III = P. Kos, A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien III, Berlin 1995. FMRSl VI = A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien VI, Wetteren 2010. lupa = UBI ERAT LUPA = F. und O. Harl, http://lupa.at (Bild­datenbank zu antiken Steindenkmälern). BOSIO, L. 1991, Le strade Romane della Venetia e dell‘Histria. – Padova. CIGLENEČKI, S. 2016, Claustra Alpium Iuliarum, tractus Italiae circa Alpes and the defence of Italy in the final part of the Late Roman period. – Arheološki vestnik 67, 409–424. FRELIH, M. 2003, Logatec – Longaticum in rimski obrambni sistem Claustra Alpium Iuliarum. S prispevkom o bitki pri reki Frigidus (Soča) leta 394. – Logatec. GIESLER, U. 1981, Aussagen der Grabungsbefunde. – V / In: T.Ulbert, Ad Pirum (Hrušica). Spätrömische Passbefestigung in den Julischen Alpen, Münchner Beiträge zur Vor- und Frühgeschichte 31, 108–120. livered from the vicinity of Aquileia.50 The first recorded military finds date to the end of the 3rd century and are associated with a permanently stationed garrison.51 Finds associated with the army reach to the beginning of the 5th century.52 The small finds related to female dress suggest that civilian inhabitants outnumbered soldiers in the 4th century.53 The material evidence also includes a substantial quantity and variety of tools, though no remains of workshops have been identified in the investigated part of the fort.54 The small finds from the thirteen houses are not diagnostic as to the function of individual buildings. Chronology of the buildings and other structures is shown on Fig. 6. Translation: Andreja Maver 50 Giesler 1981, 105. 51 Giesler 1981, 65. Kos 2015, 98. 52 Giesler 1981, 68. Pflaum 2002, 265, 275. 53 Giesler 1981, 69–73. 54 Kos 2015, 100. HITZINGER, P. 1854, Die Römerstrasse über die Julischen Alpen und deren Befestigung. – Mittheilungen des histori­schen Vereins für Krain 9, 81–87. KOS, P. 1986, The Monetary Circulation in the Southeastern alpine Region ca. 300 B. C. – A. D. 1000 / Denarni obtok na prostoru jugovzhodnih Alp 300 pr. n. š. – 1000. – Situla 24. KOS, P. 2013, Claustra Alpium Iuliarum – protecting Late Ro­man Italy. – Studia Europaea Gnesnensia 7, 2013, 233–261. KOS, P. 2014, Izgradnja zapornega sistema Claustra Alpium Iuliarum. Antični, arheološki in numizmatični viri / Con­struction of the Claustra Alpium Iuliarum fortifications. Historical, archaeological and numismatic sources. – V / In: J. Kusetič, P. Kos, A. Breznik, M. Stokin (ur / eds.), Claustra Alpium Iuliarum – med raziskovanjem in uprav­ljanjem / Claustra Alpium Iuliarum – Between Research and Management, 112–132, Ljubljana. KOS, P. 2015, Ad Pirum (Hrušica) in Claustra Alpium Iulia-rum. – Vestnik XXVI. KUSETIČ et al. 2014 = J. KUSETIČ, P. KOS, A. BREZNIK, M. STOKIN (ur. / eds.), Claustra Alpium Iuliarum. Med raziskovanjem in upravljanjem / Claustra Alpium Iuliarum – Between Research and Management. – Ljubljana. KUSETIČ, J. 2015, Pregled novejših arheoloških raziskav na poznorimskem obrambnem sistemu Claustra Alpium Iuliarum / Overview of the recent research on the Late Roman Claustra Alpium Iuliarum defence system. – Arheo 32, 23–36. MACKENSEN, M. 1981, Die römischen Fundmünzen. – V / In: T.Ulbert, Ad Pirum (Hrušica). Spätrömische Passbefestigung in den Julischen Alpen, Münchner Beiträge zur Vor- und Frühgeschichte 31, 131–152. MILAVEC, T. 2017, Defending Italy from the north-east: Claustra Alpium Iuliarum and its interpretations. – V / In: M. Bohr, M. Teska (ur. / eds.), Extra limites, 149–162, Poznań–Wrocław. MÜLLNER, A. 1893, Die “Gradišča” in Krain. Das römische Castell: “Ad Pirum” in den Julischen Alpen. – Argo 2, 165–170. PICK, K., SCHMID, W. 1922–1924, Frühgeschichtliche Befestigungsanlagen im Bereiche der Isonzofront. – Ja­hreshefte des österreichischen archäologischen Institutes 21–22, Beiblatt, 277–308. SCHALLMAYER, E. 1990, Der römische Weihebezirk von Osterburken 1, Corpus der griechischen und lateinischen Beneficiarer-Inschriften des Römischen Reiches. – For-schungen und Berichte zur Vor- und Frühgeschichte in Baden-Württemberg 40. SCHMID, W. 1932, Die römische Poststation Noreia in Einöd. – Jahreshefte des österreichischen archäologischen Institutes 27, Beiblatt, 194–222. STICOTTI, P. 1920, Notizie degli scavi di antichita 17, 101. ŠAŠEL, J. 1970–1971, Alpes Iuliana. – Arheološki vestnik 21–22, 33–44 (= Šašel, J., Opera Selecta, Situla 30, Ljubljana 1992, 728–739). TALBERT, R. J. A. 2010, Rome’s World. The Peutinger Map Reconsidered. – Cambridge. ULBERT, T. 1981, Ad Pirum (Hrušica). Spätrömische Passbe­festigung in den Julischen Alpen. – Münchner Beiträge zur Vor- und Frühgeschichte 31. VEDALDI IASBEZ, V. 1994, La Venetia Orientale e l‘Histria. Le fonti letterarie greche e latine fino alla caduta dell‘Impero Ro­mano d‘Occidente. – Studi e ricerche sulla Gallia Cisalpina 5. ZACCARIA, C. 1992, Regio X. Venetia et Histria. Tergeste – Ager Tergestinus et Tergesti adtributus. – V / In: Supple-menta Italica, n. s. 10, 139–283, Roma. Peter Kos Narodni muzej Slovenije Prešernova cesta 20 SI-1000 Ljubljana peter.kos@nms.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 77–91 LONGATICUM – LOGATEC Ahac ŠINKOVEC Izvleček Rimski Longaticum se v antičnih kartografskih in itinerarskih virih omenja kot cestna postaja (mansio) ob glavni rimski cesti med Emono in Akvilejo. Naselbina je stala na območju današnjega Logatca v osrednjem delu Logaške kotline, kjer se je od prazgodovinske jantarjeve poti, ki je vodila čez prelaz Okra, odcepila rimska itinerarska cesta čez Hrušico. Pripisujemo ji naselbinske in grobiščne ostanke, ki so bili odkriti na območju zaselka Čevica v Dolenjem Logatcu. Osrednji del naselbine s cestno postajo je domnevno ležal v okolici današnjega Narodnega doma, od koder se je območje poselitve širilo približno 400 m daleč vzdolž itinerarske ceste proti severovzhodu. V času največjega obsega naselbine je njen skraj­ni severovzhodni del verjetno segal do križišča med Tržaško in Tovarniško cesto, kjer so bili odkriti naselbinski ostanki, datirani v čas od druge polovice 1. do sredine 2. st. V drugi polovici 2. st. je bil vzhodni del naselbine opuščen, poselitev pa se je verjetno skrčila na ožje naselbinsko jedro v okolici Narodnega doma. V poznorimskem času je bilo območje Logaške kotline vključeno v sistem cestnih zapor Claustra Alpia Iuliarum. Cestna postaja je verjetno izgubila svojo funkcijo in bila opuščena v času nemirnih notranje- in zunanjepolitičnih do-godkov ob koncu 4. in na začetku 5. st. Ključne besede: Italija (10. regija), Dolenji Logatec, Longaticum, rimska doba, naselbina, mansio, cestna postaja, cesta, križišče, grobišče Abstract Roman Longaticum is known from the Roman cartographic and itinerary sources as a roadside station (mansio) along the road between Emona and Aquileia. The settlement was situated in the area of present-day Logatec in the central part of the Logatec basin, where the Roman itinerary road across Hrušica parted from the prehistoric Amber Route, which led across the Ocra pass. The archaeological remains associated with the settlement were discovered in the area of Čevica in Dolenji Logatec. It is believed that the central part of the settlement with the roadside station was located in the vicinity of the present-day Narodni dom, from where the settlement territory spread for about 400 m along the itinerary road towards the northeast. At its peak, the far northeastern part of the settlement probably extended up to the present day crossroads between Tržaška and Tovarniška cesta streets, where settlement remains dating from the second half of the 1st century to the middle of the 2nd century were discovered. In the second half of the 2nd century, the eastern part of the settlement was abandoned, and the habitation area probably narrowed down to the settlement nucleus in the vicinity of Narodni dom. During the late Roman period, the Logatec basin was integrated into the Claustra Alpium Iuliarum fortification system. The roadside station probably lost its function and was abandoned during turbulent internal and foreign political events at the end of the 4th and the beginning of the 5th centuries. Keywords: Italy (Regio X), Dolenji Logatec, Longaticum, Roman period, settlement, mansio, road station, road, crossroad, cemetery https://doi.org/10.3986/9789610502586_04 LEGA Rimska naselbina Longaticum je ležala ob glavni cesti (via publica) med Emono in Akvilejo, v vzhodnem delu akvilejskega mestnega agra.1 Območje naselbine je upravno sodilo v X. italsko regijo, ki je obsegala mejno cono Italije s strateško pomembnim ozemljem v okolici italsko-ilirskih vrat, kjer je potekala naravna prometna povezava med severnojadranskim prostorom in osred­njo Slovenijo. Ta je vodila čez prelaz Razdrto (Ocra) in se preko Pivške kotline, Planinskega polja in Logaške kotline nadaljevala proti Ljubljanski kotlini. Po njej je v prazgodovinskem času potekala starodavna jantarjeva pot, ki je povezovala Apeninski polotok s celinsko Evro­po in Balkanom. Na območju Logaške kotline se je od nje odcepila rimska magistralna cesta, ki je vodila po krajši trasi čez prelaz Hrušica proti Akvileji.2 Antično ime naselbine3 se je ohranilo v poime­novanju današnjega Logatca, mesta v osrednjem delu Logaške kotline. To je netipično kraško polje pretočno ponornega tipa z ravno oblikovanim dnom, ki ga obdaja hribovit dinarski svet, sestavljen iz narivov Hrušiškega pokrova na jugu in jugovzhodu ter Trnovskega pokrova na severu. Ta sta med seboj ločena s prelomom, vzdolž katerega sta se oblikovala naravna prehoda, ki povezuje­ta kotlino s Planinskim poljem na jugu in Ljubljanskim barjem na severovzhodu. Z neprepustnega območja na severozahodni strani polja pritekata potoka Reka in Črni potok, ki se ob vstopu na polje združita v Logaščico. Ta teče čez zahodni del polja in na območju Dolenjega Logatca, kjer poteka meja med nepropustnimi in pro-pustnimi kamninami, ponikne v požiralnik, imenovan Jačka. Dno polja v tem delu je prekrito z aluvialnimi na­nosi, v katere je Logaščica vrezala nizko teraso, ugodno za poselitev. V vzhodnem delu polja je zaradi odsotnosti površinskih voda nastala suha dolina, imenovana Pusto polje (sl. 1).4 Današnji Logatec je nastal s spajanjem manjših zaselkov v naselbinsko celoto, ki je ohranila svojevrstno bipolarno urbanistično zasnovo, vidno še danes v delitvi na Gorenji in Dolenji Logatec. Arheološke raziskave so pokazale, da sta obe poselitveni jedri obstajali že v rimskem času.5 Na območju Gorenjega Logatca je domnevno stal vicus staroselskega prebivalstva, ki naj bi se priselilo iz prazgodovinske naselbine na bližnji vzpetini Velike Bukve, medtem ko je na območju Dole-njega Logatca stala rimska cestna postaja, ki se omenja v rimskih kartografskih in itinerarskih virih. Pripisujemo 1 Vzhodna meja akvilejskega agra je potekala v bližini Bevk na Ljubljanskem barju (Šašel Kos 2002). 2 Horvat, Bavdek 2009, 148–149. 3 Ime lahko izhaja iz keltščine: Repanšek 2016, 191–197. 4 Mihevc 1992, 312; Pagon 2008, 8–10. 5 Toponim Longatico je moral prvotno označevati obe naselbinski enoti oz. kar območje celotne Logaške kotline (Frelih 2003, 8–9; Repanšek 2016, 191). THE LOCATION The Roman settlement of Longaticum was situated along the main road (via publica) between Emona and Aquileia, in the eastern part of the Aquileia ager.1 It administratively belonged to the regio X of Italy, which comprised the border zone of Italy with strategically important territory in the vicinity of the Italo-Illyrian Gate. The most convenient passage between the North Adriatic and central Slovenia took place here, leading over Razdrto pass (Ocra) and across the Pivka basin, Planinsko polje, and the Logatec basin towards the Ljub­ljana basin. Its course was followed by the prehistoric Amber Route, which connected the Apennine peninsula with continental Europe and Balkans. In the area of the Logatec basin, the main Roman road parted from its course and continued along the shorter route across the Hrušica pass in the direction of Aquileia.2 The ancient name3 of the settlement was preserved in the name of present-day Logatec, which lies in the central part of the Logatec basin. The modern town de­veloped from several small hamlets united into a larger settlement with a unique bipolar urbanistic scheme, preserved in the still present division on Gorenji Logatec and Dolenji Logatec. Archaeological research has shown that both settlement cores existed in Roman times.4 In the area of Gorenji Logatec, a vicus of the autochthonous population probably existed, which might have been connected with the prehistoric settlement on the nearby hill of Velike Bukve. A roadside station, known from the Roman cartographic and itinerary sources, must have been situated though in the area of Dolenji Logatec. Its remains were discovered in Čevica at the foot of the Naklo hill (Fig. 1).5 RESEARCH HISTORY Considering the etymological resemblance between the ancient name Longaticum and the con­temporary German name of the settlement Lohitsch6 already Humanist historians and chroniclers assumed that an ancient settlement must have been situated in the place of present-day Logatec. The first to propose a 1 The eastern border of the Aquileia ager ran near Bevke in the Ljubljansko barje marshes (Šašel Kos 2002). 2 Horvat, Bavdek 2009. 3 The name could be of Celtic origin: Repanšek 2016, 191–197. 4 The toponym Longatico must have originally refered to both settlements or even to the whole area of the Logatec basin itself (Frelih 2003, 8–9; Repanšek 2016, 191). 5 Frelih 2002, 81; 2003, 12–16. 6 Šašel 1970, 521; in 1307, the name Logatsch also occurs, while in the 16th and 17th centuries the place was referred to as Logitzio, Ligatiz, Logatis, Louitsch, Logates, Logatus, Logitsch, Loitsch, etc. (Frelih 2003, 8, 47: note 15). 7 3 6 Naklo 1 Dolenji Logatec 4 1 Dolenji Logatec (˝evica) 4 Vodice pri Kalcah12 Pusto polje 2 Gorenji Logatec (Tabor) 5 Pod Ženš°kom 13 Ja°ka 3 Pod Ostrim vrhom 6 Pod Smrekovcem 9 Velike bukve Sl. 1: Lega prazgodovinskih in rimskih arheoloških najdišč na območju Logaške kotline. Fig. 1: The position of the prehistoric and Roman archaeological sites in the area of the Logatec basin. (Vir / Source: GERK [http://rkg.gov.si]). ji naselbinske in grobiščne ostanke, odkrite na območju zaselka Čevica, ki leži ob vznožju vzpetine Naklo ob severnem robu kotline (sl. 1).6 ZGODOVINA RAZISKAV Že humanistični zgodovinarji in kronisti so na podlagi sorodnosti med antičnim imenom Longaticum in sodobnim nemškim imenom kraja Lohitsch7 domne­vali, da je na območju današnjega Logatca stala antična naselbina. Prvi, ki je poskušal lego naselbine natančneje opredeliti, je bil Peter Hitzinger, ki je leta 1852 na podlagi razdalje med Longatikom in Navportom, navedene v Tabuli Peutingeriani,8 naselbino lociral na območje Do­ 6 Frelih 2002, 81; 2003, 12–16. Šašel 1970. Iz leta 1307 je znano tudi ime Logatsch, medtem ko naj bi v 16. in 17. st. kraj označevali z imeni Lo-gitzio, Ligatiz, Logatis, Louitsch, Logates, Logatus, Logitsch, Loitsch itd. (Frelih 2003, 8, 47, op. 15). 8 Tab. Peut. III, 5 / IV, 1. more accurate location of the settlement was Peter Hitz­inger, who pinpointed its position in the area of Dolenji Logatec based on the distance between Longaticum and Nauportus quoted on the Tabula Peutingeriana.7 Nevertheless, until the first archaeological excavations, it was widely believed that Longaticum was located in the area of Gorenji Logatec,8 where a bronze statue of Mercurius and several Roman coins were discovered in the late 19th century.9 The first Roman remains in the area of Dolenji Logatec came to light in 1978 during construction works in the Čevica area at the foot of the Naklo hill, where the remains of a Roman cremation cemetery were discov­ered.10 In 1980 and 1981, seven cremation graves were documented in the area, which were partially damaged 7 Tab. Peut. III 5 / IV 1. 8 Šašel 1970, 521. For a more detailed history of research, cf. Frelih 1988, 21; 2003, 8–9. 9 Frelih 1988, 21; 2002, 79, 81; 2003, 8; Petru 1975a, 191. 10 Frelih 2003, 10. Sl. 2: Ostanki rimske naselbine, grobišča in cest na območju Dolenjega Logatca. Fig. 2: The remains of the Roman settlement, cemetery and roads in the territory of Dolenji Logatec. (Vir / Source: TTN5, © Geodetska uprava RS / Surveying and Mapping Authority of the Republic of Slovenia). lenjega Logatca. Kljub temu je vse do prvih arheoloških odkritij na tem prostoru prevladovalo mnenje, da je bil Longaticum na območju Gorenjega Logatca,9 kjer so ob koncu 19. st. odkrili bronast kipec Merkurja in več rimskih novcev.10 Do odkritja prvih rimskodobnih ostankov v Do-lenjem Logatcu je prišlo leta 1978, ko so pri gradbenih delih na območju Čevice, na južnem pobočju hriba Naklo, naleteli na ostanke žganega rimskega grobišča.11 V letih 1980 in 1981 je bilo na tem območju dokumenti­ranih sedem žganih grobov, ki so bili pri gradbenih delih delno poškodovani.12 Na pobudo arheologa in domačina Marka Freliha, ki je bil navzoč pri odkritju, je bila leta 1987 nedaleč stran izkopana manjša sonda, v kateri so bili odkriti še trije žgani in en skeletni grob (sl. 2: 1).13 Odkritje grobišča je sprožilo vprašanje o obstoju sočasne rimske naselbine v bližini obravnavanega ob-močja, s katerega so bile do tedaj znane zgolj posamične 9 Šašel 1970; Šašel 1975, 92. Za podroben pregled zgodo-vine raziskav glej Frelih 1988, 21; 2003, 8–9. 10 Petru 1975a; 1988, 21; 2002, 79, 81; 2003, 8. 11 Frelih 2003, 10. 12 Frelih 1988, 21–22, sl. na str. 24. 13 Frelih 1988, 22–23, sl. na str. 24: gr. 1; 2003, 11. during construction works.11 A local and archaeologist Marko Frelih, who witnessed the discovery, conducted small research excavations nearby in 1987, where an­other three cremation and one inhumation grave were discovered (Fig. 2: 1).12 The discovery of the graveyard indicated the exis­tence of a Roman settlement in the near vicinity, where only stray finds occurred thus far.13 Its remains came to light in 1989 during the construction works in front of the Narodni dom (Fig. 2: 2). Small excavations were car­ried out under the surveillance of Frelih in the area where remains of a part of a Roman house were examined.14 In 1987 and 1988, the remains of a Roman settle­ment were also discovered during the regulation works on a creek Reka in Mareke in Gorenji Logatec (the area of Tabor in the old village centre; Fig. 1: 2). Considering the vicinity of a near-by Iron Age settlement on Velike Bukve, they were interpreted as remains of a vicus of autochthonous population.15 11 Frelih 1988, 21–22, fig. on p. 24. 12 Frelih 1988, 22–23, fig. on p. 24: gr. 1; 2003, 11. 13 Frelih, Mihevc 1985; 1988, 21–23; 2003, 10–11, 13. 14 Frelih 1989, 16–26; 2003, 14–15. 15 Frelih 2003, 12. naključne rimske najdbe.14 Rimski naselbinski ostanki so bili odkriti leta 1989 pri urejanju razsvetljave za košarkaško igrišče pred Narodnim domom (sl. 2: 2). Pod vodstvom Freliha je bila na tem mestu izkopana manjša sonda, v kateri so bili dokumentirani ostanki dela rimskega bivalnega objekta.15 V letih 1987 in 1988 so na rimske naselbinske ostanke naleteli tudi ob regulaciji potoka Reka na Ma-rekah v Gorenjem Logatcu (območje Tabora, v starem vaškem jedru; sl. 1: 2), ki so bili zaradi bližine železno­dobne naselbine na Velikih bukvah (sl. 1: 9) opredeljeni za ostanke vasi (vicus) staroselskega prebivalstva.16 Do novih odkritij na območju Dolenjega Logatca je prišlo leta 2002, ko so bili pri gradnji bencinske črpalke nasproti Narodnega doma odkriti ostanki dela rimske naselbine in ceste (sl. 2: 3).17 Rimski naselbinski ostanki so bili odkriti tudi v okviru zaščitnih raziskav pri obnovi Tržaške ceste. Raziskave ob gasilskem domu v letu 2006 (sl. 2: 4)18 so razkrile ostanke poselitve iz zgodnjerim­skega obdobja, medtem ko so bili pri raziskavah na območju križišča med Tržaško in Tovarniško cesto v letu 2013 (sl. 2: 5) poleg naselbinskih ostankov odkriti tudi ostanki rimske in poznorimske ceste.19 Omenjene raziskave so pokazale, da je bil obseg rimske poselitve na obravnavanem prostoru bistveno večji, kot se je prvotno domnevalo. PRAZGODOVINA O najstarejši poselitvi Logaške kotline20 pričajo posamične površinske najdbe kamnitih odbitkov z obrobja Pustega polja zahodno od Ostrega vrha (sl. 1: 3) in iz Vodic pri Kalcah (sl. 1: 4), ki bi lahko sodili že v paleolitik. Nekoliko mlajši sta površinski najdbi dveh kremenovih puščičnih osti, ki sta bili odkriti ob vznožjuvzpetine Ženček (sl. 1: 5) in sta okvirno datirani v čas od mlajše kamene dobe do bronaste dobe.21 Prvi dokazi o trajni poselitvi Logaške kotline so znani z najdišča Zapolje ob vznožju vzpetine Smreko­vec v severovzhodnem delu kotline, kjer so bili odkriti 14 Frelih, Mihevc 1985; Frelih 1988, 21–23; 2003, 10–11, 13. 15 Frelih 1989, 16–26; 2003, 14–15. 16 Frelih 2003, 12. 17 Frelih 2003, 15. 18 Logatec – arheološko najdišče Tabor. Raziskave je izvedel Zavod za varstvo kulturne dediščine Slovenije, OE Ljubljana; Topličanec 2007, 46–48. 19 Raziskave na najdišču Logatec – križišče je izvedlo podjetje Arhos d.o.o.; Šinkovec 2018. 20 Podroben pregled vseh znanih arheoloških najdišč na območju Logaške kotline podaja Matjaž Krašna v diplom­skem delu (Krašna 2017). 21 Frelih 1988, 11. New discoveries in the area of Dolenji Logatec were made during the construction of a gas station in the vicinity of Narodni dom in the year 2002, where remains of a Roman settlement and a road were dis­covered (Fig. 2: 3).16 Roman settlement remains were also discovered during preventive excavations for the reconstruction of the Tržaška cesta road. Excavations in the area of Gasilski dom in the year 2006 revealed settlement remains from the early Roman period (Fig. 2: 4),17 while at the crossroads between Tržaška and Tovarniška cesta streets in 2013 remains of a Roman and late Roman road were also discovered (Fig. 2: 5).18 The excavations revealed that the extent of the Roman settlement was far larger than had been assumed. PREHISTORY The oldest evidence of human presence in the area of the Logatec basin19 is represented by stray surface finds of stone splitters, discovered at the edge of Pusto polje to the west of Ostri vrh and in Vodice pri Kalcah (Fig. 1: 3-4), which could have already been from the Palaeolithic period. The stray finds of two quartz ar­rows discovered at the foot of the Ženček hill (Fig. 1: 5) were dated to the period from the Neolithic to the Bronze Age.20 The first evidence of permanent settlement of the Logatec area is known from the location Zapolje at the foot of the Smrekovec hill in the north-eastern part of the basin, where modest settlement remains from the Middle and Late Bronze Age were discovered (Fig. 1: 6).21 In the Iron Age, the area of the Logatec basin prob­ably belonged to the territory of the Notranjska group.22 In this period, the elevated areas along the edges of the basin were occupied in order to ensure control over natural passages in the vicinity. The remains of three hillforts are known from this period: the hillfort Strmica above Zaplana, which was situated by the passage to­wards the Ljubljansko barje marshes in the northeast of the basin, (Fig. 1: 7), the hillfort Brst nad Martinj hribom in the south of the basin, at the passage towards Planinsko polje (Fig. 1: 8), and the hillfort Velike Bukve above Gorenji Logatec in the west, at the passage towards 16 Frelih 2003, 15. 17 Logatec – Arhaeological site Tabor. Research was car­ried out by the Institute for the Protection of Cultural Heri­tage of Slovenia, Regional Office Ljubljana; Topličanec 2007, 46–48. 18 Research on the Logatec – Arhaeological site križišče was carried out by the company Arhos d.o.o.; Šinkovec 2018. 19 A detailed review of known archaeological sites in the area of the Logatec basin is given by Matjaž Krašna in his BA thesis (Krašna 2017). 20 Frelih 1988, 11. 21 Olić 2006. 22 Gabrovec 1999, 183–184. skromni naselbinski ostanki iz časa srednje in pozne bronaste dobe (sl. 1: 6).22 V železni dobi je območje Logaške kotline verjetno pripadalo ozemlju notranjske skupine, za katero je bila značilna poselitev višje ležečih predelov ob robovih kotline, ki so omogočali nadzor nad naravnimi prehodi v okolici.23 Iz tega časa so znani ostanki treh utrjenih gradišč na obrobju kotline: na severovzhodu kotline, ob prehodu proti Ljubljanskemu barju, je stalo gradišče Strmica nad Zaplano (sl. 1: 7), na jugu, ob prehodu proti Planinskemu polju, gradišče Brst nad Martinj Hribom (sl. 1: 8) in na zahodu, ob prehodu proti dolini Idrijce, gradišče Velike Bukve nad Gorenjim Logatcem (sl. 1: 9).24 Ostanki železnodobne poselitve so bili odkriti tudi na območju vzpetine Ženček in pod njo ležeče ledine Tončeva ravan (sl. 1: 10),25 ter na območju vzpetine Sekirica (sl. 1: 11).26 Kot pričajo odlomki železnodobne keramike, odkrite na najdišču Pusto polje severno od Dolenjega Logatca (sl. 1: 12)27 ter na najdišču Jačka v osrednjem delu Dolenjega Logatca (sl. 1: 13),28 je bilo v tem času intenzivno poseljeno tudi osrednje, dolinsko območje kotline. Rezultati sondažnih raziskav na gra­dišču Velike Bukve (sl. 1: 9) kažejo, da so vsaj nekatere prazgodovinske lokacije ostale obljudene vse do prihoda Rimljanov ob koncu 1. st. pr. n. š.29 LITERARNI IN EPIGRAFSKI VIRI Ime kraja Longaticum je znano iz rimskih karto­grafskih in itinerarskih virov.30 Najstarejša omemba kraja izvira iz kopije antičnega zemljevida Tabula Peutingeriana, kjer je ime Long[at]ico31 navedeno ob glavni cesti med Akvilejo in Emono, med krajema In Alpe Iulia (Hrušica, Lanišče ali širše območje Kalc)32 in Nauporto (Vrhnika).33 Naselbina se omenja tudi v rimskih itinerarijih Itinerarium Antonini s konca 3. st. in Itinerarium Burdigalense iz prve polovice 4. st., kjer je Longaticum označen kot mansio.34 To je bila uradna 22 Olić 2006. 23 Gabrovec 1999, 183–184. 24 Frelih 1988, 12–14. 25 Ministrstvo za kulturo RS, Register nepremične kul­turne dediščine (RKD) [http://rkd.situla.org/], evidenčna št. enote dediščine (EŠD) 11108. 26 RKD, EŠD 11106. 27 RKD, EŠD 29552. 28 Frelih 1988, 16, op. 11. 29 Frelih 1988, 13. Pri sondiranju na gradišču Velike Buk­ve je bilo v nasipu naselbine odkritih več svinčenih želodov, ki nakazujejo na možnost rimskega obleganja naselbine. 30 Šašel 1970, 521. 31 Tab. Peut. III, 5. 32 Za lokacijo kraja In Alpe Iulia prim. Šašel 1975, 92 in Kos 2014, 41–42, s starejšo literaturo. 33 Šašel 1975, 92. 34 It. Ant. 129, 1: Longatico mansio; It. Burd. 560, 5: man- the Idrijca valley (Fig. 1: 9).23 Settlement remains from the Iron Age period were also discovered in the area of the Ženček hill and the adjacent Tončeva ravan (Fig. 1: 10)24 as well as in the area of the Sekirica hill (Fig. 1: 11).25 and in the location Pusto polje north of Dolenji Logatec (Fig. 1: 12).26 A larger quantity of Iron Age pottery was also found on the site Jačka in the central part of Dolenji Logatec (Fig. 1: 13).27 This indicates that the central lowland area of the basin was intensively inhabited during this period as well. According to the results of trench excavations at Velike Bukve hillfort, at least some prehistoric sites remained populated until the arrival of Romans at the end of the 1st century BC. 28 LITERARY AND EPIGRAPHIC SOURCES The name Longaticum is known from Roman car­tographic and itinerary sources.29 The earliest mention of the settlement originates from the copy of an ancient map Tabula Peutingeriana, where the name Long[at]ico30 is stated along the Aquileia–Emona road, between In Alpe Iulia (Hrušica, Lanišče or the wider area of Kalce)31 and Nauporto (Vrhnika).32 Longaticum is also mentioned in the Roman itineraries: the Itinerarium Antonini from the late 3rd century and the Itinerarium Burdigalense from the first half of the 4th century, where the settlement is at­tested as mansio.33 This was an official term for a roadside station, which originally served as a service facility for travelling emperors and high officials, from the 3rd century onwards though, it became an indispensable part of the Roman postal system (cursus publicus).34 In the Itiner­arium Antonini, the station is stated as being between Fluvio Frigido (Ajdovščina) and Hennoma civitas (i.e., Emona)35 and in the Itinerarium Burdigalense between ad Pirum summas Alpes (Hrušica) and mutatio Ad nonum (supposedly near Log pri Brezovici).36 23 Frelih 1988, 12–14. 24 Ministry of Culture of the Republic of Slovenia, Regis­ter of Immovable Cultural Heritage (RKD) [http://rkd.situla.org/], heritage identity number (EŠD) 11108. 25 RKD, EŠD 11106. 26 RKD, EŠD 29552. 27 Frelih 1988, 16, note 11. 28 Frelih 1988, 13. In the trench excavated on the Velike Bukve hillfort, a large number of lead slingshots was discov­ered in the rampart, indicating a possible Roman siege. 29 Šašel 1970, 521. 30 Tab. Peut. III 5 / IV 1. 31 For the location of the settlement In Alpe Iulia cf. Šašel 1975, 92 and Kos 2015, 41–42, with earlier bibliography. 32 Šašel 1975, 92. 33 It. Ant. 129, 1: Longatico mansio; It. Burd. 560, 5: man-sio Longatico. 34 Šašel Kos 1997, 24; Kolb 1999, 829. 35 Šašel 1975, 77. 36 Šašel 1975, 78. oznaka za cestno postajo, ki je bila prvenstveno name-njena oskrbi in prenočevanju cesarjev in drugih uradnih oseb, od 3. st. dalje pa je bila neposredno povezana z delovanjem rimske poštne službe (cursus publicus).35 Postaja je v Antoninskem itinerariju navedena med krajema Fluvio Frigido (Ajdovščina) in Hennoma civitas (tj. Emona),36 v itinerariju Burdigalense pa med krajema ad Pirum summas Alpes (Hrušica) in mutatio Ad nonum (domnevno v bližini Loga pri Brezovici).37 O poteku cest v okolici Longatika pričata najdbi dveh miljnikov, ki ju omenja Janez Dolničar v delu Epitome chronologica iz leta 1714. Prvi naj bi bil odkrit leta 1686 v Vodicah pri Kalcah,38 drugi, ki naj bi prav tako izviral iz bližnje okolice Logatca, pa leta 1699.39 Oba sta danes žal izgubljena, prepisa besedil pa sta nezanesljiva.40 NASELBINA V DOLENJEM LOGATCU (LONGATICUM) Longaticum je stal ob glavni rimski cesti med Akvilejo in Emono, ki je najverjetneje potekala po trasi današnjega kolovoza pod vzpetino Sekirica in se vzdolž ceste skozi Čevico nadaljevala skozi Dolenji Logatec. Na območju Čevice so že pred 2. svetovno vojno ob obnav­ljanju cestišča in izkopu jarkov za odtočne kanale nale­teli na večje kamnite plošče, ki so domnevno pripadale rimski cesti.41 Ostanki ceste pa so bili prvič arheološko dokumentirani šele leta 2002 pri gradnji nove bencinske postaje nasproti Narodnega doma (sl. 2: 3)42 ter leta 2013 v severozahodnem delu Tovarniške ceste (sl. 2: 5).43 Re-zultati raziskav na območju Tovarniške ceste so pokazali, da je bila cesta široka približno 8 m in je imela podlago iz masivnih, do 50 cm velikih apnenčevih lomljencev, ki jih je prekrival tlak iz prodnikov. V nasutju ob cesti sta bili najdeni dve železni obuvali za kopita, ki sta okvirno datirani v čas od sredine 1. do konca 2. st. (sl. 3).44 Na podlagi novcev, odkritih na tem mestu, lahko sklepamo, da je cesta ostala v uporabi še v času 3. in 4. st.45 sio Longatico. 35 Šašel Kos 1997, 24; Kolb 1999, 829. 36 Šašel 1975, 77. 37 Šašel 1975, 78. 38 CIL III, 4614. 39 CIL III, 413*. 40 Petru 1975a, 191; 1975b, 191; Šašel 1975, 83 (42 in 43); Frelih 1988, 36–40, sl. na str. 41; 2003, op. 78.41 Frelih 2003, 13, op. 29. 42 Frelih 2003, 15. 43 Arheološko najdišče Logatec – križišče; Šinkovec 2018, 42–44. 44 Šinkovec 2018, 68. 45 Novčne najdbe z raziskav na križišču med Tržaško in Tovarniško cesto je opredelila Alenka Miškec iz Narodnega muzeja Slovenije. The course of the roads around Longaticum is at­tested by two milestones mentioned by Janez Dolničar in his Epitome chronologica from the year 1714; the first one was supposedly discovered in 1686 in Vodice pri Kalcah,37 while the other one was found in the near surroundings of Logatec in 1699.38 Both have been lost, and their transcriptions are unreliable.39 THE SETTLEMENT IN DOLENJI LOGATEC (LONGATICUM) Longaticum was situated along the main road be­tween Aquileia and Emona, which followed the course of the present-day cart-route under the Sekirica hill and continued along the Čevica road through Dolenji Logatec. Before the 2nd World War, larger stone slabs were discovered during the road reconstruction in the area of Čevica, which arguably represented the remains of a Roman road.40 Nevertheless, the remains of the road were archaeologically documented for the first time yet in the area of the new gas station opposite the Narodni dom in 2002 (Fig. 2: 3)41 and in the north-western part of Tovarniška cesta (Fig. 2: 5) in 2013.42 The results of the excavations in the area of Tovarniška cesta indicate that the road was about 8 m wide and laid on a base made of up to 50 cm large limestone quarry stones, which were covered with pebble paving. By the road two metal hipposandals were found, which are roughly dated from the middle of the 1st century to the end of the 2nd century (Fig. 3).43 According to the coin finds from the excavations, it is assumed that the road stayed in function until the 3rd and 4th centuries.44 As Frelih speculated,45 the road station must have been located in the vicinity of the crossroads between the itinerary road and the ancient route via the Ocra pass. It is supposed that the latter one branched off towards south in the area of Dolenji Logatec and ran at the foot of Martinj Hrib along the course of present-day Notranjska cesta road. The contemporary settlement, which was probably adjacent to the station,46 suppos­edly spread some 400 m along the main road between 37 CIL III, 4614. 38 CIL III, 413*. 39 Petru 1975a, 191; 1975b, 191; Šašel 1975, 83 (42 and 43); Frelih 1988, 36–40, fig. on p. 41; 2003, note 78. 40 Frelih 2003, 13, note 29. 41 Frelih 2003, 15. 42 Arhaeological site Logatec – križišče; Šinkovec 2018, 42–44. 43 Šinkovec 2018, 68. 44 Coins from the excavations at the crossroads between Tržaška and Tovarniška cesta streets were determined by Alenka Miškec from the National Museum of Slovenia. 45 Frelih 1988, 26. 46 Stations were usually located outside settlements or even on their own (Šašel Kos 1997, 24). Sl. 3: Obuvalo za kopito z najdišča Logatec – križišče. Fig. 3: A hipposandal from the site Logatec - križišče. (Foto / Photo: S. Olić, © Arhos d.o.o.) Kot je domneval že Frelih,46 je cestna postaja najverjetneje ležala v neposredni bližini križišča med itinerarsko cesto in starodavno cesto čez prelaz Okra, ki se je na območju Dolenjega Logatca odcepila proti jugu in je domnevno vodila ob vznožju Martinj Hriba po trasi današnje Notranjske ceste. Sočasna naselbina, ki je najverjetneje stala ob cestni postaji,47 se je domnevno raztezala v približno 400 m dolgem pasu vzdolž glavne ceste med Narodnim domom na zahodu (sl. 2: 2) ter današnjim križiščem med Tržaško in Tovarniško cesto na vzhodu (sl. 2: 5). Na ožjem naselbinskem območju do danes niso bili odkriti ostanki starejše, predrimske poselitve, ki bi kazali na staroselsko zasnovo naselja. Najstarejši rimskodobni poselitveni ostanki so bili odkriti leta 2006 ob gasilskem domu na Tržaški cesti, kjer so bili dokumentirani ostanki lesenega objekta z maltnim tlakom in več ognjišči, ki so bili okvirno datirani v avgustejsko-tiberijsko obdobje (sl. 2: 4).48 Med najdbami so prevladovali odlomki uvožene­ga italskega posodja, najdena pa je bila tudi fibula tipa Aucissa. Nad naselbinskimi plastmi je ležala masivna konstrukcija, sestavljena iz do 50 cm velikih apnenčevih lomljencev, prekritih s slojem peska, ki jo po načinu gradnje in orientaciji lahko povežemo z že omenjenimi ostanki ceste, odkrite leta 2013 v severozahodnem delu Tovarniške ceste (sl. 2: 5).49 46 Frelih 1988, 26. Postaje so praviloma stale zunaj obstoječih naselbin, pogosto pa so bile celo na samem (Šašel Kos 1997, 24).48 Logatec – arheološko najdišče Tabor (Topličanec 2007, 46–48).49 Šinkovec 2018. Narodni dom in the west and the crossroads between Tržaška and Tovarniška cesta streets in the east. Thus far, in this area, no traces of earlier, pre-Roman settlement were found that would indicate the autoch­thonous origin of the settlement. The earliest settlement remains were discovered in the area of Gasilski dom (firehouse) in Tržaška cesta, where remains of a wooden building with mortar paving and several fireplaces, dated to the Augustan-Tiberian period, were found (Fig. 2: 4).47 A fibula of the Aucissa type was discovered there as well as the fragments of mainly imported Italian vessels. Above the settlement layers, lay a massive structure consisting of up to 50 cm large limestone quarry stones covered with a layer of sand. Regarding its construction and orientation, the structure resembled the remains of the road found in the north-western part of Tovarniška cesta.48 The central part of the Roman settlement with a road station supposedly lay in the area of the basketball court in front of Narodni dom (Fig. 2: 2). In 1989, a small trench was dug here under the surveillance of Marko Frelih, and the remains of a part of a building were 47 Logatec – Archaeological site of Tabor (Topličanec 2007, 46–48). 48 Šinkovec 2018. Osrednji del rimske naselbine s cestno postajo je domnevno ležal na območju košarkarskega igrišča in zelenice pred Narodnim domom (sl. 2: 2). Leta 1989 je bila na tem mestu pod vodstvom Freliha izkopana manjša sonda, v kateri so bili odkriti ostanki dela večje­ga bivalnega objekta, datiranega v čas od 1. do 4. st.50 Objekt je imel domnevno leseno konstrukcijo, od katere so se ohranili zgolj kamniti temelji, grajeni v suhozidni tehniki, medtem ko je bila notranjost objekta prekrita s tlakom iz zbite ilovice. Odkriti so bili ostanki dveh poselitvenih faz, ločeni med seboj s plastjo žganine, ki jo Frelih povezuje z opustošenjem naselbine ob vpadu Kvadov in Markomanov okrog leta 170. Med značilnimi predmeti naselbinske materialne kulture starejše faze, datirane v čas 1. in 2. st., je izstopala steklena skodela iz brezbarvnega stekla, okrašena z vrezanim okrasom, ki je sodila med izdelke najvišjega kakovostnega razreda zgodnjecesarske steklarske produkcije.51 Domneva se, da se je naselbina proti severovzhodu širila do današnjega križišča med Tržaško in Tovarniško cesto, kjer so bili pri raziskavah leta 2013 odkriti ostanki dela naselbinskega kompleksa (sl. 2: 5).52 Čez to območje je pred izgradnjo naselbine potekala manjša stranska cesta, ki je vodila od glavne ceste proti severu. Grajena je bila iz več plasti peska in lomljencev, ki so tvorile dobre 3 m široko cestišče, obdano na zahodni strani z enim, na vzhodni strani pa z dvema drenažnima jarkoma (sl. 4). Po opustitvi ceste v drugi polovici 1. st. je bil prostor zahodno od nje oblikovan v rahlo privzdignjen plato, na katerem je bila zgrajena naselbina. Odkriti so bili ostanki dveh naselbinskih faz. Iz starejše faze so izvirali ostanki lesenih stavb, od katerih so se ohranili tlaki iz zbite ilovice, datirani v zaključni del 1. in na začetek 2. st. V mlajši fazi je bil na njihovem mestu zgrajen večji objekt z vkopanimi kamnitimi temelji, grajenimi v suhozidni tehniki, ki je imel verjetno lesen pod in leseno nadzemno konstrukcijo, prekrito z opečnato strešno kritino (sl. 5). Opuščen je bil na začetku druge polovice 2. st. Med gradivom iz naselbinskih plasti53 so poleg odlomkov uvoženega italskega posodja, predvsem amfor, izstopali številni odlomki dvoročajnih bikoničnih skled (sl. 6), ki so verjetno predstavljale izdelke lokalnih delavnic, saj so bile do zdaj v večjem številu znane predvsem z okoliških najdišč (npr. z grobišča v Cerknici).54 Glede na odsot­nost značilnih ostankov naselbinskih aktivnosti, kot so ognjišča, objekti verjetno niso imeli bivalne funkcije. 50 Frelih 1989, 18–27; 2002, 81, sl. 2, 3; 2003, 14–16. 51 Lazar 2003, 42–44; 2004, 58, sl. 2: 1. 52 Arheološko najdišče Logatec - križišče; Šinkovec 2018. Za analizo gradiva, odkritega na najdišču Logatec – križišče, glej Šinkovec 2018. 54 Urleb 1983, t. 3: 1; 13: 4; 18: 2; 21: 6. Podobne sklede se pojavljajo tudi na Ulaki nad Starim trgom pri Ložu (Gaspa-ri 2000), na Uncu in v Ljubljani ter v strugi reke Ljubljanice (ustna informacija Andrej Gaspari, Filozofska fakulteta Uni­verze v Ljubljani, Oddelek za arheologijo). excavated.49 The structure of the building was probably wooden and only stone foundations, built-in drywall technique, preserved, while its interior was paved with beaten loam. According to the finds the building was dated to the period between the 1st and 4th centuries. Two settlement phases were detected, separated by a layer of burnt residue, which Frelih links with the destruction of the settlement during the incursion of the Marcomanni and Quadi around the year 170. Among the finds from the earlier phase (between the 1st and 2nd centuries), a bowl of decoloured glass with facet-cut decoration de­serves special attention: it can be attributed to the highest quality glass products of the early Roman Empire.50 It is assumed that at its peak the northeastern part of the settlement extended to the crossroads between Tržaška and Tovarniška cesta streets, where the remains of a part of a settlement complex were excavated in 2013 (Fig. 2: 5).51 In the period before the construction of the settlement, a smaller side road passed the area, leading from the main road to the north. The road was approximately 3 m wide and built from several layers of sand and quarry stone (Fig. 4). Three drainage canals ran along the road; one along the western edge and two along the eastern edge. After the abandonment of the road in the second half of the 1st century, the ground to the west of it was shaped into an elevated plateau, on top of which the settlement was built. The remains of two settlement phases were detected. From the earlier phase, the remains of pavements from beaten loam preserved which probably belonged to wooden buildings, dated to the end of the 1st century or the beginning of the 2nd 49 Frelih 1989, 18–27; 2002, 81, figs. 2, 3; 2003, 14–16. 50 Lazar 2003, 42–44; 2004, 58, fig. 2: 1. 51 Arhaeological site Logatec - križišče; Šinkovec 2018. Sl. 6: Dvoročajne bikonične sklede iz naselbinskih plasti na najdišču Logatec – križišče. Keramika. M. = 1:4. Fig. 6: Two-handled biconical bowls from the settlement layers on the site Logatec - križišče. Ceramics. Scale 1:4. (Risba / Drawing: J. Brečič, © Arhos d.o.o.) Grobišče: Ostanki grobišča so bili odkriti vzhodno od cerkve sv. Jožefa na območju Čevice v Dolenjem Logatcu (parc. št. 30 in 37/1, k. o. Blekova vas, sl. 2: 1). Na tem mestu sta bili pri gradbenih delih v letu 1978 najdeni dve oljenki in več odlomkov rimske keramike, ki so domnevno pripadali grobnemu inventarju. Poleg je bil odkrit 3 m dolg in 0,5 m širok suhozidni temelj, katerega funkcija in časovna pripadnost nista bili ugotovljeni. Pri nadaljevanju gradbenih del v letih 1980 in 1981 je bilo dokumentiranih sedem žganih grobov, ki so bili pri izkopu delno uničeni. Od najdb iz grobov se je ohranila zgolj ena oljenka.55 Načrtne raziskave grobišča so bile izvedene v letu 1987 nedaleč stran, kjer so bili odkriti še trije žgani in en skeletni grob. Za žgane grobove je bila značilna enostavna grobna jama, na dno katere je bila položena žganina pokojnika, medtem ko je skeletni grob vseboval slabo ohranjene ostanke lesene krste. V njem sta bila naj­dena bronasta pasna spona in lonček, ki je bil položen k nogam pokojnika.56 Na podlagi ohranjenih najdb lahko grobove okvirno datiramo v čas od 2. do 4. st., medtem ko se datacija v 1. st. zdi manj verjetna. Poznorimska cesta: Iz zaključne faze obstoja naselbine izvirajo ostanki poznorimske ceste, grajene iz oblic in posameznih večjih lomljencev, ki je bila odkrita v severozahodnem delu To­ 55 Frelih, Mihevc 1985; Frelih 1988, 21–23, sl. na str. 24; 1991, 19; 2003, 10. 56 Frelih 1988, 23, sl. na str. 24: gr. 1; 1991, 19–20; 2002, 81; 2003, 11. century, were found. In the second phase, a larger build­ing with dry stone foundations was built in their place. It probably had wooden walls and floors and was covered with tile roofing (Fig. 5). The building was abandoned at the beginning of the second half of the 2nd century. Among the small finds from the settlement layers, 52 the shards of imported Italian pottery, mostly of amphorae, are predominant. In contrast, numerous fragments of two-handled biconical bowls were discovered (Fig. 6), which were probably a product of local workshops, since they are known mainly from the sites in the region (e.g., the Cerknica cemetery).53 Given the absence of typical residuals of settlement activities, such as hearths, it is assumed that the buildings had no residential function. Cemetery: The remains of the cemetery were discovered east of the church of St Jožef in the area of Čevica in Dolenji Logatec (plots 30 and 37/1, cadastre municipality Ble­kova vas, Fig. 2: 1). During the construction works in the year 1978 two oil lamps and a few shards of Roman pottery were found there, which arguably belonged to the grave inventory. A 3 m long and 0.5 m wide wall was uncovered, the function of which is not clear. During the continuation of the works in 1980 and 1981 remains of 52 For analysis of the material discovered on the Logatec–križišče site, cf. Šinkovec 2018. 53 Urleb 1983, pl. 3: 1; 13: 4; 18: 2; 21: 6. Similar bowls oc­cur in Ulaka above Stari trg pri Ložu (Gaspari 2000), in Unec and in Ljubljana as well as in the riverbed of the Ljubljanica River (oral information Andrej Gaspari). varniške ceste (arheološko najdišče Logatec - križišče) in je tekla proti jugovzhodu, v smeri Planinskega polja (sl. 2: 5; 7).57 Njen nastanek lahko na podlagi novčnih najdb datiramo v drugo polovico 4. st.,58 verjetno pa je ostala v uporabi še v zgodnjesrednjeveškem obdobju.59 NASELBINI V GORENJEM LOGATCU IN V VODICAH PRI KALCAH Sočasno z rimsko naselbino v Dolenjem Logatcu sta na območju Logaške kotline obstajali še dve na­selbinski jedri. Prvo je ležalo na območju Gorenjega Logatca (sl. 1: 2), kjer je bilo že v sedemdesetih letih 19. st. odkrito več rimskih novcev, od tu pa naj bi izvirala tudi naključna najdba bronastega kipca Merkurja ter več rimskih napisnih kamnov.60 Na območju Tabora v starem vaškem jedru so bili pri regulaciji struge potoka Reka na Marekah v letih 1987 in 1988 po naključju od­kriti ostanki dveh objektov s kamnitimi temelji, verjetno 57 Šinkovec 2018, 45–46. 58 Novčne najdbe z raziskav na najdišču Logatec – križiš-če je opredelila Alenka Miškec (glej Šinkovec 2018).59 O uporabi ceste vsaj še tekom 5. st. pričajo odlomki več železnih podkev.60 Petru 1975a. Sl. 7: Ostanki poznorimske ceste na najdišču Logatec – križišče. Fig. 7: The remains of the late Roman road on the site Loga­tec – križišče. (Foto / Photo: M. Vujasinović, © Arhos d.o.o.) seven cremation graves were documented, which were partially damaged in the course of works. Only one oil lamp was preserved from the graves.54 Small research excavations of the cemetery were car­ried out in the near vicinity in 1987, where another three cremation and one inhumation grave were discovered. The cremation graves had simple burial pits, filled with burnt material, while in the inhumation grave remains of the wooden coffin were found. The latter comprised a bronze belt buckle and a pottery vessel, laid to the feet of the deceased.55 Based on the survived finds, the graves can be roughly dated to the period from the 2nd to the 4th centuries, while dating in the 1st century seems less likely. Late Roman road: The Late Roman road, built of pebbles and indi­vidual large quarry stones, was discovered in the north­western part of Tovarniška cesta street (archaeological site Logatec - križišče). It led towards the south-east, in the direction of Planinsko polje (Figs. 2: 5; 7).56 Accord­ing to the coin finds,57 it is dated to the second half of the 4th century, but it probably remained in use even in the early medieval period.58 THE SETTLEMENTS IN GORENJI LOGA­TEC AND IN VODICE NEAR KALCE Besides the Roman settlement in Dolenji Logatec the remains of another two contemporary settlements are known from the Logatec basin. One lay in the ter­ritory of present-day Gorenji Logatec (Fig. 1: 2), where several Roman coins, as well as a bronze statue of Mer­cury and a few inscription stones, were discovered in the 1870s.59 In the area of Tabor in the old village centre, the remains of two buildings with stone foundations were 54 Frelih, Mihevc 1985; Frelih 1988, 21–23, fig. on p. 24; 1991, 19; 2003, 10. 55 Frelih 1988, 23, fig. on p. 24: gr. 1; 1991, 19–20; Frelih 2002, 81; 2003, 11. 56 Šinkovec 2018, 45–46. 57 Coins from the excavations on the Logatec–križišče site were determined by Alenka Miškec (cf. Šinkovec 2018).58 The use of the road at least in the 5th century is attested by the finds of metal horseshoes.59 Petru 1975a. z leseno kostrukcijo in opečnato strešno kritino.61 Pri odkritju je bila poleg značilnega naselbinskega gradiva, predvsem odlomkov keramičnega in steklenega posodja ter delov noše, najdena velika količina železove žlindre, ki je bila najverjetneje povezana s kovaško dejavnostjo. Odkriti so bili tudi štirje poznorimski novci iz časa 3. in 4. st. Odkrito gradivo, predvsem keramične najdbe, kažejo na povezavo s starejšo prazgodovinsko naselbi-no na vzpetini Velike Bukve nad Gorenjim Logatcem. Domneva se, da je na tem mestu stal vicus staroselskega prebivalstva.62 Rimski naselbinski ostanki so bili odkriti tudi na območju Vodic pri Kalcah v jugozahodnem delu Loga­ške kotline (sl. 1: 4). Domnevna naselbina je ležala ob trasi rimske ceste, ki je vodila iz Logatca na Hrušico, na ravnini neposredno pred vzponom na Lanišče, kjer so bili leta 1988 dokumentirani ostanki rimske ceste, vsekane v skalnato pobočje.63 Na najdišču so bili ob gradnji plinovoda v letih 1981 do 1984 najdeni številni raznovrstni rimski predmeti in novci iz 1. st. pr. n. š. do 4. st. n. št.64 V bližini je bilo izkopanih tudi več naključnih kovinskih najdb, odkritih s pomočjo detektorja kovin.65 Najdbe verjetno pripadajo rimski obcestni postojanki ali vikusu z lesenimi zgradbami, saj ostanki zidanih objektov pri raziskavah niso bili odkriti.66 61 Frelih 1988, 25; 2003, 12. Z obravnavano naselbino lahko morda povežemo tudi naključne najdbe rimske kera-mike, ki je bila odkrita pri gradnji prizidka osnovne šole v Gorenjem Logatcu (Krašna 2017, 34–35). 62 Frelih 2003, 12. 63 Frelih 1988, 36. 64 Frelih 1988, 36; 2003, 26; Pflaum 2007, 316. 65 Naključne kovinske najdbe, ki jih hrani Vojni muzej v Logatcu, je obdelala Veronika Pflaum. Del najdb je po vsej ver­jetnosti pripadal poznorimski zakladni najdbi (Pflaum 2007). 66 Frelih 1988, 36; Pflaum 2007, 316. Nekateri avtorji (Frelih 2003, 29) domnevajo, da se je na tem mestu stala rim-ska preprežna postaja (mutatio) In Alpes Iulia, ki je omenjena na Tabuli Peutingeriani. V prid tej domnevi govori lega na­selbine tik ob vznožju vzpona na Lanišče, kjer je bil zadnji vir pitne vode pred vstopom v hribovit kraški svet Hrušice. Z zadržkom je potrebno vzeti mnenje Janeza Švajncerja, ki je na to mesto lociral cestno postajo mansio Longatico (Švajncer 2004, 13). accidentally uncovered during regulation works on the Reka creek in 1987 and 1988.60 The structure of the buildings was probably wooden and covered with tile roofing. Beside typical settlement finds, among which shards of pottery and glass, as well as parts of attire, prevailed, a large amount of iron slag was discovered, which was interpreted as the remains of a smithery. The settlement stayed inhabited until the late Roman period, as is attested by four late Roman coins dated to the 3rd and 4th centuries. According to discovered material, es­pecially pottery, it seems that the settlement was directly connected with the prehistoric settlement, located on the Velike Bukve hill. Frelih assumes that the settlement represented a vicus of autochthonous population.61 Roman settlement remains were also discovered in the territory of Vodice near Kalce in the southwestern part of the Logatec basin (Fig. 1: 4). The settlement was located on the plain along the course of the Roman road, which led from Logatec to Hrušica. From Vodice the road began to ascent in the direction of Lanišče, where a twenty-metre long section of the road cut in the rocky slope was documented in 1988.62 During construction works on the gas pipeline from 1981 to 1984 various Roman finds and coins dated from the 1st century BC to the 4th century AD were discovered on the site.63 Also, numerous chance metal finds were found in the area, acquired with the use of metal detectors.64 The finds probably belong to the Roman roadside post or vicus. Since no traces of buildings built of stone were discovered it is assumed that the settlement consisted mainly of wooden houses.65 60 Frelih 1988, 25; 2003, 12. Stray finds of the Roman pottery were also discovered during extension construction of the Elementary school in Gorenji Logatec (Krašna 2017, 34–35). 61 Frelih 2003, 12. 62 Frelih 1988, 36. 63 Frelih 1988, 36; 2003, 26; Pflaum 2007, 316. 64 Chance metal finds, which are kept in War Museum in Logatec, have been annalysed by Veronika Pflaum. A part of the finds probably belonged to the late Roman hoard (Pflaum 2007). 65 Frelih 1988, 36; Pflaum 2007, 316. Some authors (Fre­lih 2003, 29) believe that this was the location of the Roman changing station (mutatio) In Alpes Iulia, which is men­tioned in Tabula Peutingeriana. In favour of this assumption is the position of the settlement at the foot of the ascent to Lanišče, where the last source of drinking water was before entering the hilly karst region of Hrušica. In contrast, the as­sumption of Janez Švajncer, who believes that this was the location of the roadside station mansio Longatico, seems doubtful (Švajncer 2004, 13). ZGODOVINSKI RAZVOJ LONGATIKA Nastanek cestne postaje Longaticum je bil tesno povezan z izgradnjo rimske državne ceste med Emono in Akvilejo, ki se je na območju Logaške kotline odce­pila od starodavne prazgodovinske poti čez Razdrto in je vodila po najkrajši možni trasi čez prelaz Hrušica proti Italiji.67 Ta strateško pomemben cestni odsek je bil najverjetneje zgrajen v času cesarja Avgusta, o čemer poročata antična zgodovinarja Tacit (druga polovica 1. in začetek 2. st.) in Festus (druga pol. 4. st.).68 Iz časa neposredno pred izgradnjo ceste domnevno izvirajo ostanki zgodnjecesarske poselitve ob gasilskem domu v Dolenjem Logatcu (sl. 2: 4), ki jih lahko povežemo z nastanitvijo prvih italskih priseljencev. Naselbina s cestno postajo je najverjetneje ležala v bližini križišča med itinerarsko cesto, ki je vodila proti Kalcam in Hrušici, in cesto, ki se je odcepila proti jugu mimo Planinskega polja (Stari grad pri Uncu) in se na­daljevala čez prelaz Razdrto (sl. 1). Glede na dosedanje raziskave lahko sklepamo, da se je v času 1. st. in prve polovice 2. st. naselbina razprostirala na približno 400 m dolgem območju vzdolž glavne ceste od Narodnega do-ma na jugozahodu (sl. 2: 2) do križišča med Tržaško in Tovarniško cesto na severovzhodu (sl. 2: 5). Ob koncu 2. st. je bil severovzhodni del naselbine opuščen, njen obseg pa se je domnevno skrčil na ožje poselitveno jedro na območju današnjega Narodnega doma. Ta spremem­ba je bila morda posledica nemirnih časov ob vpadu Markomanov in Kvadov v drugi polovici 2. st., s katerimi Frelih povezuje tudi porušenje objektov starejše nasel­binske faze na območju Narodnega doma.69 Naselbina pozneje ni bila več obnovljena v polnem obsegu, kljub temu pa je, kot pričajo itinerarski viri iz 3. in 4. st., še vedno služila kot cestna postaja. Iz tega časa izvira tudi večji del grobov, odkritih ob cerkvi sv. Jožefa (sl. 2: 1). V poznorimskem obdobju je območje Logaške kotline dobilo novo geostrateško vlogo, povezano z izgradnjo obsežnega obrambnega sistema Claustra Alpium Iuliarum, ki se je razprostiral med Kvarnerskim zalivom na jugu in dolino reke Bače na severu. Osrednji del sistema je bil prav v okolici Logaške kotline, preko katere je potekala državna cesta proti Italiji, ki je bila varovana s tremi zaporednimi linijami obrambnih zapor. Prva linija zapor, ki je ležala pred vstopom v Logaško kotlino s severovzhoda, se je razprostirala med Verdom pri Vrhniki in Marinčevim Gričem nad Strmico. Drugo linijo sta tvorili utrdbi na Lanišču in na Brstu pri Martinj Hribu ob jugozahodnem in jugovzhodnem robu kotline, medtem ko je tretji, najpomembnejši del predstavljala 67 Za učinkovito delovanje rimskega cestnega omrežja je bil namreč poleg rednega vzdrževanja cest nujen tudi obstoj cestnih postaj, ki med seboj niso bile oddaljene več kot en dan potovanja (Šašel Kos 1997, 20). 68 Šašel 1975, 80, 96. 69 Frelih 2003, 15–16. THE HISTORICAL DEVELOPMENT OF LONGATICUM The origin of the Longaticum roadside station was closely connected with the construction of the Roman road between Emona and Aquileia, which parted from the ancient prehistoric route across the Ocra pass in the area of the Logatec basin and led along the shortest line across the Hrušica pass in the direction of Italy.66 This strategically important road section was most probably built in the time of Augustus, as is mentioned by the ancient historians Tacit (second half of the 1st and begin­ning of the 2nd century) and Festus (second half of the 4th century).67 From the period immediately before the construction of the road allegedly originate the remains of an early imperial settlement near Gasilski dom in Dolenji Logatec (Fig. 2: 4), which are attributed to the first Roman settlers. The roadside station with the surrounding settle­ment was most likely located in the immediate vicinity of the crossroads between the itinerary road leading towards Kalce and Hrušica, and the road that branched off south in the direction of Planinsko polje (Stari grad near Unec). During the 1st and the first half of the 2nd centuries, the settlement allegedly spread over an ap­proximately 400 m long area along the main road from Narodni dom in the southwest to the crossroads between the Tržaška and Tovarniška cesta streets in the northeast. At the end of the 2nd century, the northeastern part of the settlement was abandoned and the settlement area pre­sumably narrowed down to the settlement core around Narodni dom (Fig. 2: 2). This change may have been a result of the turbulent times following the incursions of the Marcomanni and Quadi in the second half of the 2nd century. According to Frelih, the consequences of this event are most probably reflected in the destruction of the first settlement phase in the area of Narodni dom.68 After that the settlement was no longer restored to its full extent; however, according to the itinerary sources from the 3rd and 4th centuries, Longaticum retained its function as a road station. From this period also a larger part of graves, discovered in the area near the church of St Jožef (Fig. 2: 1), originates. In the late Roman period, the area of the Logatec basin was given a new geostrategic role, related to the construction of the Claustra Alpium Iuliarum defence system, which extended between the Kvarner bay in the south and the Bača valley in the north. The central part of the system was located in the surroundings of the 66 In addition to the regular maintenance of the roads, the existence of the roadside stations was crucial for the ef­ficient functioning of the Roman road system; the stations should have been no more than one day journey apart (Šašel Kos 1997, 20). 67 Šašel 1975, 80, 96. 68 Frelih 2003, 15–16. trdnjava Ad Pirum, ki je ležala v višinskem zaledju na območju današnje Hrušice.70 Opustitev naselbine je bila najverjetneje povezana z burnimi notranje in zunanjepolitičnimi dogodki ob koncu 4. in v začetku 5. st., v središču katerih se je znašlo območje Logaške kotline zaradi svoje lege ob meji z Ita­lijo. Prav na tem prostoru so potekali srditi državljanski boji za oblast, ki so svoj vrhunec dosegli leta 394 v bitki pri Frigidu, v kateri sta se spopadla uzurpator Evgenij in Teodozij Veliki.71 Nedolgo za tem, v letih 401 in 408, so preko Logaške kotline v Italijo vdrli Zahodni Goti pod Alarikovim vodstvom. Kot pričajo poročila antičnih avtorjev, sistem vojaških zapor čez alpske prehode v tem času ni bil več v funkciji in je najverjetneje izgubil svoj pomen.72 To kaže na postopen razkroj političnih in ekonomskih strukutur rimske države, ki je imel za po­sledico umik prebivalstva rimskih naselbin v odročnejše predele stran od glavnih prometnih poti. 70 Najnovejša dognanja o obrambnem sistemu Claustra Alpium Iuliarum so predstavljena v Kos 2014 in Kusetič et al. 2014. Za kartografski prikaz sistema zapor glej Kusetič et al. 2014, sl. 3.99. 71 Bratož 2014, 160–183. 72 Kos 2014, 40. Logatec basin, where three consecutive lines of fortifica­tions spread, defending the passage along the main road to Italy. The first defence line extended between Verd near Vrhnika and Marinčev grič hill above Strmica, which closed the natural entrance to the Logatec basin from the northeast. The second line comprised the fortifications at Lanišče and at Brst near Martinj Hrib at the south and southwestern edge of the basin, while the third, central part of the system represented the Ad Pirum fort, located in the hilly hinterland in the area of present-day Hrušica.69 The abandonment of Longaticum was most likely connected with the turbulent political events at the end of the 4th and the beginning of the 5th centuries. Some of the most bitter civil fighting for power took place in the surroundings of the Logatec basin due to its location at the border of Italy, culminating Battle of Frigidus in the year 394 between the usurper Eugenius and Theodosius the Great.70 Soon afterwards, in the years 401 and 408, the Western Goths under Alaric passed the Logatec basin on their way to Italy. According to Roman writ­ers, the system of military fortifications of the Alpine crossings was no longer in function at that time and most likely lost its significance.71 This indicates a gradual disintegration of the political and economic structures of the Roman state, which resulted in the withdrawal of the population of Roman settlements into more remote areas away from the main traffic routes. Translation: Gregor Pobežin 69 The latest findings about the Claustra Alpium Iuliarum are represented in Kos 2014 and Kusetič et al. 2014. For a car­tographic representation of the course of the defence system, cf. Kusetič et al. 2014, fig. 3.99. 70 Bratož 2014, 160–183. 71 Kos 2014, 40. BRATOŽ, R. 2014, Med Italijo in Ilirikom. Slovenski prostor in njegovo sosedstvo v pozni antiki. – Ljubljana. FRELIH, M. 1988, Komunikacijski sistem v prazgodovini in antiki na primeru Logaške kotline. – V / In: Mladinski raziskovalni tabor Logatec '88. Zbornik poročil raziskovalnih skupin, 10–51, Logatec. FRELIH, M. 1989, Odkritje rimske cestne postaje “mansio Longatico” v Dolenjem Logatcu. – V / In: Mladinski razis­kovalni tabor Logatec 1989. Zbornik poročil raziskovalnih skupin, 13–32, Logatec. FRELIH, M. 1991, Mansio Longatico – rimska cestna postaja v Logatcu. – Diplomsko delo / BA thesis, Filozofski fakultet Sveučilišta u Zagrebu (neobjavljeno / unpublished). FRELIH, M. 2002, La mansio romana di Longaticum - Logatec (Slovenia). – Quaderni Friulani di Archeologia 12, 77–82. FRELIH, M. 2003, Logatec - Longaticum in rimski obrambni sistem Claustra Alpium Iuliarum. S prispevkom o bitki pri reki Frigidus (Soča) leta 394 (Logatec - Longaticum and the Roman defense system Claustra Alpium Iuliarum. With contribution to the battle at the Frigidus (Soča) River in 394). – Logatec. FRELIH, M., A. MIHEVC 1985, Logatec. – Varstvo spomenikov 27, 228–229. GABROVEC, S. 1999, 50 Jähre Archäologie der älteren Eisen­zeit in Slowenien. – Arheološki vestnik 50, 145–188. GASPARI, A. 2000, Ulaka. Prazgodovinska in rimska naselbina nad Starim trgom pri Ložu. – Magistrsko delo / MA thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). HORVAT, J., A. BAVDEK 2009, Okra / Ocra: Vrata med Sredozemljem in srednjo Evropo / The gateway between the Mediterranean and Central Europe. – Opera Instituti Archaeologici Sloveniae 17. KOLB, A. 1999, s. v. Mansio. – V / In: Der neue Pauly, Enzy­klopädie der Antike, Bd. 7, 829, Stuttgart, Weimar. KOS, P. 2014, Ad Pirum (Hrušica) in Claustra Alpium Iulia-rum. – Vestnik 26. KRAŠNA, M. 2017, Arheologija Logaške kotline. – Diplomsko delo / BA thesis, Oddelek za arheologijo, Filozofska fakul­teta Univerze v Ljubljani (neobjavljeno / unpublished). KUSETIČ, J., P. KOS, A. BREZNIK, M. STOKIN 2014, Claus-tra Alpium Iuliarum. Med raziskovanjem in upravljanjem / Claustra Alpium Iuliarum. Between Research and Mana­gement. – Ljubljana. LAZAR, I. 2003, Rimsko steklo Slovenije / The Roman glass of Slovenia. – Opera Instituti Archaeologici Sloveniae 7. LAZAR, I. 2004, Najdbe rimskega stekla iz Logatca (Longa­ticum). – V / In: Drobci antičnega stekla / Fragments of Ancient Glass, 57-65, Koper. MIHEVC, P. 1992, s. v. Logaško polje. – V / In: Enciklopedija Slovenije 6, 312, Ljubljana. OLIĆ, S. 2006, Logatec – Zapolje. Poročilo o zaščitnem arheo­loškem izkopavanju. – Neobjavljeno poročilo / Unpublished report, Arhos d.o.o. (Hrani / Kept in the Arhiv ZVKDS Ljubljana). PAGON, S. 2008, Geografija občine Logatec. – Diplomsko delo / BA thesis, Oddelek za geografijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). PETRU, P. 1975a, Gorenji Logatec. – V / In: Arheološka najdišča Slovenije, 191, Ljubljana. PETRU, P. 1975b, [Kalce]. – V / In: Arheološka najdišča Slove­nije, 191–192, Ljubljana. PFLAUM, V. 2007, The supposed Late Roman hoard of tools and a steelyard from Vodice near Kalce. – Arheološki vestnik 58, 285–332. REPANŠEK, L. 2016, Keltska dediščina v toponimiji jugovzhod­nega alpskega prostora (Celtic Legacy in the Toponymy of South-Eastern Alps). – Linguistica et philologica 33. ŠAŠEL, J. 1970, s. v. Longaticum. – V / In: Paulys Realencyclo­pädie der classischen Altertumswissenschaft, Suppl. 12, 521, Stuttgart. ŠAŠEL, J. 1975, Rimske ceste v Sloveniji (viae publicae). – V / In: Arheološka najdišča Slovenije, 74–99, Ljubljana. ŠAŠEL KOS, M. 1997, Pošta v antiki (Postwesen im Altertum / Postal service in the Roman period). – V / In: Pošta na slovenskih tleh, 18–43, Ljubljana. ŠAŠEL KOS, M. 2002, The boundary stone between Aquileia and Emona / Mejnik med Akvilejo in Emono. – Arheološki vestnik 53, 373–382. ŠINKOVEC, A. 2018, Končno poročilo o predhodnih arhe­oloških izkopavanjih na najdišču Logatec-križišče 2013 (13-0189). – Neobjavljeno poročilo / Unpublished report, Arhos d.o.o. (Hrani / Kept in the Arhiv ZVKDS Ljubljana). ŠVAJNCER, J. 2004, Logatec v zgodovini. – Logatec. TOPLIČANEC, M. 2007, Dolenji Logatec. – Varstvo spome­nikov 43. Poročila 2006/2007, 46–48. URLEB, M. 1983, Antično grobišče v Cerknici (La nécropole romaine a Cerknica). – Arheološki vestnik 34, 298–346. Ahac Šinkovec ahac.sinkovec@gmail.com Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 93–112 NAUPORTUS - VRHNIKA Jana HORVAT Izvleček Nauportus je ležal ob tovorni poti med Italijo in srednjim Podonavjem ter ob izvirih Ljubljanice, kjer se je začela plovna pot proti vzhodu. Prvotno naselbino keltskih Tavriskov nadzorujejo Rimljani od konca 2. ali začetka 1. st. pr. Kr. Najpozneje sredi 1. st. pr. Kr. ima Navport status vikusa na akvilejskem ozemlju, magistri vici so osvobojenci akvilejskih trgovskih družin. Na desnem bregu Ljubljanice, na Dolgih njivah, je v avgustejskem obdobju stal utrjen skladiščni kom­pleks s pristaniščem, preko katerega sta potekala tranzitni promet in oskrba legij v srednjem Podonavju. Po avgustejskem obdobju je bil kompleks na Dolgih njivah opuščen, hkrati je Navport izgubil vodilno vlogo širšega območja. Nadalje se je razvijalo naselbinsko jedro na Bregu, na drugi strani reke in ob cesti proti Emoni. Strateški pomen prostora pride do izraza ponovno v poznorimski dobi, ko sta bila morda že konec 3. st. postavljena trdnjava na Gradišču in opazovalni stolp na Turnovšču. V 4. st. je bil zgrajen obrambni zid po hribovju zahodno od Navporta, ki je predstavljal del obrambnega sistema Claustra Alpium Iuliarum. Naselbina na Bregu je bila opuščena verjetno hkrati z večino nižinskih naselij v prvi polovici 5. st. Ključne besede: Italija (10. regija), Vrhnika, Nauportus, rimska doba, naselbina, vicus, utrdba, obcestna postaja, skladišča, pristanišče, poznorimska utrdba Abstract Nauportus (modern Vrhnika) was located on the route between Italy and the Middle Danube area, near the springs of the Ljubljanica river, where the waterway towards the east started. The original settlement of the Celtic tribe of the Taurisci was controlled by the Romans from the end of the 2nd century or the beginning of the 1st century BC onwards. In the mid-1st century BC at the latest, Nauportus had the status of vicus in the territory of Aquileia. The magistri vici were freedmen of Aquileian merchant families. In the Augustan period, a fortified storehouse complex with a port for transit traffic and the supply of the legions in the Middle Danube area stood at Dolge njive on the right bank of the Ljubljanica. The Dolge njive complex was abandoned after the Augustan period, which was also the time when Nauportus lost its principal role in the wider area. The settlement at Breg on the other side of the river and along the road towards Emona kept developing. The strategic significance of the area again becomes apparent in the Late Roman period, when the fort at Gradišče and the tower at Turnovšče were built, perhaps as early as the end of the 3rd century. In the 4th century, a defence wall was built on the hills west of Nauportus as part of the Claustra Alpium Iuliarum defence system. The settlement at Breg was probably abandoned at the same time as most of the lowland settlements, i.e. in the first half of the 5th century. Keywords: Italy (Regio X), Vrhnika, Nauportus, Roman period, settlement, vicus, fortification, late Roman fort, road station, storehouses, port https://doi.org/10.3986/9789610502586_05 GEOGRAFSKA LEGA Iz opisov antičnih avtorjev izhaja, da je Navport ležal ob glavni prometni povezavi med Akvilejo in Emono, na vznožju prehodov čez Alpe in ob plovni reki. Ti podatki jasno umeščajo naselje na zahodni rob Ljubljanskega barja in ob izvire Ljubljanice, to je na ob-močje današnje Vrhnike.1 Pravilnost lociranja potrjuje razsežno območje rimskih ostankov v ravnici na obeh bregovih reke. Po Ljubljanici, ki je plovna skoraj od izvirov, se odpira rečna pot proti vzhodu. Proti zahodu pa se vzpne hriboviti svet, preko katerega vodi najugodnejša pot iz srednjega Podonavja in severnega Balkana proti severni Italiji. Vsakokratni naselbini na tem območju je torej dajala osnovni pečat tranzitna pot vzhod–zahod.2 ZGODOVINA RAZISKAV Prva arheološka izkopavanja so bila na Vrhniki v letu 1884. Sledilo jim je več manjših raziskav in dve večji: Walterja Schmida v letih 1934 in 1936 in Ive Mikl Curk 1969–1970.3 V novejšem obdobju so potekala zaščitna izkopavanja leta 1992 na Gradišču 5,4 2005 na Jelovškovi ulici 10–11,5 2007 na priključku za avtocesto,6 2013–2014 na Stari cesti,7 2016 pri Stari pošti,8 2017 v Delavskem naselju9 in 2018 na Gradišču.10 Območje Dolgih njiv je bilo proučeno z geofizikalnimi raziska­vami v letih 2002–2003.11 PISNI VIRI Pisne vire, ki omenjajo naselje Navport in reko z is-tim imenom, je podrobno analizirala Marjeta Šašel Kos.12 Strabon 7, 5, 2: Navport, ki je naselje Tavriskov (v rokopisih ohra­njen zapis v akuzativu – Naúponton; verjetno opis situacije v 2. st. pr. Kr.). 1 V starejši literaturi v rabi tudi nemško ime Oberlaibach. 2 Horvat 1990, 137–140, 240–243. Pregled raziskav do leta 1985: Horvat 1990, 37–38, 162–163. 4 Bavec, Horvat 1996. 5 Horvat, Peterle Udovič 2006; Horvat, Mušič 2007 (naj­ prej poimenovano Kočevarjev vrt).6 Horvat et al. 2016. 7 Janežič, Mulh, Černe 2017; Mulh, Černe 2018. 8 Vojaković et al. 2019. 9 Žerjal, Bekljanov Zidanšek 2018. 10 Podatki Tina Žerjal. 11 Mušič, Horvat 2007. 12 Šašel Kos 1990, 17–21, 143–148. GEOGRAPHICAL POSITION According to descriptions by Roman period authors, Nauportus was situated on the main route between Aq­uileia and Emona, on a navigable river and at the foot of Alpine passes. On the basis of these data, the settlement can be located without doubt at the western edge of the Ljubljansko barje basin and near the springs of the Ljubljanica river, i.e. the area of the present-day town of Vrhnika.1 The correctness of the location is further confirmed by an extensive area of Roman remains in the plain on both riverbanks. Navigable almost to its sources, the Ljubljanica is the beginning of the waterway to the east, while the most advantageous route from the Middle Danube area and the northern Balkans towards north Italy leads over the mountainous terrain to the west. In any period, the settlement in this location was thus characterized by the east-west transit route.2 RESEARCH HISTORY The first archaeological excavations in Vrhnika took place in 1884. They were followed by several minor and two major investigations: the excavations led by Walter Schmid in the years 1934 and 1936, and by Iva Mikl Curk in 1969–1970.3 In the more recent period, rescue excavations took place in 1992 at Gradišče 5,4 in 2005 at Jelovškova ulica 10–11,5 in 2007 at the highway exit-road,6 in 2013–2014 at Stara cesta,7 in 2016 at Stara pošta,8 in 2017 at Delavsko naselje,9 and in 2018 at Gradišče.10 The area of Dolge njive was geophysically surveyed in 2002–2003.11 WRITTEN SOURCES The written sources that mention the settlement Nauportus and the river of the same name were studiedin detail by Marjeta Šašel Kos.12 1 The German name Oberlaibach appears in older literature. 2 Horvat 1990, 137–140, 240–243. 3 An overview of research up to 1985: Horvat 1990, 37– 38, 162–163. 4 Bavec, Horvat 1996. 5 Horvat, Peterle Udovič 2006; Horvat, Mušič 2007; the site was earlier called Kočevarjev vrt.6 Horvat et al. 2016. 7 Janežič, Mulh, Černe 2017; Mulh, Černe 2018. 8 Vojaković et al. 2019. 9 Žerjal, Bekljanov Zidanšek 2018. 10 Information by Tina Žerjal. 11 Mušič, Horvat 2007. 12 Šašel Kos 1990, 17–21, 143–148. Strabon 4, 6, 10: naselje Navport (v rokopisih ohranjen zapis v akuzativu – Pámporton; verjetno opis situacije v prvi polovici 1. st. pr. Kr.). Velej Paterkul 2, 110, 4: naselje Navport (čas panonsko-delmatskega upora 6–9 po Kr.). Tacit, Annales 1, 20,1: vikus Navport, ki je bil kot municipij (čas vojaškega upora leta 14 po Kr.). Tabula Peutingeriana IV, 1: cestna postaja Navport med Longatikom in Emono. Plinij Starejši, Naturalis Historia 3, 128: reka Navport, povezana z legendo o Argonavtih. Navport se je imenovala reka Ljubljanica, ki je torej nosila enako ime kot naselje ob njej.13 PREGLEDNE OBJAVE Splošni pregledi: Horvat 1996; 2009; 2012a; Horvat, Mušič 2007; Gaspari 2017, 127–144. Arheološka topografija in drobne najdbe: Horvat 1990. Epigrafski viri: EDR128824, 128825, 156071, 156077, 156080, 156082, 156084, 156085. Komentarji: Šašel Kos 1990, 21–30, 148–156. Dopol­nitve: Šašel Kos 1997, 117–122; Šašel Kos 1998; Grassl 2017. Novci: Kos 1977, 150–153; FMRSl I, 375–378, št. 206/1–3; FMRSl III, 246–256, št. 108–110; FMRSl IV, 247–250, št. 111–115; FMRSl V, 309, št. 83; FMRSl VI, 163–164, št. 108–109; Horvat 1990, 86–96, 194–204. PRAZGODOVINA Na razglednem hribu Tičnica, približno kilometer zahodno od Ljubljanice, leži velika, z okopi obdana na­selbina (sl. 1), ki je živela od mlajše bronaste do začetka starejše železne dobe.14 Predmeti iz konca srednje in začetka mlajše bronaste dobe so bili odkriti v strugi Ljubije pri Verdu (Bd C2 do Ha A1)15 in v Ljubljanici na Vrhniki.16 Na severnem območju današnje Vrhnike je bil 13 Šašel Kos 1990, 19–20, 145–147; Šašel Kos 2017. 14 Gaspari, Masaryk 2009; Gaspari, Mlekuž 2013; Gaspa­ ri, Vinazza 2018. 15 Gaspari 2006. 16 Gaspari, Erič 2008, 410–411; Gaspari, Masaryk 2009, 196–197. Strabo 7, 5, 2: Nauportus, which is a settlement of the Taurisci (the name in the manuscripts is in the accusative case – Naúponton; probably a description of the situation in the 2nd century BC). Strabo 4, 6, 10: the settlement of Nauportus (the name in the manuscripts is in the accusative case – Pámporton; probably a description of the situation in the first half of the 1st century BC). Velleius Paterculus 2, 110, 4: the settlement of Nauportus (the time of the Pan-nonian and Dalmatian uprising, AD 6–9). Tacitus, Annales 1, 20,1: the vicus of Nauportus, which was like a municipium (the time of the military revolt of AD 14). Tabula Peutingeriana IV, 1: the road station of Nauportus between Longaticum and Emona. Pliny the Elder, Naturalis Historia 3, 128: the river of Nauportus, associated with the legend of the Argonauts. Nauportus was the name of the present-day Ljubljanica river, which therefore shared its name with the settlement on it.13 OVERVIEW PUBLICATIONS General overviews: Horvat 1996; 2009; 2012a; Horvat, Mušič 2007; Gaspari 2017, 127–144. Archaeological topography and small finds: Horvat 1990. Epigraphic sources: EDR128824, 128825, 156071, 156077, 156080, 156082, 156084, 156085. Commentaries: Šašel Kos 1990, 21–30, 148–156. Supple­ments: Šašel Kos 1997, 117–122; Šašel Kos 1998; Grassl 2017. Coins: Kos 1977, 150–153; FMRSL I, 375–378, no. 206/1–3; FMRSL III, 246–256, no. 108–110; FMRSL IV, 247– 250, no. 111–115; FMRSL V, 309, no. 83; FMRSL VI, 163–164, no. 108–109; Horvat 1990, 86–96, 194–204. PREHISTORY Situated on the Tičnica hill, which offers a good view of the surroundings, and approximately 1 km west of the Ljubljanica, lies a large settlement, surrounded by ramparts (Fig. 1), dated from the Late Bronze Age to the beginning of the Early Iron Age.14 Artefacts from the 13 Šašel Kos 1990, 19–20, 145–147; Šašel Kos 2017. 14 Gaspari, Masaryk 2009; Gaspari, Mlekuž 2013; Gaspari, Vinazza 2018. prazgodovinska naselbina / Prehistoric settlement rimska naselbina / Roman settlement rimsko grobiš~e / Roman cemetery rimska cesta / Roman road domnevna rimska cesta / presumed Roman road plovilo / vessel 0 500 m Sl. 1: Vrhnika. Fig. 1: Vrhnika. 1 Stara pošta; 2 Dolge njive; 3 Ljubljanica: deblak in ladja / logboat and ship; 4 Breg, Jelovškova ulica 10–11; 5 Breg, priključek na avtocesto / highway exit; 6 Breg, Delavsko naselje; 7 Gradišče 5; 8 Gradišče; 9 Turnovšče; 10 Stara cesta; 11 Ljubljanica: ladja /ship; 12 Mokrice najden grob ali zaklad z mečem, sekiro, iglo in keramič­nim loncem, ki sodi v srednjo bronasto dobo (Bd C).17 Raziskave na desnem bregu Ljubljanice (Dolge nji­ve) so razkrile sledove stalnih dejavnosti, kot so sekanje dreves in obdelava lesa, ki so potekale v obdobju med 4./3. st. in 1. st. pr. Kr. Niso pa ti ostanki znak naselbine v neposredni bližini (sl. 1: 2).18 17 Horvat 1990, 67–68, 180; Turk 2007, 215–216, 226–227. 18 Horvat et al. 2016, 221–223, 252–255. end of the Middle and the beginning of the Late Bronze Age were discovered in the riverbeds of the Ljubija near Verd (BA C2 to Ha A1)15 and the Ljubljanica in Vrhni­ka.16 In the northern part of the present-day Vrhnika, a grave or a hoard from the Middle Bronze Age (BA C) was discovered, containing a sword, an axe, a pin, and a ceramic pot.17 Investigations on the right bank of the Ljubljanica (site Dolge njive; Fig. 1: 2) revealed traces of permanent activities such as cutting trees and woodworking from the period between the 4th/3rd and the 1st century BC. These remains, however, are not an indicator of a settlement in the immediate vicinity.18 15 Gaspari 2006. 16 Gaspari, Erič 2008, 410–411; Gaspari, Masaryk 2009, 196–197. 17 Horvat 1990, 67–68, 180; Turk 2007, 215–216, 226–227. 18 Horvat et al. 2016, 221–223, 252–255. ARHEOLOŠKI SLEDOVI RIMSKE DOBE POZNOREPUBLIKANSKO OZ. POZNOLATENSKO OBDOBJE Na levem bregu Ljubljanice, pri Stari pošti (sl. 1: 1), so bili odkriti kosi obdelanega lesa (tram in pokončno zabit kol), ki so verjetno sled prvega urejanja rečnega brega v 2. st. pr. Kr. Višje je ležala naselbinska plast, v katero je bil vkopan lesen zabojnik (domnevno zbiralnik za vodo). Pozneje je bilo območje prekrito s tlakom iz drobnega kamenja. Glavnino najdb iz obeh zgornjih plasti predstavlja keramika, uvožena iz Italije, dobro datirana v konec 2. st. oziroma v prvo tretjino 1. st. pr. Kr. (fina namizna keramika, navadna in kuhinjska ke­ramika, oljenke, amfore). Latenske keramike je izredno malo, le nekaj koščkov grafitnih loncev.19 Kaže, da je pri Stari pošti ležalo najstarejše naselje prišlekov iz Italije, na katerega se verjetno tudi nanaša omemba pri Strabonu.20 Posamične kovinske najdbe (fibule, poznolatenski meč) in latenska keramika nakazujejo, da je morda so­časno z naselbino pri Stari pošti obstajalo tudi že prvo poselitveno jedro na desnem bregu Ljubljanice, v rečnem okljuku na Dolgih njivah (sl. 1: 2).21 Na južnem delu Dolgih njiv je bila ob Ljubljanici odkrita hodna površina s sledmi človekove dejavnosti, ki so jo predstavljali zavrženi ostanki obdelave lesa. Dejavnost je bila datirana v čas med koncem 2. st. in sredino 1. st. pr. Kr.22 V strugi Ljubljanice (sl. 1: 3) med Staro pošto in Dolgimi njivami je bil odkrit deblak iz konca 2. st. pr. Kr. Dolg je bil okoli 14,4 m in je bil verjetno namenjen za prevoz tovora. Poleg deblaka so ležali ostanki šivane ladje, datirane v 2. st. pr. Kr.23 V reki so bili odkriti tudi posamezni latenskodobni predmeti.24 Pelodna analiza kaže na postopen umik gozda v širši okolici Vrhnike v 2. in 1. st. pr. Kr.25 DOLGE NJIVE V AVGUSTEJSKEM OBDOBJU Naselbino na Dolgih njivah (sl. 1: 2) so raziskovali večkrat,26 večja in pomembnejša izkopavanja pa staizvedla Walter Šmid v letih 1934 in 193627 ter Iva Mikl 19 Vojaković et al. 2019. 20 Strabon 4, 6, 10. 21 Horvat et al. 2016, 223, 254. 22 Horvat et al. 2016, 222–223, 253–255 (faza 1D). 23 Gaspari 2017. 24 Gaspari, Masaryk 2009, 196–198. 25 Andrič 2016, 269, 274–275. 26 Horvat 1990, 49–57, 97–132, 171–173, 205–235. 27 Schmid 1943, 9–13: Horvat 1990, 50–51, 97–109, 172, 207–211. ARCHAEOLOGY OF THE ROMAN PERIOD LATE REPUBLICAN OR LATE LA TENE PERIOD On the left bank of the Ljubljanica, near Stara pošta (Fig. 1: 1), fragments of worked wood (a beam and a verti­cal post) were discovered; probably a trace of the earliest riverbank engineering in the 2nd century BC. Above this there was a settlement layer, into which a wooden contain­er (supposedly a water tank) was dug. The area was later covered with a pavement of small stones. The majority of the finds from the top two layers are Italian-imported pottery, which can be reliably dated to the end of the 2nd / first third of the 1st century BC (fine tableware, coarse pottery and kitchenware, oil lamps, amphorae). La Tene pottery is very scarce and includes just a few fragments of graphite pots.19 It seems that the earliest settlement of the newcomers from Italy was at Stara pošta, and this is probably the settlement referred to by Strabo.20 Contemporary to the Stara pošta settlement, stray metal finds (fibulae, Late La Tene sword) and La Tene pottery indicate the possible existence of another settle­ment area on the right bank of the Ljubljanica, in the river meander at Dolge njive (Fig. 1: 2).21 In the southern part of Dolge njive, near the Lju­bljanica, a walking surface with traces of human activi­ties – woodworking waste – was discovered. The activity dates to the time between the end of the 2nd century and the middle of the 1st century BC. 22 A logboat from the end of the 2nd century BC was discovered in the riverbed of the Ljubljanica between Stara pošta and Dolge njive (Fig. 1: 3). Approximately 14.4 m long, it was probably intended for freight trans­port. Remains of a sewn boat, dated to the 2nd century BC, were discovered next to the logboat.23 In addition to that, stray finds from the La Tene period were discovered in the river.24 Pollen analysis indicates a gradual removal of the forest in the wider surroundings of Vrhnika in the 2nd and 1st centuries BC.25 DOLGE NJIVE IN THE AUGUSTAN PERIOD While several investigations26 were conducted at the Dolge njive settlement (Fig. 1: 2), the largest and mostsignificant excavations were performed by Walter Šmid 19 Vojaković et al. 2019. 20 Strabo 4, 6, 10. 21 Horvat et al. 2016, 223, 254. 22 Horvat et al. 2016, 222–223, 253–255 (Phase 1D). 23 Gaspari 2017. 24 Gaspari, Masaryk 2009, 196–198. 25 Andrič 2016, 269, 274–275. 26 Horvat 1990, 49–57, 97–132, 171–173, 205–235. obrambni jarek / defence ditch lesen pomol / wooden pier obmo˛je raziskovanja 2007 / research area 2007 Sl. 2: Vrhnika. Naselbina na Dolgih njivah. Avgustejsko obdobje. Fig. 2: Vrhnika. The Dolge njive settlement. Augustan period. Curk 1969.28 Z geofizikalnim pregledom je bil pridobljen in 1934 and 193627 and by Iva Mikl Curk in 1969.28 A tloris celotne naselbine (sl. 2).29 geophysical survey provided a plan of the entire settle- Drobno arheološko gradivo z vseh območij na-ment (Fig. 2).29 selbine (skladišča, obzidje, tlak pristanišča) kaže, da Small archaeological finds from all parts of the je naselbina na Dolgih njivah živela od zgodnjega do settlement (storehouses, walls, the pavement of the port) poznoavgustejskega obdobja, potem pa je bila opušče­ 27 Schmid 1943, 9–13: Horvat 1990, 50–51, 97–109, 172, 207–211. 28 Mikl Curk 1974; Horvat 1990, 97–99, 205. 28 Mikl Curk 1974; Horvat 1990, 97–99, 205. 29 Mušič, Horvat 2007; Horvat 2008. 29 Mušič, Horvat 2007; Horvat 2008. na.30 Obstaja možnost, da ležijo na območju okljuka še starejše faze naselbine (glej zgoraj). Naselje je imelo obliko nepravilnega pravokotnika (130 × 144 m). Z dveh strani ga je oblivala Ljubljanica, na vzhodni in verjetno tudi na južni strani pa obrambni jarek, napolnjen z vodo (širok okoli 7 m in globok 3,5 m). Na južni strani je bilo obzidje široko 2 m, na vzhodni 3 m. Fronti sta bili grajeni iz kamnov lomljencev, vmesni prostor je bil zapolnjen z ilovico in drobnejšimi kamni, povezovali so ju prečni zidci in vodoravna lesena bruna. Prostor med obzidjem in obrambnim jarkom je bil delo-ma tlakovan. Vzdolž reke je bilo obzidje široko približno1 m in nanj so bile prizidane stavbe. Štirje kvadratni stolpi v obzidju so ščitili vogale in dva vhoda v naselbino. Severni vhod je pokrival stolp, skozi katerega je potekala močno tlakovana pot. Vzhodni vhod sta ščitila na eni strani kvadratni stolp, na drugi pa okrepljen del obzidja. Sredi naselja je ležal trg, tlakovan s kamni in obdan s stebriščem. Zajemal je skoraj tretjino površine nasel­bine (okoli 5.500 m2). Okoli trga so stala skladišča (sl. 2: 1–5,11–20,22– 23). Sestavljali so jih dolgi in ozki prostori (dolžina od 20 do 26 m, širina 6 m), ki so bili grajeni posamič, v paru ali v nizu štirih. Med skladišči so bili ozki prehodi, široki do 0,5 m (ambitus). Posebnost Navporta so široko odprti vhodi skladišč, ki so drugače značilni predvsem za taberne (trgovine). Nekateri prostori imajo široko odprte vhode z dveh strani (sl. 2: 20,22), nekateri so bili pozneje predeljeni. Ob južnem obzidju je bila odkrita stavba z vrsto manjših prostorov, ki po obliki predstavlja prave taberne (sl. 2: 21). Skupna površina skladiščnih in 31 trgovskih prostorov je bila okoli 6.400 m2. Po ostankih prežganega glinastega ometa je Schmid sklepal, da so bile nekatere stene lesene in ometane z ilovico. Kritina je bila opečna. Pod enim izmed temeljev na severni strani so bili odkriti leseni piloti.32 Sredi trga sta stali dve stavbi. Ena (sl. 2: 25) je imela pravokotno ploščad, zidano v dveh nivojih, obdajal jo je plitvo temeljen obodni zid. Stavbo razlagamo kot sve­tišče, z dvignjenim podijem in stopniščem na vzhodu. Funkcija druge stavbe sredi trga ni jasna (sl. 2: 24).33 Namembnosti manjših stavb v severozahodnem vogalu naselbine ne moremo razbrati zgolj iz tlorisov. Geofizikalne raziskave nakazujejo možnost več grad-benih faz. Na območju stavbe 8 (sl. 2: 8) je bil najden barvan omet in arhitekturni okras iz kamna: akroterij, oblikovan kot preplet akantovih listov in cvetov, ter tanke plošče, okrašene z dvojnim S, ki so bile verjetno 30 Horvat 1990; Mušič, Horvat 2007, 254–261, 278–279; Horvat et al. 2016, 223–225, 255–257. 31 Mušič, Horvat 2007. 32 Horvat 1990, 101–108, 207–211. 33 Mušič, Horvat 2007, 249–250, 264, 278, 281. indicate that the Dolge njive settlement existed from the Early to the Late Augustan periods, after which it was abandoned.30 It is possible, however, that even earlier settlement phases are present in the area of the river meander (see above). The settlement was in the form of an irregular rectangle (130 x 144 m). It was surrounded by the Lju­bljanica on two sides, while on the east and probably also on the south there was a water-filled moat (ca. 7 m wide and 3.5 m deep). The walls were 2 m wide on the south and 3 m on the east. The space between the faces of the walls was filled with clay and small stones, while the faces themselves were made of roughly worked stones and bound together with narrow transverse walls and horizontal wooden beams. The area between the wall and the moat was partly paved. The wall along the river was approximately 1 m wide with buildings leaning on it. Four square towers in the wall guarded the corners and two entrances to the settlement. A paved path led through the tower above the northern entrance. The eastern entrance was guarded by a square tower on one side and a fortified part of the wall on the other. In the centre of the settlement lay a square, paved with stones and enclosed with a colonnade. It extended over almost a third of the settlement surface (about 5,500 m2). The square was surrounded by storehouses (Fig. 2: 1–5,11–20,22–23). These buildings had long and narrow rooms (20-26 m in length, 6 m in width) and could be constructed as individual rooms, in a pair, or in a row of four. The storehouses were separated by narrow (up to 0.5 m wide) passages (ambitus). A special feature of the Nauportus storehouses are wide entrances, which are otherwise more typical of tabernae (shops). While some rooms had wide-open entrances on both sides (Fig. 2: 20,22), others were later divided into several rooms. A building with a series of small rooms with the form of the real tabernae was discovered near the south wall (Fig. 2: 21). The total surface of both storage and commercial 31 premises was about 6,400 m2. Remains of burnt clay wall plaster led Schmid to the conclusion that some walls were made of wood and covered with clay plaster. The roof was covered with tiles. Wooden pilots were discovered under a foundation on the north side.32 Two buildings stood in the middle of the square, one of them (Fig. 2: 25) with a rectangular platform in two levels, surrounded by a wall with shallow foundations. The building is interpreted as a sanctuary with a raised podium and a staircase on the east side. The function of 30 Horvat 1990; Mušič, Horvat 2007, 254–261, 278–279; Horvat et al. 2016, 223–225, 255–257. 31 Mušič, Horvat 2007. 32 Horvat 1990, 101–108, 207–211. okras strešnega roba.34 Schmid je stavbo interpretiral kot svetišče lokalne boginje Ekorne.35 Med naključnimi najdbami z Dolgih njiv, ki niso podrobneje locirane, so še kapitel z napisom, različni od­lomki kamnitega arhitektonskega okrasa, kosi poslikane malte ter črni in beli mozaični kamenčki.36 Pristanišče je ležalo vzdolž naselbine, na desnem bregu Ljubljanice. Obrežni pas je bil prekrit z velikimi kamni lomljenci, čez pa je bila nasuta debela plast peska. Prvotni tlak je bil še dvakrat popravljen. Gre za pristajal-no obalo, dolgo vsaj 270 m.37 Breg je bil dodatno utrjen s posameznimi lesenimi koli. K pristanišču sodi verjetno tudi lesena konstrukcija, ki je stala v reki pred severnim vhodom v naselbino. Pravokotno območje, veliko 6 × 4,5 m, je bilo gosto pilotirano in je predstavljalo verjetno temelj masivnega objekta, morda lesenega pomola.38 Zaklad enega velikega in 23 malih keltskih srebr­nikov je bil verjetno odkrit v prehodu med stavbama 4 in 5.39 Iz skladiščnega prostora (4a) domnevno izvira zaklad svinčenih želodov, približno 600 kosov.40 Na Dolgih njivah so bile najdene vse zvrsti uvožene rimske keramike, od fine namizne do kuhinjske. Pojavlja se tudi groba lokalna lončenina. Nekaj je latenske fine keramike, za katero ni jasno, ali sodi v avgustejsko obdobje ali pa je pokazatelj starejše faze na Dolgih njivah.41 Rimske najdbe iz reke ob Dolgih njivah so številne in med njimi posebej izstopajo orožje in večje število kovinskih posod.42 Na dveh napisnih kamnih, katerih prvotna lega ni znana in sta datirana v prvo polovico ali sredino 1. st. pr. Kr., sta omenjena svetišče boginje Ekorne in portik (sl. 7).43 Glede na močno naselitveno jedro na Dolgih njivah, ki je vključevalo portik in verjetno vsaj eno svetišče, bi lahko iskali te objekte prav tu. 34 Horvat 1990, 102–105, 112–113, 208, 215–216, t. 6: 1–3,5 (stavba V).35 Schmid 1943, 11–12. 36 Horvat 1990, 22, 50, 101, 149, 172, 207. 37 Horvat et al. 2016, 223–225, 255–257. 38 Logar 1986; Horvat 1990, 100–101; Horvat 2012a, 290; Horvat et al. 2016, 224–225, 256. 39 Horvat 1990, 89–90, 106, 197–198, 209. 40 Horvat 1990, 106, 209. 41 Horvat 1990, 115–132, 218–234; Mušič, Horvat 2007, 257–261, 278–279. 42 Horvat 1990, 58–61, 133–135, 173–175, 236–239. Za nekatere kose orožja in orodja (Horvat 1990, t. 27–29) je bilo pozneje ugotovljeno, da izvirajo iz Ljubljanice pri Bev­kah: Istenič 2003; Bras Kernel 2006. 43 EDR128824, 156071; Šašel Kos 1990, 22–23, 25–27, 149, 152–155; Šašel Kos 1998. the other building in the middle of the square is unclear (Fig. 2: 24).33 The purpose of the small buildings in the northwest corner of the settlement cannot be discerned on the basis of their plans alone. Geophysical surveys indicate the possibility of several phases of construction. In the area of Building 8 (Fig. 2: 8), painted plaster was discovered, as well as architectural ornaments made of stone: an acroterion with intertwined acanthus leaves and flowers, and thin slabs decorated with a double S, which probably served as roof ornaments.34 Schmid interpreted the build­ing as the sanctuary of the local deity Aecorna.35 The stray finds from Dolge njive that are not pre­cisely located include a capital with an inscription, vari­ous stone fragments of architectural ornaments, pieces of painted mortar, and black and white mosaic tiles.36 The port was located on the right bank of the Lju­bljanica along the length of the settlement. The riparian strip was paved with large, roughly worked flagstones, covered with a thick layer of sand. The original pavement was repaired twice. This was the landing shore and it was at least 270 m long.37 The bank was reinforced with separate wooden posts. A wooden construction in the river in front of the northern entrance to the settlement probably belonged to the port as well. The rectangular area measured 6 x 4.5 m, it stood on densely placed pilots, and was probably the foundation of a massive building, perhaps a wooden pier.38 A hoard consisting of one large and 23 small Celtic silver coins was probably discovered in the passage be­tween Buildings 4 and 5.39 A hoard of ca. 600 lead sling shots supposedly came from Storing room 4a.40 All types of imported Roman pottery, from fine tableware to kitchenware, were discovered at Dolge njive. Local kitchenware also occurs. There is some La Tene fine pot­tery, but it is unclear whether it belongs to the Augustan period or whether it is an indicator of an earlier phase at Dolge njive.41 Especially prominent among the numerous Roman finds from the river at Dolge njive are weapons and a considerable number of metal vessels.42 33 Mušič, Horvat 2007, 249–250, 264, 278, 281. 34 Horvat 1990, 102–105, 112–113, 208, 215–216, Pl. 6: 1–3, 5 (Building V). 35 Schmid 1943, 11–12. 36 Horvat 1990, 22, 50, 101, 149, 172, 207. 37 Horvat et al. 2016, 223–225, 255–257. 38 Logar 1986; Horvat 1990, 100–101; Horvat 2012a, 290; Horvat et al. 2016, 224–225, 256. 39 Horvat 1990, 89–90, 106, 197–198, 209. 40 Horvat 1990, 106, 209. 41 Horvat 1990, 115–132, 218–234; Mušič, Horvat 2007, 257–261, 278–279. 42 Horvat 1990, 58–61, 133–135, 173–175, 236–239. It has been established for several weapons and tools (Horvat 1990, t. 27–29) that they originate from the Ljubljanica near Bevke: Istenič 2003; Bras Kernel 2006.x LEVI BREG LJUBLJANICE MED 1. IN 4. ST. PO KR.: BREG IN GRADIŠČE Cesta Akvileja–Emona naj bi bila po pisnem viru zgrajena v poznoavgustejskem obdobju.44 Trasa je bila na območju Navporta ugotovljena na več mestih. Potekala je po levem bregu Ljubljanice.45 Ob cesti se je konec 1. st. pr. Kr. razvilo naselbinsko jedro na Bregu, ki je obstajalo do pozne rimske dobe. Sle­dovi poselitve pokrivajo prostor ob Ljubljanici, ki meri okoli 600 m v dolžino od jugozahoda do severovzhoda, v širino pa okoli 200 m (sl. 1).46 Podrobnejše raziskave so potekale na treh območjih: Jelovškova ulica 10–11 (sl. 1: 4), priključek na avtocesto (sl. 1: 5) in Delavsko naselje (sl. 1: 6).47 Na območju Jelovškove ulice 10–11 (prej Ljubljanska cesta 9, Kočevarjev vrt; sl. 1: 4), verjetno tik ob rimski cesti, so bile v srednje- ali poznoavgustejskem obdobju postavljene lesene stavbe, grajene s pokončnimi stojkami (sl. 3). Med njimi so ležali jarki z lesenimi palisadami, napolnjeni z vodo. Istočasno so delovali trije vodnjaki, ki so bili zgrajeni s pomočjo lesenih sodov (dendrokrono­loške datacije sodov po letu 3 in 10 po Kr.).48 V obdobje prve naselbine sodi svinčena ploščica z omembo Arija iz Navporta (sl. 8). Polizdelek bronaste fibule in odrezki bronaste pločevine so sledi kovinarske proizvodnje. Po sredini 1. st. so bile lesene stavbe podrte in jarki zasuti. V drugi polovici 1. st. po Kr. sta bili na poravna­nem prostoru zgrajeni dve stavbi s kamnitimi temelji (širina 14 m oziroma 15 m, dolžina pa več kot 30 m), brez predelnih sten (sl. 4). Morda sta imeli skladiščno funkcijo. V 2. st. sta bili stavbi verjetno porušeni in na istem prostoru je bila postavljena velika stavba s kamni­timi temelji (33 × več kot 41 m) in kamnitim notranjim tlakom. Vrste stebrov so nosile streho.49 Na območju Delavskega naselja so odkrili podobne ostanke kot na Jelovškovi ulici 10–11: stavbe, grajene s kamnitimi temelji ali lesenimi sohami, tlake in odvodne jarke. Ugotovljene so bile tri glavne naselbinske faze, datirane od začetka 1. do 4. st. po Kr.50 Na območju priključka na avtocesto (sl. 1: 5) je bila odkrita dolga ozka stavba, datirana po drobnih najdbah v avgustejsko obdobje. Morda gre za podoben skladiščni prostor, kot so stali na nasprotnem bregu, na Dolgih nji­vah.51 Izkopavanja 2007 so tik ob Ljubljanici odkrila rob naselbine z lesenimi objekti in odvodnimi jarki ter ure­ditev obrežja z leseno palisado in več peščenimi nasutji. 44 Festus, Breviarium, 7; Šašel 1975, 80. 45 Horvat 1990, 40–45, 165–168, sl. 3. 46 Horvat 1990, 61–65, 175–179, pril. 2. 47 Žerjal, Bekljanov Zidanšek 2018; Bekljanov Zidanšek, Žerjal 2018. 48 Čufar et al. 2019. 49 Horvat, Mušič 2007, 167–172; Horvat 2012b. 50 Bekljanov Zidanšek, Žerjal 2018. 51 Mikl Curk 1974, 376–378; Horvat 1990, 64–65, 178–179. Two inscription stones from the first half or the middle of the 1st century BC, whose original positions are unknown, mention the sanctuary of Aecorna and a portico (Fig. 7).43 Considering the fact that there was an important settlement area at Dolge njive, which included a portico and probably at least one sanctuary, this could be the location of the two buildings. THE LEFT BANK OF THE LJUBLJANICA BETWEEN THE 1ST AND 4TH CENTURIES AD: BREG AND GRADIŠČE According to a written source, the Aquileia–Emona road was built in the Late Augustan period.44 Its course has been identified in several locations in the area of Nauportus. It ran along the left bank of the Ljubljanica.45 Near the road, a settlement developed at Breg at the end of the 1st century BC and it existed until the Late Ro­man period. Traces of occupation cover an approximately 600 m long and 200 m wide area along the Ljubljanica, oriented southwest-northeast (Fig. 1).46 More detailed in­vestigations were carried out in three locations: Jelovškova ulica 10–11 (Fig. 1: 4), the highway exit-road (Fig. 1: 5), and Delavsko naselje (Fig. 1: 6).47 In the area of Jelovškova ulica 10–11 (earlier: Lju­bljanska cesta 9, Kočevarjev vrt; Fig. 1: 4), wooden build­ings with vertical posts were constructed in the Middle or Late Augustan period (Fig. 3), probably in the immediate vicinity of the Roman road. Between them, there were water-filled ditches with wooden palisades. Three wells made with wooden barrels were in use in the same time (dendrochrolological terminus post quem dates: AD 3 and AD 10).48 A lead tablet mentioning an Arius from Nauportus belongs to the first period of the settlement (Fig. 8). There are traces of metalworking: a semi-finished bronze fibula and scraps of sheet bronze. After the middle of the 1st century AD, the wooden buildings were torn down and the ditches filled. In the second half of the 1st century AD, two build­ings with stone foundations (14 or 15 m in width, more than 30 m in length) and without partitions were con­structed in the levelled area (Fig. 4). They might have functioned as storehouses. In the 2nd century, the build­ings were probably demolished and a large construction with stone foundations (33 x more than 41 m) and an 43 EDR128824, 156071; Šašel Kos 1990, 22–23, 25–27, 149, 152–155; Šašel Kos 1998. 44 Festus, Breviarium, 7; Šašel 1975, 80. 45 Horvat 1990, 40–45, 165–168, Fig. 3. 46 Horvat 1990, 61–65, 175–179, App. 2. 47 Žerjal, Bekljanov Zidanšek 2018; Bekljanov Zidanšek, Žerjal 2018. 48 Čufar et al. 2019. V1 V3 V2 jarek / ditch V1-V3 vodnjak / well stojka / pit 01 10 m les / wood Sl. 3: Vrhnika, Jelovškova ulica 10–11. Lesene stavbe, odvodni jarki in vodnjaki iz prve polovice 1. st. po Kr. M. = 1:400. Fig. 3: Vrhnika. Jelovškova ulica 10–11 Wooden buildings, drainage ditches, and wells, first half of the 1st century AD. Scale = 1:400. Najstarejša poselitvena faza sodi v poznoavgustejsko dobo. V drugi polovici 1. st. so v bližini porušili stavbo, iz katere izvirajo ostanki maltnih tlakov in barvanega stenskega ometa. Od 2. st. dalje na raziskanem območju ni bilo več arheoloških sledov.52 Na severnem območju naselbine na Bregu so mo­rale stati bolj kakovostne stavbe, saj so bili naključno najdeni mozaični kamenčki in barvan omet.53 V 1. in 2. st. je na Gradišču (sl. 1: 8; 6) ležalo ločeno naselbinsko jedro (okoli 1200 m oddaljeno od naselbine na Bregu). Prva zidana stavba, tlakovanje in izravnave na 52 Horvat, Peterle Udovič, Žerjal 2007; Žerjal, Peterle Udovič 2008. 53 Horvat 1990, 61–63, 175–178: na območju ni bilo modernih raziskav. interior stone pavement was built in the same location. The roof was supported by rows of pillars.49 Settlement remains similar to those in Jelovškova ulica 10–11 were discovered in the site of Delavsko na­selje: buildings constructed with wooden posts or stone foundations, pavements and drainage ditches. Three main settlement phases are dated between the beginning of the 1st and the 4th century AD.50 In the area of the highway exit­road (Fig. 1: 5), a long narrow building was discovered. It is dated to the Augus­tan period on the basis of small finds. The building might have been a storehouse, not unlike the ones at Dolge njive on the opposite riverbank.51 In 2007, excavations right 49 Horvat, Mušič 2007, 167–172; Horvat 2012b. 50 Bekljanov Zidanšek, Žerjal 2018. 51 Mikl Curk 1974, 376–378; Horvat 1990, 64–65, 178–179. 102 1. st. / 1st cent. 2. st. / 2nd cent. 3.-4. st. / 3rd - 4th cent. temelj stebra, 2. st. column base, 2nd cent. 0 1 10 m Sl. 4: Vrhnika. Vrhnika, Jelovškova ulica 10–11. Stavbe s kamnitimi temelji, 1.–4. st. po Kr. M. = 1:400. Fig. 4: Vrhnika. Jelovškova ul. 10–11 Stone foundations, 1st–4th cent. Scale = 1:400. along the Ljubljanica unearthed the edge of a settlement with wooden buildings and drainage ditches. A wooden palisade and several layers of sand served as a reinforce­ment of the riverbank. The earliest settlement phase can be attributed to the Late Augustan period. A nearby building, from which the remains of mortar pavements and wall painted plaster originate, was demolished in the second half of the 1st century. There are no archaeological traces in the investigated area from the 2nd century onwards.52 In the northern part of the settlement at Breg, the buildings must have been better built, judging by the stray finds of mosaic tiles and painted plaster.53 52 Horvat, Peterle Udovič, Žerjal 2007; Žerjal, Peterle Udovič 2008. 53 Horvat 1990, 61–63, 175–178: there have been no modern investigations in the area. Sl. 6: Vrhnika. Trdnjava na Gradišču in stolp na Turnovšču. M. = 1 : 4.000. Fig. 6: Vrhnika. The fort at Gradišče and the tower at Turnovšče. Scale 1:4,000. (Dopolnjeno po / Supplemented after Horvat 1990, pril. / Map 5) lokaciji Gradišče 5 (sl. 1: 7)sodijo v sredino oz. v drugo polovico 1. st. Sledi več gradbenih faz v drugi polovici 1. in v 2. st. V notranjosti raziskane stavbe sta ležali dve ognjišči, ki predstavljata ostanke podrobno nedoločene obrtniške dejavnosti. V ruševinah in nasutjih so bili najdeni deli barvanega stenskega ometa in mozaični kamenčki, ki kažejo na bližino bolj razkošnih stavb.54 Na Gradišču 5 je bila v drugotni rabi najdena kam­nita, z reliefi okrašena preklada (sl. 5). Po dimenzijah sodeč izvira iz pomembne zgradbe, morda svetišča.55 POZNA RIMSKA DOBA V pozni rimski dobi je še živela naselbina na Bre­gu (sl. 1: 4).56 Posamične najdbe iz 4. st. izvirajo tudi z nasprotnega brega Ljubljanice, z Dolgih njiv (sl. 1: 2), in so lahko znak ponovne rabe prostora.57 Morda sodi na območje Dolgih njiv tudi novčni zaklad, zakopan po letu 270.58 Na Gradišču je stala trdnjava petkotne oblike (sl. 1: 8). Obrambni zidovi so bili široki 2,3 m, z okroglimi stolpi v vogalih in pravokotnimi na stranicah (sl. 6).59 Najdeni so bili arhitektonski členi iz lehnjaka in nekaj 54 Bavec, Horvat 1996. V letu 2018 so bili v bližini odkriti ostanki še ene stavbe iz 1. st.; podatek T. Žerjal. 55 Bavec, Horvat 1996, 109–110. 56 Horvat, Mušič 2007, 170. Novci: Horvat 1990, 87–96, 195–204; Mušič, Horvat 2007, 261, 279. 58 Horvat 1990, 93–94, 201–202 (najdišče Verd). 59 Horvat 1990, 74–77, 185–187. Natančen potek severo­vzhodnega obzidja je bil ugotovljen leta 2018; podatek T. Žer­jal. In the 1st and 2nd centuries, a separate settlement area developed at Gradišče (Fig. 1: 8; 6), about 1200 m from the Breg settlement. The first masonry building, pave­ment, and levelled ground at the present-day location of Gradišče 5 (Fig. 1: 7) date to the middle or second half of the 1st century. Several construction phases followed in the second half of the 1st century and in the 2nd century. Two fireplaces, which represent the remains of an unidentified craft activity, were found in the interior of the investigated building. Fragments of painted wall plaster and mosaic tiles were discovered in the ruins and backfills, indicating the existence of more luxurious buildings in the vicinity.54 A relief-decorated stone architrave (Fig. 5) was found in a secondary position at Gradišče 5. Judging by its dimensions, it originates from an important building, possibly a temple.55 LATE ROMAN PERIOD The settlement at Breg still existed in the Late Roman period (Fig. 1: 4).56 Stray finds from the 4th century also come from Dolge njive on the opposite bank of the Lju­bljanica (Fig. 1: 2), and could be an indication of a renewed use of the area.57 A hoard of coins, buried after the year 270, could also have come from the area of Dolge njive.58 54 Bavec, Horvat 1996. In 2018, remains of another build­ing from the 1st century AD were discovered in the vicinity; information by Tina Žerjal. 55 Bavec, Horvat 1996, 109–110. 56 Horvat, Mušič 2007, 170. 57 Coins: Horvat 1990, 87–96, 195–204; Mušič, Horvat 2007, 261, 279. 58 Horvat 1990, 93–94, 201–202 (the Verd site). novcev (republikanski srebrnik ter novci od Vespazijana do Teodozija).60 Po obliki in legi sodeč je bila trdnjava morda zgrajena že konec 3. st.61 Na Turnovšču (sl. 1: 9; 6), kopastem pomolu okoli 30 m nad utrdbo na Gradišču in nad rimsko cesto, je bil postavljen kvadratni stolp (stranica 11,5 m) z debelim temeljem (1,6 m). Lehnjak je bil uporabljen za gradnjo okvirjev oken in vrat. Drobne najdbe sodijo v pozno rimsko dobo. Stolp je imel dober pregled nad okolico in je bil verjetno funkcionalno povezan s trdnjavo na Gradišču.62 GROBIŠČA Grobišče je ležalo vzdolž ceste južno od naselbine na Bregu. Naključno so bili odkriti posamezni žgani in skeletni grobovi.63 Na pobočju Sv. Trojice nad rimsko cesto so v letih 2013–2014 izkopali 55 žganih in verjetno osem skeletnih grobov, ki so datirani od druge polovice 1. do 4. st. Več žganih grobov je imelo obliko skrinje, sestavljene iz kamnitih plošč ali opek, v dveh primerih je šlo za sežig na mestu pokopa – t. i. bustum (sl. 1: 10).64 Verjetno je še eno grobišče ležalo ob cesti severno od naselbine na Bregu (sl. 1: 12).65 Osamljeni grobovi v okolici Vrhnike morda kažejo na manjše zaselke oziroma na posamezne vile.66 Z grobov v Navportu ali njegovi širši okolici verjetno izvirata tudi nagrobnika, vzidana v samosta-nu Bistra, ki sodita v drugo polovico 2. st. oz. v prvo polovico 3. st.67 OKOLICA NAVPORTA Gradbeni napis, vzidan v samostan v Bistri, omenja Neptunovo svetišče s portikom, ki ga je v 1. st. po Kr. dal postaviti priseljenec iz Akvileje, Lucij Servilij Sabin. Svetišče je domnevno stalo na izvirih Bistre.68 Pri Sinji gorici, okoli 460 m vzhodno od Dolgih njiv, je bila v strugi Ljubljanice odkrita ladja iz začetka 1. st. po Kr. (sl. 1: 11).69 60 Horvat 1990, 75–76, 92, 186, 200. 61 Pröttel 1996, 138–139. 62 Horvat 1990, 77–79, 187. 63 Horvat 1990, 72–73, 183–184. 64 Parc. št. 2215/2, k. o. Vrhnika; Janežič, Mulh, Černe 2017; Mulh, Černe 2018. 65 Horvat 1990, 66–67, 179–180. 66 Horvat 1990, 80–82, 188–189. 67 EDR156080, 156082; Šašel Kos 1990, 23–24, 29 150, 155–156. 68 EDR156077; Šašel 1960–1961, 188–189; Šašel Kos 1990, 23, 29, 150, 155–156; Horvat 1990, 82–83, 190–191. 69 Erič et al. 2014. A pentagonal fort stood at Gradišče (Fig. 1: 8). It had 2.3 m wide defence walls, round towers in the corners and rectangular ones on the sides (Fig. 6).59 Architectural ele­ments made of tufa and some coins (a republican silver coin and coins from Vespasian to Theodosius) were discovered there.60 Judging by its shape and position, the fort could have been built as early as the end of the 3rd century.61 A square tower stood on the domed promontory of Turnovšče (Fig. 1: 9; 6), some 30 m above the fort at Gradišče and the Roman road. Its sides measured 11.5 m and it had a massive foundation (1.6 m). Window and door frames were made of tufa. Small finds belong to the Late Roman period. The tower commanded a good view of the surroundings and was probably functionally connected with the fort at Gradišče.62 CEMETERIES A cemetery was situated along the road south of the settlement at Breg. Individual cremation and inhu­mation graves were discovered by chance.63 In the years 2013–2014, 55 cremation and probably eight inhumation graves, dating from the second half of the 1st century to the 4th century, were excavated on the slope of the Sv. Trojica hill above the Roman road. Several cremation graves were in the form of a cist, made of stone slabs or bricks. In two cases, cremation took place at the location of the burial – the so-called bustum (Fig. 1: 10).64 Another cemetery was probably situated along the road north of the settlement at Breg (Fig. 1: 12).65 Isolated graves in the surroundings of Vrhnika might indicate smaller settlements or individual villas.66 Two tombstones from the second half of the 2nd century or first half of the 3rd century, which are built into the Bistra monastery, probably originate from the graves in Nauportus or its wider surroundings.67 THE SURROUNDINGS OF NAUPORTUS A construction inscription, built into the Bistra monastery, mentions a sanctuary with a portico, dedi­cated to Neptune and built in the 1st century AD by an 59 Horvat 1990, 74–77, 185–187. The exact course of the northeast wall was identified in 2018; information by Žerjal. 60 Horvat 1990, 75–76, 92, 186, 200. 61 Pröttel 1996, 138–139. 62 Horvat 1990, 77–79, 187. 63 Horvat 1990, 72–73, 183–184. 64 Plot no. 2215/2, cadastral municipality Vrhnika; Jane­žič, Mulh, Černe 2017; Mulh, Černe 2018. 65 Horvat 1990, 66–67, 179–180. 66 Horvat 1990, 80–82, 188–189. 67 EDR156080, 156082; Šašel Kos 1990, 23–24, 29 150, 155–156. ETIMOLOGIJA IMENA Po mnenju Luke Repanška je ime Nauportus lahko ali v celoti latinsko, v celoti predlatinsko substratno ali hibridno. V prvem členu imena je najverjetneje ohra­njeno vulgarnolatinsko nau- “ladja”, v drugem členu portu­ pa je ohranjen pomen “prehod”, “pristanišče” ali “skladišča”.70 Možno je, da so rimski naseljenci na osnovi napačne latinske etimologije domače predkeltsko ime prilagodili svojemu jeziku. Na to bi kazali različni 71 zapisi imena pri Strabonu, kot Naúpontos in Pámportos. STATUS NASELJA Strabon opisuje Navport kot postojanko na trgovski poti iz Akvileje proti vzhodu. Tu so blago preložili z vo­zov na ladje, ki so plule po rekah Ljubljanici in Savi proti Segestiki.72 Strateški pomen naselja na meji med Italijo in Ilirikom se pokaže v času panonsko-delmatskega upora (6–9 po Kr.), ko so uporniki načrtovali napad na Italijo čez Navport.73 Strabon v 7. knjigi omenja, da je bil Navport v ro­kah Tavriskov, za kar je verjetno uporabil starejše vire, v katerih je bila opisana situacija v sredini 2. st. pr. Kr.74 Na podlagi pomena imena je Jaroslav Šašel domneval, da je v Navportu obstajala tavriskijska mitninska postaja.75 Iz drugega odlomka pri Strabonu izhaja, da so Navport v 1. st. pr. Kr. nadzorovali Rimljani.76 Literarni Repanšek 2016, 197–199; Šašel Kos 2017, 227–228. Tudi: Šašel Kos 1990, 20, 146–147. 71 Šašel Kos 2017, 228. 72 Strabon 4, 6, 10; 7, 5, 2. 73 Velej Paterkul 2, 110, 4. 74 Strabon 7, 5, 2; Šašel Kos 1990, 18–19, 144–145. 75 Šašel 1966. 76 Strabon 4, 6, 10; Šašel Kos 1990, 17–19, 143–145. immigrant from Aquileia, Lucius Servilius Sabinus. The sanctuary was supposedly located at the springs of the Bistra steam.68 Near Sinja gorica, some 460 m east of Dolge njive, a boat from the beginning of the 1st century AD was dis­covered in the riverbed of the Ljubljanica (Fig. 1: 11).69 ETYMOLOGY OF THE NAME According to Luka Repanšek, the name Nauportus could be a fully Latin name, a fully pre-Latin substrate name, or a hybrid of the two. The meaning preserved in the first segment of the name is most likely the vulgar Latin nau- “ship”, while the meaning preserved in the sec­ 70 ond segment portu­ is “passage”, “port”, or “storehouses”.It is possible that, on the basis of a wrong etymology, the Roman settlers adapted the autochthonous pre-Celtic name to their own language. This might be indicated by the two different writings of the name in Strabo: Naúpon­tos and Pámportos.71 SETTLEMENT STATUS Strabo describes Nauportus as a post on the trade route from Aquileia to the east. This is where the goods were loaded from carts onto the boats navigating the rivers Ljubljanica and Sava towards Segestica.72 The strategic significance of the settlement on the border between Italy and Illyricum became important in the time of the Pannonian-Dalmatian uprising (AD 6–9), when the insurgents planned an attack upon Italy via Nauportus.73 Strabo mentions in his seventh book that Nauportus was held by the Taurisci. His information probably came from earlier sources, where the situation in the 2nd century BC was described.74 Based on the meaning of the name,Jaroslav Šašel speculated that there had been a Tauriscan toll station in Nauportus.75 The second excerpt of Strabo’s writing implies that Nauportus was controlled by the Romans in the 1st century BC.76 The written source is supported by two inscription stones mentioning the heads of the village (magistri vici). On the basis of the archaic language and the form of the letters, the stones can be dated to the 68 EDR156077; Šašel 1960–1961, 188–189; Šašel Kos 1990, 23, 29, 150, 155–156; Horvat 1990, 82–83, 190–191. 69 Erič et al. 2014. 70 Repanšek 2016, 197–199; Šašel Kos 2017, 227–228. Also: Šašel Kos 1990, 20, 146–147. 71 Šašel Kos 2017, 228. 72 Strabo 4, 6, 10; 7, 5, 2. 73 Velleius Paterculus 2, 110, 4. 74 Strabo 7, 5, 2; Šašel Kos 1990, 18–19, 144–145. 75 Šašel 1966. 76 Strabo 4, 6, 10; Šašel Kos 1990, 17–19, 143–145. vir potrjujeta dva napisna kamna, na katerih so omenjeni vaški načelniki (magistri vici). Kamna sta po arhaičnem jeziku in obliki črk datirana v prvo polovico ali sredino 1. st. pr. Kr. (sl. 7).77 Tacit piše, da je bil Navport vikus, podoben municipiju, kar govori o njegovem velikem pomenu v letu 14.78 Mejnik med akvilejskim in emonskim mestnim ozemljem, ki je bil odkrit pri Bevkah, priča, da je v zgodnji rimski dobi Navport ležal na akvilejskem ozem­lju in je meja prečila Ljubljansko barje nekaj več kot 5 km vzhodno od naselja.79 Verjetno je tudi cesta, ki je povezovala Akvilejo z Navportom, v celoti potekala po akvilejskem ozemlju.80 Tabula Peutingeriana omenja v Navportu cestno postajo.81 PREBIVALCI Na štirih napisnih kamnih, ki so datirani v pozno­republikanski ali zgodnjeavgustejski čas, se pojavljajo imena pripadnikov akvilejskih oziroma italskih trgov­skih družin (sl. 7). V pretežni meri gre za osvobojence, ki imajo vlogo vaških načelnikov (magistri vici).82 Italiki so omenjeni tudi na štirih svinčenih ploščicah iz prve polovice 1. st. (sl. 8).83 Na prevlado italskega prebivalstva v predavgustej­skem in avgustejskem času kaže tudi sestava drobnih najdb, med katerimi popolnoma prevladuje iz Italije uvožena keramika, vključno z grobim kuhinjskim posodjem.84 Tacit omenja v letu 14 v Navportu prisotnost vojske,85 ki jo tudi nakazujejo vojaške najdbe iz naselbine in iz Ljubljanice86 ter napisi na svinčenih ploščicah.87 Avtohtono prebivalstvo je arheološko slabo opri­jemljivo.88 Najbolje ga predstavlja čaščenje boginje Ekorne, katere ime kaže na predrimsko in predkeltsko lokalno božanstvo. Kult so prevzeli italski naseljenci že v 1. st. pr. Kr. in je bil vezan na Navport in na Emono.89 77 Šašel Kos 1990, 22–28, 148–155; Šašel Kos 1997, 117– 120; Šašel Kos 1998, 101–104. 78 Tacit, Ann. 1, 20, 1; Šašel Kos 1990, 21, 148. 79 Šašel Kos 2002a; 2002b; 2014; o meji tudi Horvat, Sa­ gadin 2017, 203–204.80 Šašel Kos 2002b, 256. 81 Tab. Peut. IV, 1. 82 EDR128824, 128825, 156071, 156084; Šašel Kos 1990, 21–30, 148–156; Šašel Kos 1997, 117–122; Šašel Kos 1998, 101–106. 83 Grassl 2017. 84 Horvat 1990, 115–132, 218–235; Vojaković et al. 2019. 85 Tacit, Ann. 1, 20, 1. 86 Horvat 1990, 114, 217; Istenič 2009a. 87 Grassl 2017. 88 Horvat 1990, 123–131, 226–234. 89 Šašel Kos 1992; 1999; Šašel Kos 2000b, 39–40. first half or to the middle of the 1st century BC (Fig. 7).77 Tacitus writes that Nauportus was a vicus that resembled a municipium, which indicates its great significance in the year AD 14.78 A boundary stone between the city territories of Aquileia and Emona, discovered near the nearby vil­lage of Bevke, shows that in the Early Roman period Nauportus lay in the Aquileian territory. The boundary crossed Ljubljansko barje slightly more than 5 km east of the settlement.79 Similarly, the road between Aquileia and Nauportus probably ran entirely over the Aquileian territory.80 A road station in Nauportus is mentioned in the Tabula Peutingeriana.81 INHABITANTS Names of members of Aquileian or Italian merchant families appear on four inscription stones, dated to the Late Republican or Early Augustan time (Fig. 7). These people were mostly freedmen and they had the role of magistri vici.82 Italian inhabitants are also mentioned in four lead tablets from the first half of the 1st century AD (Fig. 8).83 The predominance of the Italian population in the Pre-Augustan time is indicated also by the structure of small finds, among which Italian-imported pottery, including coarse kitchenware, prevails.84 Tacitus mentions a military presence in Nauportus in the year AD 14.85 His claim is supported by military finds from the settlement and from the Ljubljanica,86 as well as by inscriptions on lead tablets.87 The autochthonous population is archaeologically poorly documented.88 It is best expressed in the worship of the goddess Aecorna, whose name indicates a pre-Roman and pre-Celtic local deity. The cult was adopted by the settlers from Italy in Nauportus and Emona as early as the 1st century BC.89 77 Šašel Kos 1990, 22–28, 148–155; Šašel Kos 1997, 117– 120; Šašel Kos 1998, 101–104. 78 Tacitus, Ann. 1, 20, 1; Šašel Kos 1990, 21, 148. 79 Šašel Kos 2002a; 2002b; 2014; for the boundary see also Horvat, Sagadin 2017, 203–204. 80 Šašel Kos 2002b, 256. 81 Tab. Peut. IV, 1. 82 EDR128824, 128825, 156071, 156084; Šašel Kos 1990, 21–30, 148–156; Šašel Kos 1997, 117–122; Šašel Kos 1998, 101–106. 83 Grassl 2017. 84 Horvat 1990, 115–132, 218–235; Vojaković et al. 2019. 85 Tacitus, Ann. 1, 20, 1. 86 Horvat 1990, 114, 217; Istenič 2009a. 87 Grassl 2017. 88 Horvat 1990, 123–131, 226–234. 89 Šašel Kos 1992; 1999; Šašel Kos 2000b, 39–40. GOSPODARSTVO Na podlagi pisnih virov in materialnih ostankov lahko domnevamo, da sta bili v 1. st. pr. Kr. in v av-gustejski dobi najpomembnejši dejavnosti v Navportu trgovina ter prevozništvo po kopnem in reki. Svinčene napisne ploščice (sl. 8) dodatno osvetlijo razgibano gospodarsko življenje v prvi polovici 1. st. po Kr. Ena priča o trgovanju s sadikami vinske trte. Na drugih dveh je omenjen uvoz začinjene ribje omake za vojaško posadko ter dobava majhnih tunik, morda za potrebe vojske. Četrta ploščica govori o pošiljki volne za potrebe tekstilne obrti, ki je morda delovala v samem Navportu.90 Na območju Brega (Jelovškoval ul. 10–11) so bili odkriti sledovi predelave brona. RAZVOJ NAVPORTA V pozni prazgodovini in na začetku rimske dobe je bil Navport trgovska postojanka na tranzitni poti Akvi­leja–Segestika. V pristanišču na Ljubljanici so preložili na ladje tovore za pot proti vzhodu, tiste, usmerjene na zahod, pa na vozove.91 Naselje Tavriskov iz 2. st. pr. Kr., ki ga omenja Strabon, še ni bilo ugotovljeno. Konec 2. ali na začetku 1. st. pr. Kr. je bila na levem bregu Ljubljanice, pri Stari pošti, postavljena prva rimska naselbina (sl. 1: 1). Zelo verjetno je ob njej ležalo pristanišče. Sočasna plovila, odkrita v reki, pričajo o živahnem prometu (sl. 1: 3). Navport je imel od prve polovice ali sredine 1. st. pr. Kr. dalje status vikusa na akvilejskem mestnem ozemlju. Vodilno vlogo so imele trgovske družine iz Akvileje, saj so bili njihovi osvobojenci magistri vici. Postavljena sta bila svetišče Ekorne in portik, ki pa še nista locirana. Močno navezanost na Italijo kaže tudi drobna materi­alna kultura, npr. keramika iz naselbine pri Stari pošti, in najstarejši napisni kamni, izklesani iz nabrežinskega apnenca.92 Staroselci so bili potisnjeni v ozadje. 90 Grassl 2017. Promet po Ljubljanici: Istenič 2009b; Gaspari 2017, 127–155. O pomenu Navporta tudi: Šašel Kos 2000a, 294–297; Šašel Kos 2005, 480–482. 92 Šašel Kos 1997, 117, 120; Šašel Kos 1998, 101, 105. Sl. 8: Vrhnika, Jelovškova ulica 10–11. Svinčena ploščica z imenom Arius Nauportanus. Fig. 8: Vrhnika. Jelovškova ul. 10–11. Lead tablet with the name Arius Nauportanus. ECONOMY On the basis of written sources and material remains it can be surmised that the two main economic activities in Nauportus in the 1st century BC and the Augustan period were trade and freighting on land and on the river. Lead inscription tablets (Fig. 8) give us a further insight into the varied economic life in the first half of the 1st century AD. One of them testifies to the trade with vine seedlings. Two mention the import of a spicy fish sauce for the garrison and the supply of small tunics, perhaps for the needs of the army. The fourth tablet mentions a ship­ment of wool for textile manufacture, which was perhaps located in Nauportus itself.90 Traces of bronze working were discovered in the Breg area (Jelovškova ulica 10–11). DEVELOPMENT OF NAUPORTUS In the late Prehistoric period and at the beginning of the Roman Age, Nauportus was a trade station on the Aquileia–Segestica transit route. In the port on the Ljubljanica, cargo was loaded onto boats travelling east and onto carts travelling west.91 The 2nd century BC settlement of the Taurisci mentioned by Strabo has not been identified so far. At the end of the 2nd / beginning of the 1st century BC, the first Roman settlement was built on the left bank of the Ljubljanica, at Stara pošta (Fig. 1: 1). It is very likely that the port was situated next to it. Judging by the contem­porary vessels discovered in the river, the traffic was busy (Fig. 1: 3). From the first half or the middle of the 1st century BC on, Nauportus had the status of vicus in the Aquileian city territory. Merchant families from Aquileia played the leading role, as their freedmen were the magistri vici. The sanctuary of Aecorna and a portico were built, both not yet located. A strong connection with Italy is further indicated by small material culture, e.g. the pottery from the settlement at Stara pošta, and the earliest inscription 90 Grassl 2017. 91 For the traffic on the Ljubljanica: Istenič 2009b; Gas-pari 2017, 127–155. For the significance of Nauportus also: Šašel Kos 2000a, 294–297; Šašel Kos 2005, 480–482. V četrtem ali tretjem desetletju pr. Kr. je bila na Dolgih njivah v okljuku Ljubljanice zgrajena utrjena naselbina, katere jedro je predstavljal trg, obdan z veli­kimi skladišči (sl. 1: 2; 2). Ob njej je ležalo pristanišče. Zasnova naselbine in posameznih stavb se veže na vzorce iz poznorepublikanske severne Italije in na arhitekturo pristanišč na širšem prostoru imperija. Šlo je za utrjeno prekladalno postajo in pristanišče, prek katerih sta pote­kala tranzitni promet in trgovina. Velikost skladišč kaže na izjemno količino tovorov. Postojanka je morala igrati pomembno vlogo v oskrbi legij na prostoru srednjega Podonavja in severnega Balkana. Oskrba je bila verjet-no v rokah akvilejskih trgovcev.93 Prisotnost vojske in vojaške prevoze posredno dokazujejo najdbe orožja iz naselbine, še bolj pa iz reke Ljubljanice.94 Naselbina na Dolgih njivah je bila opuščena po avgustejskem obdobju, ko se je zaradi konsolidacije rimske oblasti v srednjem Podonavju zmanjšala količina tovorov, hkrati pa je novozgrajena cesta prevzela del blaga, ki je prej potovalo po reki.95 Od 1. st. po Kr. dalje je Navport izgubil na pomenu in je kolonija Emona prevzela vodilno prometno in upravno vlogo širšega območja.96 V srednjem ali poznoavgustejskem obdobju se je začela razvijati tudi naselbina na nasprotnem bregu Ljubljanice, na Bregu, ki je imelo prometno ugodnejši položaj, hkrati ob reki in tik ob novi cesti Akvileja–Emo­na. Območje je bilo pozidano s skromnejšimi stavbami, kot starejši naselbinski predel na Dolgih njivah. Na začetku je prevladovala lesena gradnja (Jelovškova ul. 10–11). Ob priključku na avtocesto (avgustejska doba; sl. 1: 5) in na Jelovškovi ul. 10–11 (druga polovica 1. in 2. st.; sl. 1: 4) so bile zgrajene velike stavbe s kamnitimi temelji, ki so morda tudi imele funkcijo skladišč. V okolici Navporta so nastala manjša naselbinska jedra. Naselbina na Bregu je bila verjetno zapuščena ta­krat kot večina nižinskih naselij v jugovzhodnoalpskem prostoru, v prvi polovici 5. st.97 Morda je bila že konec 3. st. postavljena trdnjava na Gradišču (sl. 1: 8). Bližnji opazovani stolp na Turnovšču je datiran v poznorimsko dobo (sl. 1: 9). Obe utrdbi sta ščitili glavno pot proti Italiji. V 4. st. je bil zgrajen po hribovju zahodno od Navporta Ajdovski zid, to je 10 km dolg obrambni zid z opazovanimi stolpi, ki je zapiral poti proti zahodu in je 98 bil del obrambnega sistema Claustra Alpium Iuliarum.V času njegovega delovanja je morala biti trdnjava na Gradišču že opuščena. Najpozneje sredi 5. st. je promet po stari tranzitni poti čez Navport zamrl. 93 Mušič, Horvat 2007, 267, 283. 94 Istenič 2009a; 2009b; 2009c; 2012. 95 Mušič, Horvat 2007, 267, 283. 96 Šašel Kos 2012. 97 Ciglenečki 2012. 98 Šašel, Petru 1971, 77–81; Kusetič et al. 2014, 72–77. stones made of the Aurisina limestone.92 The autochtho-nous population was pushed into the background. In the fourth or third decade BC, a fortified settle­ment was built in the meander of the Ljubljanica at Dolge njive. Its centre was a square, surrounded by large store­houses (Fig. 1: 2; 2). Next to the settlement there was a port. The layout of the settlement and of individual build­ings is related to models from the Late Republican north Italy and to the architecture of Roman ports. The site of Dolge njive was a fortified loading station and port for transit traffic and trade. The size of the storehouses indi­cates large amounts of cargo. The post must have played an important role in the supply of the legions in the Middle Danube area and the northern Balkans. The supply was most likely in the hands of Aquileian merchants.93 Army presence and military transports are indirectly confirmed by the weapons discovered within the settlement and even more so in the Ljubljanica.94 The Dolge njive settlement was abandoned after the Augustan period, when, due to the consolidation of the Roman rule in the Middle Danube area, the amount of cargo decreased. At the same time, some of the traffic was transferred from the river to the newly constructed road.95 In the 1st century AD, Nauportus lost its significance and the colony of Emona assumed the leading role in traffic and administration in the wider area.96 The Middle and Late Augustan periods also saw the beginning of the development of the settlement at Breg on the opposite bank of the Ljubljanica. Traffic-wise, the new settlement had a better position, since it was situated both near the river and on the new Aquileia–Emona road. The buildings were more modest than those in the earlier settlement at Dolge njive and wooden constructions pre­dominated at first (Jelovškova ulica 10–11). In the area of the highway exit-road (Augustan period; Fig. 1: 5) and at Jelovškova ulica 10–11 (second half of the 1st century and the 2nd century; Fig. 1: 4), large buildings with stone foundations were constructed, possibly also functioning as storehouses. Small settlement areas emerged in the surroundings of Nauportus. The settlement at Breg was probably abandoned at the same time as most of the lowland settlements in the southeast Alpine area, i.e. in the first half of the 5th cen­tury.97 The fort at Gradišče was built perhaps as early as the end of the 3rd century (Fig. 1: 8). The nearby tower at Turnovšče is dated to the Late Roman period (Fig. 1: 9). The fort and the tower guarded the main route towards Italy. In the 4th century, a 10 km long defence wall with towers (“Ajdovski zid”) was built in the hills west of Nau­portus. It blocked the way to the west and was part of the 92 Šašel Kos 1997, 117, 120; Šašel Kos 1998, 101, 105. 93 Mušič, Horvat 2007, 267, 283. 94 Istenič 2009a; 2009b; 2009c; 2012. 95 Mušič, Horvat 2007, 267, 283. 96 Šašel Kos 2012. 97 Ciglenečki 2012. ANDRIČ, M. 2016, Človekov vpliv na rastlinstvo zahodnega Ljubljanskega barja v pozni prazgodovini (pribl. 1000–50 pr. n. št.). Primer: Vrhnika (Dolge njive) / Human impact on the vegetation of the western Ljubljansko barje in late prehistory (ca. 1000–50 cal. BC). Case study: Vrhnika (Dolge njive). – Arheološki vestnik 67, 259–275. BAVEC, U., J. HORVAT 1996, Vrhnika, Gradišče 5. – V / In: Vrhniški razgledi 1, 107–110. BRAS KERNEL, H. 2006, “Vrhniški zaklad” je del zbirke najdb iz Ljubljanice pri Bevkah (The “Vrhnika hoard” belongs to the collection of finds from the Ljubljanica near Bevke). – Argo 49/2, 11–24. BEKLJANOV ZIDANŠEK, I., T. ŽERJAL 2018, Poročilo o ar­heoloških raziskavah pri projektu “Infrastruktura Delavsko naselje”. – Ljubljana (neobjavljeno poročilo / unpublished report). CIGLENEČKI, S. 2012, Spremenjena podoba poznoantičnih urbanih središč – prispevek k transformaciji poselitvene slike v jugovzhodnoalpskem prostoru. – V / In: I. Lazar, B. Županek (ur. / eds.), Emona med Akvilejo in Panonijo / Emona between Aquileia and Pannonia, Annales Mediter­ranei, 459–478, Koper. ČUFAR et al. 2019 = ČUFAR, K., J. HORVAT, T. TOLAR, T. BERDEN, M. MERELA 2019, Raziskovalni potencial lesa sodov iz rimskih vodnjakov (Research potential of wood of barrels from Roman water wells). – Les / Wood 68/1, 47–60. EDR = Epigraphic Database Roma. Skrbnik / Service provider: DigiLab Centro interdipartimentale di ricerca e servizi, Sapienza Universita di Roma) [http://www.edr-edr.it]. ERIČ et al. 2014 = ERIČ, M., A. GASPARI, K. ČUFAR, F. SO­LINA, T. VERBIČ 2014, Zgodnjerimska ladja iz Ljubljanice pri Sinji Gorici / Early Roman barge from the Ljubljanica River at Sinja Gorica. – Arheološki vestnik 65, 187–254. FMRSl I = P. Kos, Die Fundmünzen der römischen Zeit in Slowenien I. – Berlin 1988. FMRSl III = P. Kos, A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien III. – Berlin 1995. FMRSl IV = A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien IV. – Berlin 1998. FMRSl V = A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien V. – Mainz am Rhein 2004. FMRSl VI = A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien VI. – Wetteren 2010. GASPARI, A. 2006, Bronastodobne najdbe iz potoka Ljubija pri Verdu / Bronze age finds from the Ljubija stream near Verd. – V / In: A. Gaspari (ur. / ed.), Zalog pri Verdu, Opera Instituti Archaeologici Sloveniae 11, 205–221. Claustra Alpium Iuliarum defence system.98 In that time, the Gradišče fort must have already been abandoned. In the mid-5th century at the latest, traffic on the old transit route through Nauportus died out. Translation: Meta Osredkar 98 Šašel, Petru 1971, 77–81; Kusetič et al. 2014, 72–77. GASPARI, A. 2017, Deblak s konca 2. stoletja pr. n. št. iz Ljub­ljanice na Vrhniki / The late 2nd century BC logboat from the Ljubljanica River at Vrhnika. – Ljubljana. GASPARI, A., M. ERIČ 2008, Arheološke raziskave struge Ljubljanice med Verdom in Vrhniko / Ricerche archelo­giche nel letto del fiume Ljubljanica fra Verd e Vrhnika. – Annales, Series historia et sociologia 18/2, 407–430. GASPARI A., R. MASARYK 2009, Na sledi prazgodovinskega Navporta. Gradišče na hribu Tičnica na Vrhniki / Tracing the prehistoric Nauportus. The hillfort on Tičnica near Vrhnika. – Arheološki vestnik 60, 195–206. GASPARI, A., D. MLEKUŽ 2013, Gradišče na hribu Tičnica. – Vrhniški razgledi 14, 9–18. GASPARI, A., M. VINAZZA 2018, Prazgodovinska naselbina na vrhu Tičnica (Vrhnika – Arheološko najdišče Naupor­tus). – V / In: Arheologija v letu 2017, dediščina za javnost, Ljubljana, 22. GRASSL, H. 2017, Die Rolle von Nauportus (Vrhnika) im Lich­te neuer Textfunde (Pomen Navporta v luči novih napisov na svinčenih ploščicah). – Arheološki vestnik 68, 459-469. HORVAT, J. 1990, Nauportus (Vrhnika). – Dela 1. razreda SAZU 33. HORVAT, J. 1996, Nauportus. Vrhnika v arheoloških obdobjih. – V / In: Vrhniški razgledi 1, 89–106. HORVAT, J. 2008, Early Roman horrea at Nauportus. – Mélanges de l’Ecole française de Rome, Antiquité 120/1, 111–121. HORVAT, J. 2009, Nauportus – naselje na začetku transportne poti po Ljubljanici. – V: P. Turk, J. Istenič, T. Knific, T. Na-bergoj (ur.), Ljubljanica – kulturna dediščina reke, 89–94, Ljubljana = Nauportus – a settlement at the beginning of the transportation route along the Ljubljanica. – In: P. Turk, J. Istenič, T. Knific, T. Nabergoj (eds.), The Ljubljanica – a River and its Past, 96–101, Ljubljana. HORVAT, J. 2012a, Zgodnjerimska utrjena naselbina na Dol­gih njivah na Vrhniki (Early Roman fortified settlement at Dolge njive in Vrhnika). – V / In: A. Gaspari, M. Erič (ur. / eds.), Potopljena preteklost, 287–292, Radovljica. HORVAT, J. 2012b, Skupek keramike iz prve polovice 1. stoletja iz Navporta (Assemblage of ceramic ware from the first half of the 1st century AD from Nauportus). – V / In: I. Lazar, B. Županek (ur. / eds.), Emona med Akvilejo in Panonijo / Emona between Aquileia and Pannonia, Annales Mediter­ranei, 273–299, Koper. HORVAT, J., B. MUŠIČ 2007, Nauportus, a commercial settlement between the Adriatic and the Danube. – V / In: M. Chiaba, P. Maggi, C. Magrini (ur. / eds.), Le Valli del Natisone e dell’Isonzo tra Centroeuropa e Adriatico, Studi e ricerche sulla Gallia Cisalpina 20, 165–174. HORVAT, J., P. PETERLE UDOVIČ 2006, Vrhnika. – Varstvo spomenikov 42. Poročila 2005, 188–190. HORVAT, J., M. SAGADIN 2017, Emonsko podeželje /Emona‘s countryside. – V / In: B. Vičič, B. Županek (ur. / eds.), Emona MM. Urbanizacija prostora – nastanek mesta / Emona MM. Urbanisation of space – beginning of a town, 201–223, Ljubljana. HORVAT et al. 2016 = HORVAT, J., P. PETERLE UDOVIČ, T. TOLAR, B. TOŠKAN 2016, Območje pristanišča v Nav­portu / The port area of Nauportus. – Arheološki vestnik 67, 177–258. HORVAT, J., P. PETERLE UDOVIČ, T. ŽERJAL 2007, Poročilo o zaščitnih arheoloških izkopavanjih na območju rekonstruk­cije in razširitve mostu VA 0141 čez Ljubljanico pri Vrhniki. – Ljubljana (neobjavljeno poročilo / unpublished report: arhiv IZA, ZRC SAZU, Ljubljana). ISTENIČ, J. 2003, The Early Roman “Hoard of Vrhnika”: A Collection of Finds from the River Ljubljanica (Zgodnje­rimski “Zaklad z Vrhnike”: zbirka najdb iz reke Ljubljanice). – Arheološki vestnik 54, 281–298. ISTENIČ, J. 2009a, The early Roman military route along the River Ljubljanica (Slovenia). – V / In: A. Morillo, N. Hanel, E. Martín (ur. / eds.), Limes XX, XXth International Congress of Roman Frontier Studies, 855–865, Madrid. ISTENIČ, J. 2009b, Ljubljanica – rimska trgovska in prometna pot. – V: P. Turk, J. Istenič, T. Knific, T. Nabergoj (ur.), Ljubljanica – kulturna dediščina reke, 74–80, Ljubljana = The Ljubljanica – a Roman trade and transport route. – In: P. Turk, J. Istenič, T. Knific, T. Nabergoj (eds.), The Ljublja­ nica – a River and its Past, 79–85, Ljubljana. ISTENIČ, J. 2009c, Ljubljanica in rimska vojska. – V: P. Turk, J. Istenič, T. Knific, T. Nabergoj (ur.), Ljubljanica – kul­turna dediščina reke, 81–85, Ljubljana = The Ljubljanica and the Roman army. – In: P. Turk, J. Istenič, T. Knific, T. Nabergoj (eds.), The Ljubljanica – a River and its Past, 86–91, Ljubljana. ISTENIČ, J. 2012, Reka Ljubljanica v rimski dobi (The River Ljubljanica in the Roman period). – V / In: A. Gaspari, M. Erič (ur. / eds.), Potopljena preteklost, 283–286, Radovljica. JANEŽIČ, M., T. MULH, M. ČERNE 2017, Roman lamps from the graves in Vrhnika (Slovenia). – V / In: Rimske i kasnoantičke svjetiljke: proizvodnja i distribucija, kontakti na Mediteranu / Roman and Late antique lamps: production and distribution, contacts on the Mediterranean, Zbornik Instituta za arheologiju 7, 57–67. KOS, P. 1977, Keltski novci Slovenije / Keltische Münzen Slowe­niens. – Situla 18. KUSETIČ et al. 2014 = J. KUSETIČ, P. KOS, A. BREZNIK, M. STOKIN 2014, Claustra Alpium Iuliarum. Med raziskovan­jem in upravljanjem / Claustra Alpium Iuliarum. Between Research and Management. – Ljubljana. LOGAR, N. 1986, Vrhnika / Dolge njive. – Arheološki pregled 1985, 126–127. MIKL CURK, I. 1974, Utrdbe Nauporta ob Ljubljanici na Vrhniki (Nauportus fortresses by the river Ljubljanica at Vrhnika). – Arheološki vestnik 25, 370–386. MULH, T., M. ČERNE 2018, Bustum pokop v Navportu (Vrhnika) / The bustum burial in Nauportus (Vrhnika). – V / In: Nova odkritja med Alpami in Črnim morjem / New Discoveries between the Alps and the Black Sea, Monografije CPA 6, 205–227 [http://www.zvkds.si/sites/www.zvkds.si/ files/uploads/files/publication/zbornik_ptuj_2015.pdf]. MUŠIČ, B., J. HORVAT 2007, Nauportus - an Early Roman trading post at Dolge njive in Vrhnika / Nauportus – zgodnjerimska trgovska postojanka na Dolgih njivah na Vrhniki. – Arheološki vestnik 58, 219–283. PRÖTTEL, Ph. M. 1996, Mediterrane Feinkeramikimporte des 2. bis 7. Jahrhunderts n. Chr. im oberen Adriaraum und in Slowenien. – Kölner Studien zur Archäologie der römischen Provinzen 2, Espelkamp. REPANŠEK, L. 2016, Keltska dediščina v toponimiji jugovzhod­nega alpskega prostora (Celtic Legacy in the Toponymy of South­Eastern Alps). – Linguistica et philologica 33. SCHMID, W. 1943, Das Eindringen der römischen Kultur in Noricum. – Das Joanneum 6, 7–28. ŠAŠEL, J. 1960–1961, Epigraphica. – Arheološki vestnik 11–12, 187–210. ŠAŠEL, J. 1966, Keltisches portorium in den Ostalpen (zu Plin. n. h. III 128). – V / In: Corolla memoriae Erich Swoboda dedicata, Römische Forschungen in Niederösterreich 5, 198–204, Graz, Cologne (= Opera selecta, Situla 30, 1992, 500–506). ŠAŠEL, J. 1975, Rimske ceste v Sloveniji. – V / In: Arheološka najdišča Slovenije, 74–99, Ljubljana.ŠAŠEL, J., P. PETRU (ur. / eds.) 1971, Claustra Alpium Iuliarum I. Fontes. – Katalogi in monografije 5. ŠAŠEL KOS, M. 1990, Nauportus: antični literarni in epigrafski viri / Nauportus: literary and epigraphical sources. – V / In: J. Horvat, Nauportus (Vrhnika), Dela 1. razreda SAZU 33, 17–33, 143–159. ŠAŠEL KOS, M. 1992, Boginja Ekorna v Emoni (The goddess Aecorna in Emona). – Zgodovinski časopis 46/1, 5–12. ŠAŠEL KOS, M. 1997, The Roman Inscriptions in the National Museum of Slovenia. – Situla 35. ŠAŠEL KOS, M. 1998, Caesarian inscriptions in the Emona basin? – V / In: G. Paci (ur. / ed.), Epigrafia romana in area Adriatica, 101–112, Pisa, Roma. ŠAŠEL KOS, M. 1999, The goddess Aecorna in Emona. – V / In: Pre­Roman divinities of the eastern Alps and Adriatic, Situla 38, 47–61. ŠAŠEL KOS, M. 2000a, Caesar, Illyricum, and the hinterland of Aquileia. – V / In: G. Urso (ur. / ed.), L’ultimo Cesare, 277–304, Roma. ŠAŠEL KOS, M. 2000b, Sacred places and epichoric gods in the southeastern Alpine area – some aspects. – V / In: C. Delplace, F. Tassaux (ur. / eds.), Les cultes polythéistes dans l‘Adriatique romaine, Ausonius–Publications, Études 4, 27–51, Bordeaux. ŠAŠEL KOS, M. 2002a, The boundary stone between Aquileia and Emona / Mejnik med Akvilejo in Emono. – Arheološki vestnik 53, 373–382. ŠAŠEL KOS, M. 2002b, Il confine nord-orientale dell‘Italia romana. – Aquileia Nostra 73, 245–260. ŠAŠEL KOS, M. 2005, Appian and Illyricum. – Situla 43. ŠAŠEL KOS, M. 2012, Colonia Iulia Emona – the genesis of the Roman city / Colonia Iulia Emona – nastanek rimskega mesta. – Arheološki vestnik 63, 79–104. ŠAŠEL KOS, M. 2014, The Problem of the Border between Italy, Noricum, and Pannonia. – Tyche 29, 153–164. ŠAŠEL KOS, M. 2017, Antična imena (ime?) za Ljubljanico / The ancient name(s) for the Ljubljanica River. – V / In: B.Vičič, B. Županek (ur. / eds.), Emona MM. Urbanizacija prostora – nastanek mesta. Urbanisation of space ­ beginning of a town, 225–234, Ljubljana. TURK, P. 2007, Bronasti polnoročajni meč iz Jabelj v osrednji Sloveniji / A Bronze Solid-Hilted Sword from Jablje in Cen­tral Slovenia. – V / In: M. Blečić et al. (ur. / eds.), Scripta Praehistorica in honorem Biba Teržan, Situla 44, 209–229. VOJAKOVIĆ et al. 2019 = VOJAKOVIĆ, P., I. BEKLJANOV ZIDANŠEK, B. TOŠKAN 2019, Poznorepublikanski Nav-port: območje Stare pošte (Late Republican Nauportus: the Stara pošta site). – Arheološki vestnik 70, 93–126. ŽERJAL, T., I. BEKLJANOV ZIDANŠEK 2018, Rimska na­selbina na Vrhniki - ulici Delavsko naselje in Partizanski tabor. – V / In: Arheologija v letu 2017, dediščina za javnost. Zbornik povzetkov, 73, Ljubljana [http://www.arheologija. si/files/2018/02/Arheologija-v-letu-2017.pdf].ŽERJAL, T., P. PETERLE UDOVIČ 2008, Vrhnika. – Varstvo spomenikov 44. Poročila 2007, 300–301. Jana Horvat Znanstvenoraziskovalni center SAZU Inštitut za arheologijo Novi trg 2 SI-1000 Ljubljana jana.horvat@zrc-sazu.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 113–121 VIPAVA Vesna TRATNIK Izvleček Rimska naselbina v Vipavi je v geografskem pogledu antičnih piscev ležala na prostoru Julijskih Alp, na vzhodnem obrobju akvilejskega agra. V tem času je naselbina domnevno obsegala vznožje vzpetine Stari grad in ravnico pod njo. Del hiše iz 1. in 2. st. je bil raziskan na Beblerjevi ulici. Odkrito je bilo tudi grobišče iz tega časa, za katerega so značilni pokopi v skupnih, morda družinskih grobnicah. Na severnem obrobju Vipave, na najdiščih Bela in Grublje so bili odkriti naselbinski ostanki iz časa od 1. do 4. st, z zidanimi in lesenimi stavbami. Najmlajše posamične rimske najdbe so datirane v drugo polovico 4. in na začetek 5. stoletja. Ključne besede: Italija (10. regija), Vipava, rimska doba, naselbina, grobišče, družinske grobnice Abstract The Roman settlement in Vipava lies in an area that ancient geographers considered part of the Julian Alps, at the eastern edge of the territory of Aquileia. Archaeological evidence has shown it stretched across the foot of the hill of Stari grad and the flatland below it. Part of the settlement was excavated at the Beblerjeva street, revealing a building from the 1st–2nd centuries. Other excavations revealed the contemporary cemetery, characterised by burial in common, presumably family tombs. At the northern outskirts of Vipava, the habitation remains from the 1st to the 4th century take the form of masonry and wooden buildings. The last recorded Roman artefacts are stray finds from the second half of the 4th and the early 5th century. Keywords: Italy (Regio X), Vipava, Roman period, settlement, cemetery, family tombs https://doi.org/10.3986/9789610502586_06 GEOGRAFSKI IN HISTORIČNI ORIS Vipava leži ob vznožju planote Nanosa, na vzhod­nem obrobju Vipavske doline. Naselje označujejo številni izviri reke Vipave in impozantna veduta z ru­ševinami Starega gradu v ozadju. Klimatsko in reliefno ugodno nižinsko območje je primerno za poljedelstvo, dobro uspeva vinska trta. Planota Nanosa je bila nekdaj primerna za pašništvo in druge dopolnilne gospodarske dejavnosti (les, lov, oglarstvo, smola). Plovnost reke Vi­pave do sedaj še ni bila dokazana. Ime Vipava (lok. Trg, nem. Wippach, it. Vipacco) etimološko ni pojasnjeno, ena izmed več možnosti je, da izhaja iz latinskega oseb­ 1 nega imena Vippius. Antično ime kraja ni poznano. Naselje leži ob izvi­rih reke Vipave, ki je na Tabuli Peutingeriani označena kot fl. Frigidus.2 V geografskem pogledu antičnih piscev so na ta prostor segale Julijske Alpe, na najnižjem delu Alp pa je bil prelaz Okra, čez katerega so v poznem re­publikanskem obdobju vodile trgovske poti iz Akvileje in iz Tergesta do Navporta.3 Po Šašlovi utemeljitvi lahko Okro enačimo s pogorjem Nanosa, ki sega od Vipave do Šmihela. Ta prostor naj bi poseljevali Karni, ki so pri Okri mejili na Japode.4 Vipavska dolina je verjetno že v prvih dveh deset­letjih 1. st. pr. n. št pripadala provinci Galiji Cisalpini. Po bitki pri Filipih, leta 42 pr. n. št., je bila Cisalpina priključena Italiji in domnevno se je takrat vzhodna meja Italije premaknila na reko Rižano.5 Po avgustejski raz­delitvi Italije na upravne regije je akvilejski ager obsegal območje Furlanske nižine od Tilmenta (it. Tagliamento)do Devina, severozahodni del Krasa (Šempolaj, Praprot), Vipavsko dolino, čez Razdrto je segal do Bevk pri Vrhni­ki, kjer je mejil na upravno območje kolonije Emone.6 Po Dioklecijanovi upravni reformi v 4. stoletju je območje 10. regije spadalo v provinco Venetia et Histria. ZGODOVINA RAZISKAV Glavni strokovnjaki, zadolženi za arheološko dediščino na območju Vipavske doline od sredine 19. st. dalje so bili dopisniki dunajske Centralne komisije (k. k. Central-Commission zur Erforschung und Erhaltung der kunst-und historischen Denkmale): Karl Moser je leta 1891 poročal, da so bili pri Vipavi, na travniku z ledinskim imenom Police, najdeni odlomki žar in rimski novci, grobišče je nameraval raziskati.7 Josip Mantuani je 1 Snoj 2009, 457. 2 Tab. Peut. III 5. 3 Strabon IV 6, 10; VII 5, 2. Zbrani in prevedeni odlomki: Horvat, Bavdek 2009, 19-22. 4 Šašel 1974, 9–17; Horvat, Bavdek 2009, 21. 5 Zaccaria 2007, 130; Šašel Kos, 2000, 281. 6 Šašel Kos 2000; 2002a; 2002b, 251, sl. 3. 7 Moser 1891, 35. GEOGRAPHIC AND HISTORICAL OUTLINE Vipava lies at the foot of Mt. Nanos, on the eastern fringes of the Vipava Valley. It is set among the numerous karst springs of the River Vipava against the backdrop of the ruins of an old castle (Stari grad) perched on a spur above. The climate and terrain of the flatland to the west are suitable for agriculture, also for growing grape vines. Mt. Nanos rising to the east was once exploited for pasturing and other complementary activities (forestry, hunting, charcoal and resin production). The River Vipava is not used for river traffic today and we have no evidence to suggest it was navigable in the past. Neither is the etymology of the name Vipava (locally known as Trg, Wippach in German and Vipacco in Italian) clear; it has been suggested it may originate in the Latin name of Vippius, among other possibilities.1 The name of the ancient settlement is unknown. It was located at the springs of the River Vipava marked as fl. Frigidus on the Tabula Peutingeriana.2 The ancient geographers considered that the area formed part of the Julian Alps, with the pass known as Ocra located in their lowest part. In the Late Republic, caravan routes from Aquileia and Tergeste led across this pass and continued towards Nauportus.3 According to Šašel, the name Ocra pertains not only to the pass, but to the whole range of Mt. Nanos that extended from Vipava to Šmihel. It was an area inhabited by the Carni, a tribe that bordered the Iapodes at Ocra.4 The Vipava Valley formed part of the province of Cisalpine Gaul probably as early as the beginning of the 1st cent. BC. It was included into Italy after the Battle of Philippi in 42 BC; this is probably the occasion when the eastern border of Italy moved to the River Rižana.5 Following the Augustan regional division of Italy, the territory of Aquileia as part of regio X comprised the Friuli plain from the River Tagliamento to Duino/Devin, the north­western part of the Carso/Kras, the Vipava Valley, reached across the Razdrto Pass to Bevke near Vrhnika, where it bordered the territory of Emona.6 This administrative arrangement changed in the 4th century, when Diocletian’s reforms included regio X into the province of Venetia et Histria. HISTORY OF RESEARCH The main experts and those responsible for the archaeological heritage of the Vipava Valley in the 1 Snoj 2009, 457. 2 Tab. Peut. III 5. 3 Strabo IV 6, 10; VII 5, 2. 4 Šašel 1974, 9–17; Horvat, Bavdek 2009, 21. 5 Zaccaria 2007, 130; Šašel Kos 2000, 281. 6 Šašel Kos 2000; 2002a; 2002b, 251, Fig. 3. leta 1912 poročal, da so bili v Vipavi najdeni poškodova­ni grobovi.8 Alberto Puschi je domneval rimsko utrdbo na mestu poznejšega srednjeveškega gradu.9 Leta 1960 so v Vipavi naključno odkrili zakladno najdbo novcev iz druge polovice 3. st.10 V šestdesetih letih prejšnjega stoletja je bil ustanovljen Zavod za spomeniško varstvo v Novi Gorici, pod okriljem katerega je prostor Vipave raziskovala arheologinja Nada Osmuk s sodelavci.11 Leta 2005 je bil izkopan del grobišča na Laurinovi ulici.12 Na najdišču Bela in Grublje je bil del rimskodobne naselbine raziskan leta 2008.13 Leta 2011 je bila raziskana parcela pod Starim gradom,14 v letu 2017 je bila odkrita rimska hiša na Beblerjevi ulici.15 PRAZGODOVINA Prazgodovinska naselbina se domneva na vzpetini Stari grad (sl. 1: 3) nad Vipavo, potrjujeta jo srp iz bro­naste dobe in lončenina, odkrita ob strojnem izkopu pobočja na spodnji grajski terasi.16 Pri širitvi starega kamnoloma, na lokaciji Za Tabrom (sl. 1: 12), so leta 1912 odkrili skeletne grobove iz železne dobe. Ohranili so se prstan, en manjši in dva para velikih bronastih uhanov ter kosti.17 Kamnito orodje in prazgodovinska keramika so bili najdeni tudi med izkopavanji na naj­dišču Bela in Grublje (sl. 1: 1,2), na parceli Tomažič (sl. 1: 4) in na rimskodobnem grobišču v Laurinovi ulici (sl. 1: 5).18 Marchesetti je domneval naselbino tudi na Gradišču pri Vipavi, vendar je dosedanje raziskave še niso potrdile.19 8 Najdišče Na Plazu (Mantuani 1912; 1913). 9 Puschi 1902, 135. 10 Pegan 1967, 207; Kos 1986, 124–125, 132; FMRSl I, 69–73. 11 Raziskave na arheološkem najdišču Bela in Grublje v letih 1984 (Osmuk 1985) in v letih od 2003 do 2005 (Osmuk 2000–2004; 2005a); pod Starim gradom leta 2000 (Osmuk 1999). 12 Osmuk 2005b; Tratnik 2014. 13 Vodja raziskav Patricija Bratina (Tratnik 2012). 14 Fabec, Tratnik, Vinazza 2012. 15 Gruden 2018. 16 Bratina 2010, 174; Osmuk 1999. 17 V Arheoloških najdiščih Slovenije je najdišče dvakrat navedeno. Napačno pod imenom Podnanos (Petru S. 1975, 126) in pravilno: Vipava (Petru P. 1975, 126). Glej Mlinar 1999, 14, t. 29: 1–8; Mantuani 1912; 1913, 26. Najdbe hrani Narodni muzej v Ljubljani (inv. št. P 13023–13030). 18 Tratnik 2012, t. 1: 1; 2014, 284–285 in op. 94. 19 Vzpetina na južnem obrobju Vipave (Marchesetti 1903, 91; Bratina 2010, 169, št. 20). mid­19th century were correspondents of the Austro­Hungarian k. k. Central-Commission zur Erforschung und Erhaltung der kunst-und historischen Denkmale in Vienna. One of them was Karl Moser, who in 1891 noted that fragments of urns and Roman coins were discovered near Vipava, near a meadow known as Po­lice; he intended to investigate the site.7 In 1912, Josip Mantuani reported on damaged graves from Vipava.8 Alberto Puschi posited a Roman fort on the hill of the later medieval castle (Stari grad).9 Much later, in 1960, a hoard of Roman coins from the second half of the 3rd century was found by chance in Vipava.10 The 1960s is also the time when the regional office of the heritage protection institute was founded in Nova Gorica (here­inafter ZVKDS OE Nova Gorica), which was thereafter charged with investigations in the Vipava area under the leadership of Nada Osmuk and her colleagues.11 They excavated part of the cemetery under the street of Laurinova ulica in 2005.12 Later investigations revealed part of the Roman settlement at the Bela and Grublje sites in 2008,13 part of the area below Stari grad in 201114 and, most recently in 2017, a Roman house at the Beblerjeva street.15 PREHISTORY It is presumed that the prehistoric settlement at Vipava lay on the hill of Stari grad. Its existence is supported by a Bronze Age sickle and pottery sherds that came to light during machine excavations on the slope of the lower terrace below the medieval castle.16 Inhumation burials from the Iron Age were found in 1912 while extending the ancient stone quarry at Za Tabrom. Only a finger ring, one small and two pairs of large bronze earrings were kept alongside the bones, the rest discarded.17 Stone tools and prehistoric pottery 7 Moser 1891, 35. 8 The Na Plazu site (Mantuani 1912, 282; 1913). 9 Puschi 1902, 135. 10 Pegan 1967, 207; Kos 1986, 124–125, 132; FMRSl I, 69–73. 11 Investigations at the Bela and Grublje sites in 1984 (Osmuk 1985) and again in 2003–2005 (Osmuk 2000–2004; 2005a); investigations below Stari grad in 2000 (Osmuk 1999). 12 Osmuk 2005b; Tratnik 2014. 13 Investigations led by Patricija Bratina; Tratnik 2012. 14 Fabec, Tratnik, Vinazza 2012. 15 Gruden 2018. 16 Bratina 2010, 174; Osmuk 1999. 17 The finds are mentioned twice in Arheološka najdišča Slovenije, erroneous at site of Podnanos (Petru S. 1975, 126) and correct at Vipava site (Petru P. 1975, 126). Mlinar 1999, 14, Pl. 29: 1–8. Mantuani 1912; 1913. Small finds kept in the National Museum of Slovenia in Ljubljana (Inv. Nos. P 13023–13030). 1 2 5 6 11 9 3 4 8 7 10 12 ARHEOLOŠKI SLEDOVI RIMSKE DOBE Trško jedro Vipave je skromno raziskano. Strnje­no rimsko poselitev, ki so jo zamejevale struge izvirov in manjših pritokov reke Vipave, na podlagi zbranih podatkov domnevamo ob vznožju Starega gradu, na lokacijah Na hribu in na Trgu Pavla Rušta. Posamične Sl. 1: Vipava. Obravnavana arheološka najdišča. M. = 1:10.000. (Dopolnjeno po: Tratnik 2014, sl. 1). (Osnova: TTN 5, 2004, © Geodetska uprava RS) Fig. 1: Vipava. Map of the Vipava area with marked sites mentioned in the text. Scale 1:10,000. (Supplemented after: Tratnik 2014, fig. 1). 1 – Ob Beli; – 2 Grublje; – 3 Stari grad; – 4 Tomažič; – 5 La­urinova ulica; – 6 Zdravstveni dom; – 7 Grabrijanova ulica; – 8 Parc. št. 2510 in 1511, k. o. Vipava; – 9 Police – 10 Glavni trg; – 11 Beblerjeva ulica 1 (Župnišče); – 12 Za Tabrom; – 13 Stara gora (Gradišče) were also found during the investigations at the Bela and Grublje sites (Fig. 1: 1,2), Tomažič site (Fig. 1: 4), as well as at the Roman­period cemetery at Laurinova ulica (Fig. 1: 5).18 Marchesetti suggested that another prehistoric settlement stood on Gradišče pri Vipavi, a hill at the southern edge of Vipava, but its existence has not been confirmed.19 ROMAN PERIOD Vipava is poorly investigated. Based on available archaeological evidence, we presume that the centre of the Roman settlement was located at the foot of Stari grad (the Na hribu and Trg Pavla Rušta sites), in the area delimited by the springs and tributaries of the River Vipava. Stray Roman finds have also come to light on the slopes of Stari grad (Fig. 1: 4),20 during earthworks at Grabrijanova ulica (Fig. 1: 7)21 and on other lots (Fig. 1: 8).22 A layer with Roman­period structures and finds (stone floor, pottery sherds, animal bones) was recorded in the house at Glavni trg 2 (Fig. 1: 10).23 Part of a Roman house presumably from the 1st–2nd centuries and fitted with mortar floor and plastered walls was excavatedat Župnišče (Beblerjeva street, Nr. 1; Fig. 1: 11).24 The summit of Stari grad has thus far only yielded stray finds that tell very little of the nature of human presence there (Fig. 1: 3).25 The cemetery associated with the Roman settle­ment probably extended between the church of St Stephen and the Police site, possibly even further. Graves from the 1st and 2nd centuries were excavated at 18 Tratnik 2012, Pl. 1:1; 2014, 284–285 and Fn. 94. 19 Marchesetti 1903, 91; Bratina 2010, 169, No. 20. 20 Osmuk 1999, 142; Republican as and a coin from the 4th century (FMRSl IV, 36); sherds of Early Roman amphorae (Fabec, Tratnik, Vinazza 2012, 34, Add. 6, Pl. 4: 1-4).21 Tratnik 2014, 256. 22 Tratnik 2014, 256. 23 Tratnik 2014, 256. 24 Gruden 2018. 25 Osmuk 1999. rimske najdbe so bile odkrite ob raziskavah na pobočju Starega gradu20 (sl. 1: 4), ob delih na Grabrijanovi ulici21 (sl. 1: 7) in na posameznih parcelah22 (sl. 1: 8). Plast z rimskodobnimi najdbami (kamnit tlak, odlomki lon­čenine, živalske kosti) je bila odkrita ob delih v hiši na Glavnem trgu 223 (sl. 1: 10). Del rimske hiše, domnevno iz 1. in 2. st., z estrihom in ometanimi stenami, je bil izkopan v nekdanjem župnišču na Beblerjevi ulici 1 (v nadaljevanju Župnišče; sl. 1: 11).24 Rimska poselitev na vrhu Starega gradu še ni potrjena, poznane so posamične najdbe (sl. 1: 3).25 Grobišče je verjetno obsegalo vsaj območje med današnjo cerkvijo sv. Štefana in ledino Police. Grobovi iz 1. in 2. st. so bili odkriti na Laurinovi ulici26 (sl. 1: 5), odlomki žar in rimski novci pa so bili najdeni na ledini Police27 (sl. 1: 9). Na severnem obrobju Vipave, na najdiščih Bela in Grublje (sl. 1: 1,2) je drugo poselitveno območje, kjer je dokazana kontinuiteta od 1. do 4. st. Tu domnevamo manj strnjeno poselitev z zidanimi in lesenimi stavbami ter go­spodarskimi objekti.28 Na južnem obrobju Vipave, na Stari gori (sl. 1: 13) so bile odkrite posamične rimske najdbe.29 TRASA RIMSKE CESTE Rimska cesta v Vipavi še ni odkrita. Po terenskih opažanjih arheologov Nade Osmuk in Božidarja Slap­šaka je rimska cesta od Vipave do Razdrtega vodila višje kot današnja magistralna cesta. Južno od Vipave naj bi se začela vzpenjati po pobočju Nanosa, od Gradišča pri Vipavi, mimo cerkvice sv. Nikolaja, čez Barnice za Gra­diščem nad Hraščami proti Gradišču nad Razdrtim.30 Odsek trase rimske ceste z delno ohranjenimi kolesni­cami je bil odkrit na Sušču pod Razdrtim.31 20 Raziskave po gradnji gozdarske poti (Osmuk 1999; republikanski as in novec iz 4. st. (FMRSl IV, 36); odlomki zgodnjerimskih amfor na parceli Tomažič (Fabec, Tratnik, Vinazza 2012, 34, priloga 6, t. 4: 1-4. 21 Neobjavljeno, Nada Osmuk, dokumentacija v arhivu ZVKDS OE Nova Gorica, mapa Vipava. Opomba: Tratnik 2014, 256. 22 Glej op. 21. 23 Glej op. 21. 24 Gruden 2018. 25 Osmuk 1999. 26 Tratnik 2012. 27 Moser 1891. 28 Tratnik 2012. 29 Osmuk 1985. 30 Slapšak 1998. 31 Svoljšak 2003. Laurinova ulica (Fig. 1: 5).26 At Police, urn sherds and Roman coins were found towards the end of the 19th century (Fig. 1: 9).27 Another centre of habitation has been recorded on the northern outskirts of Vipava, at Bela and Grublje (Fig. 1: 1,2). Evidence suggests it was a less densely in­habited area of masonry and wooden buildings, as well as outhouses,28 that was continuously occupied from the 1st to the 4th century. Stray Roman finds also came to light on the southern outskirts of Vipava, at Stara gora (Fig. 1: 13).29 ROMAN ROADS No traces of the Roman roads in Vipava have as yet been found. The field surveys by Nada Osmuk and Božidar Slapšak have suggested that the road from Vipava towards Razdrto led higher than the modern main road. It presumably began its ascent east of Vipava and led along the foot of Mt. Nanos from Gradišče pri Vipavi, past the church of St Nicholas, across Barnice east of Gradišče nad Hraščami towards Gradišče nad Razdrtim.30 A section of the Roman road with partially surviving wheel ruts was unearthed at Sušec below Razdrto.31 HOUSING Part of a Roman house was investigated atŽupnišče, in Beblerjeva ulica (Fig. 1: 11; 2). Excavations uncovered a paved court with a stone­framed hearth. It was associated with a building to the north, slightly raised above the level of the court. The house had a mortar floor. One of its rooms was partitioned with a brick wall in opus spicatum. There was a semicircular corner niche built against the partition wall. The walls were plastered. The house dates to the 1st–2nd century.32 CEMETERY Part of the Roman cemetery was excavated at Lau­rinova ulica (Fig. 1: 5; 3). The nine investigated tombs span from the 1st to the second half of the 2nd century.33 The anthropological analysis has shown that the deceased were cremated (crematio) on a common pyre 26 Tratnik 2012. 27 Moser 1891. 28 Tratnik 2012. 29 Osmuk 1985. 30 Slapšak 1998. 31 Svoljšak 2004. 32 Gruden 2018. 33 Tratnik 2014. Sl. 2: Vipava. Pogled na rimsko hišo na Beblerjevi ulici 1 (Župnišče) (foto: Tilen Podobnik). Fig. 2: Vipava. The Roman house at Župnišče, Beblerjeva ulica 1 (photo: Tilen Podobnik). POSAMEZNI OBJEKTI V NASELJU V Župnišču, na Beblerjevi ulici 1 je bil raziskan del rimske hiše (sl. 1: 11; 2). Večji del izkopanega območja je zavzemalo tlakovano dvorišče z ognjiščem, ki je imelo kamnit obod. Na severnem robu je stala stavba, ki je bila nekoliko dvignjena nad nivo dvorišča. Tlakovana je bila z estrihom in en prostor je bil predeljen z opečnato steno, grajeno v tehniki ribje kosti. Stene so bile ometane. Ob predelni steni je bila zidana polkrožna niša. Stavba je datirana v 1. in 2. st. 32 GROBIŠČE Del rimskega grobišča (9 grobov) je bilo izkopanih na Laurinovi ulici (sl. 1: 5; 3), ob cerkvi sv. Štefana. Gro­bovi so datirani v čas od 1. st. do druge polovice 2. st.33 Opravljene so bile antropološke in arheozoološke analize ostankov. Analiza grobov z Laurinove ulice je pokazala, da so umrle sežigali (crematio) na skupni grmadi (ustrina). Del žganih kosti je bil shranjen v žare skupaj s pepelom in ogljem z grmade. Kot žara so bili uporabljeni lonci kuhinjske keramike, v enem primeru 32 Gruden 2018. 33 Tratnik 2014. (ustrina). Part of the cremated bones was then col­lected into urns together with the ash from the pyre. They used kitchenware jars as urns, in one case more precisely an Auerberg type jar. Some of the urns were covered with a small stone slab, a sherd of an amphora or a thin­walled cup. Many of the graves also held ash and cremated bones outside the urn, strewn on the bot­tom of the grave pit, or in another vessel in the grave. The cremated bones included those of animals, which are seen as traces of the burial ritual. Archaeozoologi­cal analyses have shown that most belonged to sheep or goats and cattle. The composition of the grave goods is quite uniform; an individual burial in an urn was associ­ated with a drinking set consisting of a jug and a cup or beaker. Coins were also placed in the graves. The deceased were buried either in a cist composed of stone slabs, of tegulae or a combination of tegulae and stone slabs, under an amphora with the top cut off or in simple pits covered with a tegula. A particular feature of the cemetery is ‘family tombs’, with two or more burials in the same tomb. One such tomb contained three burials laid one beside or on top of the other and spanning two centuries (Fig. 3).34 34 Tratnik 2014. lonec tipa Auerberg. Nekaj žar je bilo pokritih: ali z manjšo kamnito ploščo, kosom ostenja amfore ali sko­delico tankih sten. Pogosto se je nekaj žganine in žganih kosti nahajalo tudi zunaj žar, na dnu grobne skrinje ali v drugih posodah v grobu. Med žganimi kostmi so bile tudi posamezne živalske kosti, ki jih razlagamo kot sled pogrebnih obredov. Največ kosti pripada drobnici in govedu. Sestava grobnih pridatkov je zelo enotna. K posameznemu pokopu v žari je bilo pridano pivsko posodje: en vrč s skodelico ali čašo. V grobove so bili pridani tudi novci. Grobna arhitektura je bila različna: pokop v skri­nji iz kamnitih plošč, v skrinji sestavljeni iz tegul ali v kombinaciji tegul in kamnitih plošč. Drugi način je žarni pokop pod amforo z odrezanim zgornjim delom, tretji način pa je grob z žaro v preprosti grobni jami, pokrit s tegulo. Značilnost grobišča so “družinski grobovi” z dvema ali več pokopi v isti grobni skrinji. Odkrita je bila tudi družinska grobnica s tremi grobovi, ki so ležali tik ob in drug na drugem, v katero so pokopavali dve stoletji34 (sl. 3). NAJDBE Obravnava odkritih predmetov z grobišča in naselbine na lokaciji Grublje je pokazala, da v zgodnje cesarskem obdobju prevladuje posodje, ki je značilno za severovzhodno Italijo in severni Jadran. Navadno namizno posodje in skodelice tankih sten so bile naj­verjetneje delane v akvilejskih delavnicah, amfore so v tem obdobju večinoma severnojadranskega izvora. Pri­sotni so posamezni importi iz vzhodnega Sredozemlja, datirani v 2. in 3. st. Med uvoženimi izdelki s konca 3. in iz 4. st. prevla­dujejo importi iz severne Afrike. V drugi polovici 4. st. se na najdiščih Bela in Grublje izrazito poveča število novčnih najdb. Na ustju amfore z grobišča v Laurinovi ulici je žig C. Lekanija Basa,35 na fibuli tipa Aucissa pa žig z napisom [P] VALE.36 Na odlomku tegule z najdišča Grublje je 37 delno ohranjen žig …ARNI. DRUŽBA Konkretnih podatkov o družbi in družbenem statusu posameznikov iz Vipave nimamo. Na odkritem grobišču se odraža romanizirani sloj prebivalcev, posa­mezni elementi, npr. pokop v grobno skrinjo iz kamnitih plošč, pa se interpretirajo kot sled staroselske tradicije. 34 Tratnik 2014. 35 Tratnik 2014, 280, 281, 298, t. 18: 145. 36 Tratnik 2014, 275, 297, t. 22: 198. 37 Tratnik 2012, 119, sl. 13. SMALL FINDS The analysis of the artefacts recovered from the cemetery and from the settlement at the Grublje site shows that most of those from the Early Imperial pe­riod consist of pottery characteristic of north­eastern Italy and the northern Adriatic. The plain tableware and thin­walled cups were presumably produced in Aquileian workshops, while the amphorae mainly came from the northern Adriatic. The artefacts also include imports from the eastern Mediterranean, dated to the 2nd and 3rd centuries. The imports from the end of the 3rd and the 4th century mainly comprise objects from northern Africa. At Bela and Grublje, there is a marked increase of coin finds in the second half of the 4th century. Some of the products are stamped. The rim of an amphora found in the cemetery at Laurinova ulica bears the stamp of C. Laecanius Bassus,35 an Aucissa brooch has a stamped inscription [P] VALE,36 while a fragment of a tegula from Grublje bears a partially surviving stamp of …ARNI.37 SOCIETY We have no concrete information on the society and social status of the people living in Roman Vipava. The finds from the cemetery reflect a Romanised popula­tion, while certain elements, for example burial in cists of stone slabs, are seen as traces of the indigenous tradition. 35 Tratnik 2014, 280, 281, 298, Pl. 18: 145. 36 Tratnik 2014, 275,297, Pl. 22: 198. 37 Tratnik 2012, 119, Fig. 13. FUNKCIJA IN STATUS NASELJANA OSNOVI ARHEOLOŠKIH VIROV Za rimsko naselbino v Vipavi nimamo dovolj podatkov, da bi lahko utemeljeno sklepali o njenem statusu. Na podlagi dosedanjih arheoloških raziskav domnevamo, da se je naselbina razvila na temeljih prazgodovinske poselitve, na ugodni geografski legi in ob pomembni prometni povezavi. Odkrita stavba z ometom, maltnim estrihom in ognjiščem morda priča tudi o višjem standardu življenja posameznikov in ustreznem poznavanju gradbenih tehnik. Del raziskane naselbine Bela in Grublje, na severnem obrobju Vipave, z odkritimi lesenimi in z nekaj zidanimi objekti pa kaže bolj podeželski značaj. RAZVOJ NASELBINE Na poselitev Vipave v 1. st. pr. n. št. oz. v avgustejski dobi kažejo posamične najdbe novcev in zgodjerimskih amfor, odkritih na pobočju Starega gradu in v plasti rečnega nanosa na Laurinovi ulici.38 V prvem stoletju je bila zgrajena nova državna cesta čez Hrušico,39 ki je obšla Vipavo, domnevamo pa, da je stara, daljša trasa ceste čez Razdrto še vedno ostala v uporabi. Domnevno v tem času so bile v Vipavi zgrajene posamezne stavbe, ki ustrezajo višjemu bivalnemu standardu.40 Iz 1. in 2. st. je tudi grobišče na Laurinovi ulici. Na severnem obrobju Vipave (Bela in Grublje) so poselitveni ostanki datirani od 1. do konca 4. st. Najmlajše posamične rimske najdbe so datirane v drugo polovico 4. st. in morda še na začetek 5. st.,41 v Vipavi je bil najden tudi Justinjanov zlatnik.42 O nadaljevanju rabe prostora v zgodnjem srednjem veku priča najdba bronastega pasnega okova avarskega izvora, z najdišča Grublje.43 Stari grad: republikanski as, datiran v prvo polovico 2. st. pr. n. št. (FMRSl V, 36. št. 15); odlomki amfor Lamboglia 2/ Dressel6A (Fabec, Tratnik, Vinazza 2012, 34, priloga 6, t. 4: 1­4). Laurinova ulica: mali keltski srebrnik, Tip Eis (Alenka Miškec, Novčne najdbe, dodatek v: Tratnik 2014 [str. 291]. 39 Vedaldi Iasbez 1994, 72–73: izgradnja v avgustejskem obdobju, Rufius Festus, Brev. 7.40 Stavba na Beblerjevi ulici (Župnišče), 1. in 2. st. 41 Odlomki amfor, novci: Tratnik 2012, 113, 121,131. 42 FMRSl I, 69. 43 Tratnik 2012, 120, sl. 15, t. 7: 7. FUNCTION AND STATUS OF THE SETTLEMENT The evidence on this subject is scarce. We may presume that the Roman settlement developed as a con­tinuation of the prehistoric habitation in a geographi­cally advantageous location and along an important lineof communication. The house excavated at Župnišče, with plastered walls, mortar floors and a hearth may be the lodgings of a relatively affluent individual with the knowledge of the appropriate Roman building techniques. In contrast, the part of the settlement in­vestigated at Bela and Grublje, with wooden and only few masonry buildings, shows a more rural character. DEVELOPMENT OF THE SETTLEMENT Stray finds of coins and pottery from the slopes of Stari grad and the alluvial layer at Laurinova ulica show that Vipava was already settled in the 1st century BC.38 In the 1st century AD, a new main road was constructed that led across the Hrušica Pass39 and avoided Vipava, but we presume that the old road across the Razdrto Pass remained in use. In this time, buildings were put up that reveal a high standard of living.40 Also from the 1st and 2nd centuries is the burial grounds at Laurinova ulica. The habitation remains on the northern outskirts of Vipava (Bela and Grublje) span from the 1st to the end of the 4th century. The last of the stray finds date to the second half of the 4th or possibly the early 5th century;41 there was also a gold coin of Justinian that was found in Vipava.42 The continued human presence in the medieval period is attested by a bronze belt fitting of Avar origin, which was recovered at the Grublje site.43 Translation: Andreja Maver 38 Republican as dated to the first half of the 2nd century BC (FMRSl V, 36, No. 15); fragments of Lamboglia 2/Dressel 6A amphorae (Fabec, Tratnik, Vinazza 2012, 34, Add. 6, Pl. 4: 1-4); Laurinova ulica: small Celtic silver coin of the Eis type (Alenka Miškec, Novčne najdbe, Supplement in: Tratnik 2014, p. 291). 39 Vedaldi Iasbez 1994, 72-73, construction dated in Au­gustan age, Rufius Festus, Brev. 7. 40 Building at Beblerjeva street (Župnišče), 1st and 2nd centuries. 41 Amphorae sheards, coins; Tratnik 2012, 113, 121, 131. 42 FMRSl I, 69. 43 Tratnik 2012, 120, Fig. 15, Pl. 7: 7. BRATINA, B. 2010, Nova arheološka odkritja v Vipavski dolini. – Goriški letnik 33-34/1, 155-180. FABEC, T., V. TRATNIK, M. VINAZZA 2012, Poročilo o opravljenih arheoloških izkopavanjih na območju predvidene gradnje stanovanjske hiše Kunaver Tomažič. – Poročilo / Re­port; ZVKDS, Center za preventivno arheologijo, Ljubljana (neobjavljeno / unpublished). FMRSl I = P. Kos, Die Fundmünzen der römischen Zeit in Slowenien I, 68-69, Berlin 1988. FMRSl IV = A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien V, 36, Berlin 1998. FMRSl V = A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien 5, Mainz 2004. HORVAT, J., A. BAVDEK 2009, Okra. Vrata med Sredozemljem in Srednjo Evropo / Ocra. The gateway between the Mediter­ranean and Central Europe. – Opera Instituti Archaeologici Sloveniae 17. GRUDEN, G. 2018, Vipava – arheološko najdišče ob zdravstve­nem domu. Vipava – trško jedro. – Varstvo spomenikov. Poročila 53, 283-284. KOS, P. 1986, The Monetary Circulation in the Southeastern alpine Region ca. 300 B. C. – A. D. 1000 / Denarni obtok na prostoru jugovzhodnih Alp 300 pr. n. š. – 1000. – Situla 24. MANTUANI J. 1912, Tätigkeitsbericht. - Mitteilungen der K.K. Zentral-Kommissions für Denkmalpflege, Bd. 11, Nr. 12, 282. MANTUANI J. 1913, Poročilo deželnega muzeja Rudolfinum v Ljubljani za leti 1922 in 1912, 26 (št. 9./VII). MARCHESETTI, C. 1903, I Castellieri preistorici di Trieste e della regione Giulia. – Trieste. MLINAR, M. 1999, Vipavska dolina v prazgodovini. – Diplom­sko delo / BA thesis. Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). MOSER, K. 1891, Funde im Wippachthale. – Mittheilungen der Anthropologischen Gesellschaft in Wien 21, Sitzungs­berichte, 35. OSMUK, N. 1985, Stara gora (Vipava). – Varstvo spomenikov 27, 263. OSMUK, N. 1999, Vipava. Stari grad. – Varstvo spomenikov. Poročila 38 (2001), 142. OSMUK, N. 2000–2004, Vipava ­ arheološko najdišče Bela in Grublje. – Varstvo spomenikov. Poročila 39–41 (2006), 235–236. OSMUK, N. 2005a, Vipava – arheološko najdišče Bela in Grublje. – Varstvo spomenikov. Poročila 42 (2006), 183. OSMUK, N. 2005b, Vipava. Laurinova ulica. – Varstvo spome­nikov. Poročila 42 (2006), 184–185. PEGAN, E. 1967, Najdbe novcev v Sloveniji. – Arheološki vestnik 18, 203–217. PETRU, P. 1975, Vipava. – V / In: Arheološka najdišča Slovenije, 126, Ljubljana. PETRU, S. 1975, Podnanos. – V / In: Arheološka najdišča Slovenije,126, Ljubljana. PUSCHI, A. 1902, I valli romani delle Alpi Giulie. – Archeo­grafo Triestino. Nuova serie, Vol. 24, Supplemento, 119-150. SLAPŠAK, B. 1998, Poročilo o dodatnih arheoloških raziskavah za odločanje o morebitni racionalizaciji trase hitre ceste M 10-7 odsek 0374 Razdrto – Selo na širšem področju varovanja območja kulturne in naravne dediščine Podnanos – Gradišče nad Hraščami, 1998. – Raziskovalno poročilo / Report; Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). SNOJ, M. 2009, Etimološki slovar slovenskih zemljepisnih imen. – Ljubljana. SVOLJŠAK, D. 2003, Sušec pri Razdrtem. – V: B. Djurić et al. (ur.), Zemlja pod vašimi nogami. Arheologija na avtocestah Slovenije. Vodnik po najdiščih, 262-264. SVOLJŠAK, D. 2004, Sušec pri Razdrtem. – In: B. Djurić et al. (eds.), The Earth Beneath Your Feet. Archaeology on the Motorways in Slovenia. Guide to Sites, 270-272, Ljubljana. ŠAŠEL, J. 1974, Okra. – Kronika 22, 9–17. ŠAŠEL KOS, M. 2000, Caesar, Illyricum, and the hinterland of Aquileia. – V / In: G. Urso (ur. / ed.), L'ultimo Cesare. Scritti, riforme, progetti, poteri, congiure. Atti del convegno internazionale, Cividale del Friuli, 16-18 settembre 1999, Monografie / Centro ricerche e documentazione sull‘ant. class. 20, 277–304, Roma. ŠAŠEL KOS, M. 2002a, The boundary stone between Aquileia and Emona / Mejnik med Akvilejo in Emono. – Arheološki vestnik 53, 373–382. ŠAŠEL KOS, M. 2002b, Il confine nord­orientale dell›Italia romana. Riesame del problema alla luce di un nuovo do­cumento epigrafico. – Aquileia nostra 73, 245-260. TRATNIK, V. 2012, Sledovi rimskodobne poselitve na Gru­bljah pri Vipavi (Traces of Roman settlement at Grublje near Vipava). – Arheološki vestnik 63, 105–138. TRATNIK, V. 2014, Rimsko grobišče na Laurinovi ulici v Vi­pavi (The Roman cemetery at Laurinova ulica in Vipava). – Arheološki vestnik 65, 255-322. ZACCARIA, C. 2007, Tra Natisone e Isonzo. Aspetti am­ministrativi in etá romana. – V / In: V. Chiaba, P. Maggi, C. Magrini (ur. / eds.), Le Valli del Natisone e dell'Isonzo tra Centroeuropa e Adriatico, Studi e ricerche sulla Gallia Cisalpina 20, 129–144. Vesna Tratnik Narodni muzej Slovenije Prešernova cesta 20 SI­1000 Ljubljana vesna.tratnik@nms.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 123–140 GRADIŠČE NAD KNEŽAKOM Boštjan LAHARNAR, Edisa LOZIĆ, Alenka MIŠKEC Izvleček Številne rimske drobne najdbe in osvetlitev kompleksne in strnjene pozidave z arheološko interpretacijo lidarskih podatkov nakazujejo, da je bilo Gradišče nad Knežakom v rimski dobi zelo obljudeno. Načrt in oblika naselbinskega rastra dopuščata domnevo o obstoju naselbine urbanega videza. V tem hipu ni jasen odnos Gradišča do morebitne antične arhi­tekture ob poti neposredno pod njim, pomen in moč naselbine v pozni rimski dobi pa poudarja tudi gradnja obrambnega stolpa na bližnji vzpetini Reber. Kontinuiteta in intenzivnost poselitve Gradišča iz železne v rimsko dobo pričata o ekonomski moči prebivalcev, morda delno povezani z neko usmerjeno kmetijsko dejavnostjo. Na to nakazujejo številni sledovi intenzivne rabe tal v neposredni okolici. Podobno kontinuiteto iz železne v rimsko dobo ugotavljamo tudi na nekaterih drugih gradiščih v soseščini. Vendar v tej fazi raziskav ni dovolj jasna dinamika njihove poselitve v celotnem rimskem obdobju. V primeru Gradišča nad Knežakom številne drobne najdbe ne nasprotujejo domnevi, da so arhitekturni sledovi, prepoznani z analizo lidarskih podatkov, predvsem poznorimska gradnja iz druge polovice 3. in 4. st., toda med najdbami so tudi predmeti iz celotnega 1. in 2. st. Ključne besede: Italija (10. regija), Gradišče nad Knežakom, rimska doba, utrjene naselbine, lidarski podatki, drobne najdbe Abstract The multitude of Roman artefacts, as well as the complex and dense architectural remains unveiled by the analysis of the LiDAR data show that the hilltop site above the village of Knežak (Gradišče above Knežak) was intensely populated in the Roman period. The layout of the architectural remains even suggests an urban settlement. It is as yet unclear whether it was associated with possible Roman buildings below the hill, near the routeway, while a tower on the adjacent hill of Reber supports its importance in the Late Roman period. The intensity and continuity of settlement from the Iron Age into the Roman period speak of Gradišče’s economic prosperity, which may in part have been due to a specialised agricultural activity. This is suggested by the evidence of intensive land use in the surrounding landscape. A similar continuity has been observed on several other hillforts in the region though there is as yet insufficient evidence regarding the dynamics of settlement throughout the Roman period. For Gradišče above Knežak, there are small finds that date to the 1st and 2nd centuries, but the latest items are from the Late Roman period and suggest that the architectural remains also mainly date to this period, more precisely to the second half of the 3rd and the 4th century. Keywords: Italy (Regio X), Gradišče above Knežak, Roman period, fortified settlements – hillforts, ALS (Airborne LiDAR) data, small finds https://doi.org/10.3986/9789610502586_07 GEOGRAFSKO-PROMETNA LEGA Arheološko najdišče Gradišče nad Knežakom (tudi Gradišče, Gradišče Knežak, Pod studencem) leži jugozahodno od Knežaka, na vzpetini z ledinskim ime­nom Gradišče (742 m). Sodi v niz gradišč, ki se vrstijo od južnega zaledja Hruševja mimo Slavine in vzdolž Taborskega grebena, izrazite geografske ločnice med porečjema apnenčaste doline reke Pivke in flišne doline reke Reke, mimo Šembij vse do Trnovega in Sv. Ahaca pri Ilirski Bistrici.1 Naselbina leži ob južnem robu geološke posebnosti, t. i. knežaškega tektonskega okna, kjer krpa rodovitnega flišnega sveta vpada pod apnenčaste plasti in ustvarja strma, ponekod prepadna pobočja.2 Gostota in razvrstitev najdišč vzdolž Pivke do Ilirskobistriškega s kvarnerskim zaledjem jasno zarisuje osnovno pot, ki je povezovala tukajšnje naselbine. Ta se je pri prelazu Razdrto (Okra) odcepila od glavne prometnice čez t. i. Iliro-italska vrata proti jugu, v smeri Kvarnerja. V nepo­sredni bližini Knežaka se srečata še dve poti. Prvo, ki je povezovala Tergeste, Divačo, Pivško in prek Javornikov Loško dolino s porečjem Kolpe, je Jaroslav Šašel ime­noval japodska cesta. Druga je vodila iz smeri Ljubljane prek Javornikov na Pivško ter naprej v Kvarner.3 TOPOGRAFIJA NAJDIŠČA Gradišče nad Knežakom leži na enem od vrhov polagoma vzpenjajočega se pobočja, ki vodi iz smeriŠembij proti južnemu robu tektonskega okna pri Kne­žaku. Severni rob naselbine se tako naveže na strmo, po­nekod prepadno pobočje tektonskega okna. Obrambne gradnje v tem delu niso bile potrebne, zahodni, južni in vzhodni rob naselbine pa oklepajo ruševine kamnitega obzidja (sl. 1). Gradnja gradišč ob prepadnih strminah in postavitev obrambne arhitekture na lažje pristopnih smereh sta bili značilnosti tudi sosednjih naselbinskih najdišč Taborskega grebena, kot so Kerin nad Pivko,Šilentabor, Grmada, Gradišče na Čepni in Gradišče pri Šembijah.4 Osrednji del naselbine je trikotne oblike oz. oblike krožnega izseka in se deli na širšo severno (obseg 525 m; 1,39 ha) in ožjo južno (obseg 469 m; 0,68 ha) poselitveno teraso, ki ju deli notranji okop (sl. 1: a). V tem hipu ni mogoče z gotovostjo trditi, da je bila najprej poseljena severna terasa in se je naselbina razširila proti jugu po­zneje. To bi bilo možno, če je zunanji okop latenski, kot domneva Battaglia (glej Zgodovina raziskav). Podobno vprašanje se poraja glede ovalne utrditve in vključitve zahodnega območja (obseg 320 m; 0,6 ha); utrditev je od 1 Urleb 1975; Horvat 2005. 2 Pleničar 1959; Šebela 2005. 3 Šašel 1975, 75, 96; Ciglenečki 1985, 269–270; Gaspari 2020, v tej knjigi. 4 Guštin 1978, 102. SURROUNDING GEOGRAPHY AND COMMUNICATIONS Gradišče above Knežak (also Gradišče, Gradišče Knežak, Pod studencem) is an archaeological site SW of the village of Knežak, on an elevation known as Gradišče (742 m a.s.l). It is a hillfort located at the southern edge of a geologic feature known as the Knežak tectonic window, where a patch of fertile flysch is exposed under the limestone strata, creating steep and in places pre­cipitous slopes.1 It is one in a series of hillforts dotting the landscape from the southern hinterland of Hruševje, past Slavina and along Taborski greben, which is a ridge separating the limestone valley of the River Pivka fromthe flysch valley of the River Reka, then past Šembije to Trnovo and the hill of Sv. Ahac near Ilirska Bistrica.2 The density and distribution of sites here and further on in the hinterland of the Kvarner Bay clearly trace a line of communication. It forked off the main route through the Illyrian-Italic Gate at the Razdrto Pass and headed towards the Kvarner Bay. Two other routes met in the immediate vicinity of Knežak. One led from Tergeste, Divača, the Pivka area, across the Javorniki Hills to the Lož Valley and the Kolpa watershed; Jaroslav Šašel named it the ‘Iapodic route’. The other led from Ljubljana across the Javorniki Hills to the Pivka area and again onwards to the Kvarner Bay.3 SITE TOPOGRAPHY Gradišče above Knežak is sited on a spur of an evenlyrising terrain leading from Šembije towards the southern edge of the Knežak tectonic window. The northern edge of the hillfort leans against the steep and in places precipitous slope of the tectonic window, eliminating the need to built ramparts. In contrast, the hillfort’s western, southern and eastern perimeter received a stonework rampart. Building hillforts by using precipices and fortifying the rest of the more easily accessible sides with ramparts also character-ises the neighbouring sites on the Taborski greben, suchas Kerin above Pivka, Šilentabor, Grmada, Gradišče at Čepna and Gradišče near Šembije.4 The central part of the hillfort is triangular or seg­mental in plan. It is divided with an internal rampart (Fig. 1: a) into the wider and higher north (circumference 525 m; 1.39 ha) and the narrower and lower south terrace (cir­cumference 469 m; 0.68 ha). It is as yet not clear whether the north terrace was inhabited first and the settlement spread southwards later. This would be possible if the outer rampart were of Late Iron Age date, as Battaglia 1 Pleničar 1959; Šebela 2005. 2 Urleb 1975, 62–71; Horvat 2005. 3 Šašel 1975, 75, 96; Ciglenečki 1985, 275–276; Gaspari 2020, in this book. 4 Guštin 1978, 119. Sl. 1: Gradišče nad Knežakom. Prikaz digitalnega modela reliefa na podlagi lidarskih podatkov (zgoraj) in interpretacija arheoloških sledov (spodaj). Fig. 1: Gradišče above Knežak. Digital surface model based on ALS data (above) and its archaeological interpretation (below). jedra naselbine ločena z izrazitim pobočjem oziroma vr­tačo (sl. 1: b). Prekinitve v obzidjih večinoma razumemo kot vhode. Glavni in najširši vhod (za tovorne živali in morda vozove) v zunanjem okopu je bil s potjo povezan do prehoda skozi notranji okop. Arheološka interpreta­cija lidarskih podatkov je razkrila izjemno gosto mrežo, večinoma pravokotnih objektov na obeh poselitvenih terasah. Prepoznaven je raster naselbine, ki smo ga pred tem zgolj slutili v seriji polkrožno potekajočih teras s številnimi poglobitvami in manjšimi kotanjami.5 Gre za sledove stavb, ki stojijo samostojno ali pa se povezujejo v skupine in nakazujejo stavbe z več prostori. Na dveh območjih zgornjega dela naselbine zaradi odsotnosti sledov objektov domnevamo obstoj manjših odprtih prostorov, nekakšnih trgov (sl. 1: c). Od gostega rastra s prevladujočo večprostorno zasnovo stavb odstopa skupina podolgovatih in vzporedno potekajočih sledov v severovzhodnem delu spodnje poselitvene terase (sl. 1: d), morda dokaz njihove posebne namembnosti (skladišča?). Pomenljivo je občutno manjše število sle­dov objektov v dodatno utrjenem zahodnem območju, ki je s pobočjem oziroma vrtačo ločeno od osrednjega dela naselbine (sl. 1: e). Tukaj smemo domnevati de­lavniško četrt, morda predvsem dejavnosti, povezane z uporabo odprtega ognja (metalurške delavnice, kovači­je), ki so bile zaradi nevarnosti požara odmaknjene od najgosteje naseljenega središča, a še vedno sestavni in branjeni del gradišča. ZGODOVINA RAZISKAV TOPOGRAFSKA POROČILA IN SONDIRANJA Prvi objavljeni opis najdišča Gradišče nad Kneža­kom najdemo v topografskem pregledu Alfonsa Müll­nerja iz leta 1880. Prikazal ga je na dveh skicah, v opisu pa poudarja pomen bližine vodnega izvira ter omenja številne ostanke zidov in keramike v notranjosti nasel­bine. Med najdbami omenja srebrnik cesarja Avgusta.6 V maju istega leta je Ferdinand Schulz poslal Karlu Dežmanu pismo s skico gradišča.7 Ta natančneje kot pri Müllnerju označuje dvojni okop na jugu, utrditev dodatnega območja na zahodu naselbine in tudi več podolgovatih vzporednih sledov v severovzhodnem delu spodnje poselitvene terase. Na Gradišču nad Knežakom je raziskoval Raffaele Battaglia.8 Gradišče opisuje kot obsežno naselbinsko območje nepravilne trikotne oblike, ki ga obdaja velik nasip. Tudi on pravilno ugotavlja obstoj dveh vzporednih 5 Urleb 1958–1959a; Urleb 1975, 66; Laharnar 2012, 66–67. 6 Müllner 1880, 23, sl. 8–9. 7 Pismo Schulza Dežmanu z dne 26. 5. 1880, Arhiv RS, fond Privata A. VIII, Dežman Karl, fasc. 4. 8 Battaglia 1927, 107–113. presumes (see History of research). A similar question pertains to the curved fortification and extension in the west (circumference 320 m; 0,6 ha); it is separated from the settlement core by a steep slope of a sinkhole (Fig. 1: b). The breaks in the ramparts are for the most part interpreted as entrances. The main and widest entrance (for pack animals and possibly carts) in the outer ram­part hosted a path that continued to the inner rampart. In the interior, the archaeological interpretation of the ALS data has revealed a dense network of mostly rectan­gular buildings on both terraces. It also revealed more clearly the internal layout of the settlement, previously only presumed in the series of terraces with numerous depressions.5 These represent the remains of buildings that are either detached or contiguous with two or more rooms. Two areas without buildings on the upper terrace suggest the existence of open spaces or squares (Fig. 1: c). Another particular feature is the group of parallel linear traces in the NE part of the lower terrace that differs considerably from the dense layout of mainly multi-room buildings (Fig. 1: d) and suggests a special function, for example as storage facilities. In contrast to the densely spaced buildings in the central part, the west extension has considerably less discernible building remains (Fig. 1: e); it may have been a workshop area with open fires (metalworkers, blacksmiths) separated from the central area for safety reasons, but still an integral and defended part of the hillfort. HISTORY OF RESEARCH TOPOGRAPHIC REPORTS AND RESULTS OF TRIAL TRENCHING The first description of the hillfort at Gradišče above Knežak is a topographic survey that Alfons Mül­lner published in 1880. Illustrated with two sketches, his description emphasises the proximity of a water source and mentions numerous remains of walls and pottery finds in the interior of the settlement. Mentioned among the small finds is a silver coin of Augustus.6 In the same year, Ferdinand Schulz wrote a letter to Karl Dežman accompanied by a sketch of Gradišče.7 The sketch is more precise than Müllner’s and marks the double rampart in the south, the fortification of the west extension and also several parallel linear features in the NW part of the lower terrace. 5 Urleb 1958–1959a; Urleb 1975, 66; Laharnar 2012, 66–67. 6 Müllner 1880, 23, Figs. 8–9. 7 Schulz's letter to Dežman dated 26 May 1880, Archives of the Republic of Slovenia, fond Privata A. VIII, Dežman Karl, fasc. 4. okopov z vmesno široko teraso (polico), ki branita južno in jugovzhodno, lažje pristopno stran naselbine. Opazil je izrazito preoblikovano površje v notranjosti naselbine in ga povezoval z ostanki zidov, ki so potekali vzpored-no s skalnim robom in nasipi gradišča. Mesta, potek in rezultati njegovih sondiranj niso podrobno znani. Ugotovitve povzemamo iz opisov v omenjeni objavi, ki pa žal ni opremljena s slikami in načrti. V zgornjem delu gradišča omenja odkritje zidu z malto, nekaj, kar, kot pravi, skupaj z najdbami rimskih novcev v vrhnjih plasteh priča o poselitvi v rimski dobi. Preostale sonde pa so tako v zgornjem delu gradišča kot na njegovem obodu razkrile zidove v suhozidni gradnji in kamnite ruševine s prazgodovinskimi odlomki (lončenine?) in tudi kakšnim kovinskim predmetom. Več podatkov mu je uspelo zbrati s sondiranji zunanjega okopa, a kot piše, mu teh sond zaradi prekinitve izkopavanj ni uspelo podaljšati prek notranjega okopa na zgornji del naselbine. Battaglia je izmeril, da je zunanji okop z naklonom 20 stopinj visok 25 m, terasa (polica) za njim pa do notranjega okopa široka 30 m. Skozi zunanji okop je izkopal sedem sond do geološke osnove. Ugotovil je nepravilen potek skalne osnove z izrazitimi poglobitva-mi, ki jo je prekrivala rdečkasta gruščasta prst. Nanjo je bil postavljen močan, temeljni in do 2 m visok nasip iz apnenčastih blokov, ki je v ruševini prekril tudi večji del notranje terase. Nizek zidec oziroma nekakšno tlakovanje je to teraso ločevalo od notranjega pobočja nasipa. Pod ruševinami okopa na notranji terasi je od­kril 50 cm debelo črno plast z odlomki lončenine, nekaj koščenimi predmeti in razsekanimi živalskimi kostmi. Domneval je, da zunanji okop ni ruševina obzidja, saj v njegovi notranjosti niso odkrili nikakršne zidane gradnje. Po njegovem mnenju je okop nasutje, škarpa, ki je bila morda zgrajena ob izravnavi terena oziroma pripravi spodnje naselbinske terase. Battaglia je izrazil presenečenje nad dokaj maloštevilnimi najdbami iz teh sond, izrecno omenja le železno sulično ost in pozno­latensko fibulo vrste Jezerine.9 Svoje poročilo sklene z zanimivo tezo, da je bila arhitektura obrambnih gradenj pivško-postojnskih gradišč (omenja bližnje gradišče na Obrobi in Gradu pri Šmihelu pod Nanosom) drugačna od tiste v Istri (omenja gradišče Vrčin pri Vodnjanu10). Prva gradišča naj bi bila značilna za japodsko območje v mlajši železni dobi in naj ne bi imela obzidij, ampak le nasipe, druga so gradišča Histrov z grajenimi, celo t. i. kiklopskimi obzidji, ki so nastala že v bronasti dobi.11 Danes lahko njegovo trditev glede nasipa na Gradupri Šmihelu pod Nanosom popravimo, saj je Mehtilda Urleb med zaščitnimi raziskavami dela SZ oboda nasel­bine vendarle ugotovila ostanke starejšeželeznodobne obrambne arhitekture, ki je bila morda primerljiva z obzidjem na Cvingerju nad Virom pri Stični in IV. fazo 9 Battaglia 1927, 111. 10 Prim. Buršić-Matijašić 1988–1989. 11 Battaglia 1927, 111–113. Raffaele Battaglia investigated the site in the period between the two wars.8 He describes Gradišče as a vast settlement area of irregular triangular plan and enclosed with a substantial rampart. He also established two parallel ramparts separated by a wide terrace defending the more easily accessible S and SE sides. He observed a heavily transformed interior, which he associated with the remains of the walls running parallel to the rocky ridge and the ramparts of the hillfort. The exact spots, course and results of his trial trenching are unknown, and references are taken from his publication that un­fortunately lacks illustrations and plans. He mentions discovering a mortar-bound wall in the upper part of the hillfort, which he sees as evidence of Roman habita­tion in addition to the finds of Roman coins in the upper layers of the site. Other trenches in the upper part of the hillfort and on the rampart revealed drywalls and stone ruins with prehistoric fragments (of pottery?) and the scarce metal objects. He was able to gather additional information by trial trenching the outer rampart, though he could not extend these trenches into the interior and across the inner rampart because he had to terminate the investigations. Battaglia established that the outer rampart with the incline of 20 degrees measured 25 m in height, while the terrace between the outer and inner ramparts measured 30 m in width. He dug seven trenches across the outer rampart and uncovering an irregular bedrock with numerous depressions and covered with reddish gravelly soil. Constructed on top of this soil were thick and up to 2 m high foundations of the rampart composed of limestone blocks, which in its dilapidated state covered a large part of the lower terrace. A low wall or a sort of a pavement separated this terrace from the interior slope of the rampart. He found a 50 cm thick black layer con­taining pottery sherds, bone objects and chopped animal bones under the ruins of the rampart on the lower terrace. He presumed that the outer rampart was not the ruins of the stonework as he found no stone construction at its core; he rather interpreted it as a bank possibly created while levelling or preparing the terrain of the lower ter­race. Battaglia was surprised to recover very few small finds in these trenches, of which he only mentions an iron spearhead and a Late La Tene Jezerine brooch.9 He concludes his report with an interesting hypothesis that the defensive architecture in the Pivka and Postojna areas (mentioning the nearby hillfort at Obroba and Grad nearŠmihel pod Nanosom) differs from that in Istria (men­tioning the hillfort at Vrčin near Vodnjan).10 The former were presumably characteristic of the Iapodic area in the Late Iron Age and not enclosed with stonework ramparts, but only earthworks, while the latter are the hillforts of the Histri with stonework (cyclopean fortifications) ramparts 8 Battaglia 1927, 107–113. 9 Battaglia 1927, 111. 10 Cf. Buršić-Matijašić 1988–1989. obzidja na Libni,12 njegovo trditev pa lahko popravimo tudi glede nasipov na Obrobi, saj so bržkone iz bronaste oziroma pozne bronaste dobe.13 Urlebova poroča, da so na Gradišču med obema svetovnima vojnama marsikatero jamo s poudarjenim robom, kot imenuje sledove stavb, prekopali razni ama­terji. Sama pa je v eni od njih izkopala sondo in odkrila odlomke rimske keramike, živalske kosti, železne žeblje in koščke stekla.14 POSAMEZNE NAJDBE Od sredine osemdesetih let 20. stoletja so na Gra­dišču nad Knežakom veliko najdb našli z iskalnikom kovin. Med njimi izstopa novčna zakladna najdba 386 rimskih republikanskih asov iz sredine 2. st. pr. n. št., ki so jo odkrili avgusta 1986. Depo je bil v jugo-jugo­vzhodnem delu gradišča, približno 10 m od zunanjega nasipa. Novci so bili položeni v jamo premera 30 cm, ki je bila obložena z apnenčastimi, kot pest velikimi kamni, in pokrita s kamnito flišno ploščo (25 × 25 cm). Ta je bila približno 10 cm pod površjem. Nekaj novcev je bilo raztresenih brez pravega reda, v težišču depoja (25 cm globoko) pa so bili urejeni v nevzporedne vrste, znamenje, da so bili prvotno spravljeni v zvitkih oz. tul­cih iz organskega materiala. Depo so prerasle korenine, ki so verjetno nekaj novcev raznesle v krogu do 1 m od večine depoja.15 Leta 1991 je ekipa Numizmatičnega kabineta Narodnega muzeja Slovenije z detektorjem kovin pregledala mesto odkritja in našla več predmetov približno 150 m od mesta odkritja depoja. Med njimi je bil rimski ključ z bronastim ročajem in železno brado.16 Korpus detektorskih najdb, ki jih je v zadnjih treh de­setletjih uspelo pridobiti Narodnemu muzeju Slovenije, je obsežen. Najdbe so izkopali različni iskalci, žal pa so podatki o najdiščnih okoliščinah večinoma zelo skrom­ni. Večji del tega gradiva je bil že vključen v študije,17 posebej so bile obravnavane štiri rimske najdbe, ki so morda povezane z rimsko vojsko.18 12 Urleb 1990, 97–100. 13 Horvat 2005, 231–232; Bratina 2002–2004. 14 Urleb 1958–1959a. 15 FMRSl III, 53/2; zapisnik najditelja v arhivu Inštituta za arheologijo ZRC SAZU.16 Istenič 1994–1995. 17 Laharnar 2012. 18 Laharnar 2015, 32, t. 4: 13–16. dating back to the Bronze Age.11 The knowledge available to us today allows us to correct his assertion with regardsto the rampart at Grad near Šmihel pod Nanosom, as Mehtilda Urleb identified the remains of Early Iron Age defensive architecture during the rescue excavations of the NW part of the hillfort’s perimeter. The defensive architecture is possibly similar to the stonework rampart at Cvinger above Vir pri Stični and the Phase IV rampart at Libna.12 We can also correct Battaglia on the subject of the hillfort at Obroba, the ramparts of which probably date to the Bronze or more precisely Late Bronze Age.13 Urleb reports that amateurs between the two world wars dug up many of the depressions representing the remains of buildings. She trial trenched one of them and found Roman pottery sherds, animal bones, iron nails and glass fragments.14 STRAY FINDS From the mid-1980s onwards, numerous artefacts were found at Gradišče above Knežak with the help of metal detectors. Particularly important is the hoard of 386 Roman Republican asses from the mid-2nd century BC. It came to light in August 1986, in the S–SE part of the hillfort roughly 10 m from the outer rampart. The coins were placed in a pit measuring 30 cm across, lined with pieces of limestone as large as a fist and covered with a flysch slab measuring 25 x 25 cm. The slab was found some 10 cm below the surface. Some of the coins were lying haphazardly in the pit, while those in the centre of the hoard (25 cm deep) were arranged in differently oriented rows, suggesting they were originally stored in rolls or cases of organic material. The hoard was eventually disturbed by tree roots that prob­ably pushed individual coins away from the main group in a circumference of roughly a metre.15 In 1991, a team from the Numismatic Cabinet of the National Museum of Slovenia examined the area using a metal detector and found several object some 150 m from the findspot of the hoard including a Roman key with a bronze bow and an iron bit.16 Over the following three decades, the National Museum obtained quite a vast body of finds from the site. They were dug up by different amateurs who provided very scarce context data. Most of the artefacts have already been examined,17 including four from the Roman period that received special attention as they indicate a connection with the Roman army.18 11 Battaglia 1927, 111–113. 12 Urleb 1990, 97–100. 13 Horvat 2005, 231–232; Bratina 2002–2004. 14 Urleb 1958–1959a. 15 FMRSl III, 53/2; notes of the finder in the archives of the Institute of Archaeology ZRC SAZU. 16 Istenič 1994–1995. 17 Laharnar 2012. 18 Laharnar 2015, 32, Pl. 4: 13–16. ARHEOLOGIJA ANTIČNE NASELBINE IN NJENE NEPOSREDNE OKOLICE DROBNE NAJDBE Med dokumentiranimi posamičnimi najdbami z Gradišča nad Knežakom so predmeti in odlomki predmetov iz železa, svinca, brona in drugih bakrovih zlitin. Gre za nakit in različne predmete noše, orodje, orožje in vojaško opremo, posodje ter razne nedoločljive predmete. Najdbe datiramo v pozno bronasto dobo (npr. odlomek bronastega srpa), starejšo železno (npr. odlom­ki trortaste, kačaste in certoških fibul), mlajšo železno (npr. notranjske različice certoških fibul vrste VIIIf, odlomki fibul latenskih shem) in rimsko dobo (sl. 2). Rimske najdbe so iz avgustejske dobe in 1. st. (npr. fibule vrst Almgren 67, 70/73 in Aucissa, ključa, zvonec cilindrične oblike) ter iz 2. do 4. st. (npr. različne vrste močno profiliranih fibul, odlomek kolenčaste fibule, obročasta fibula z izrastkom, pasna okova in pasna spona,19 rovnica, lemež). Pomembne so numizmatične najdbe. Ob omenjeni zakladni najdbi 386 republikanskih asov so z najdišča objavljeni: skupna najdba štirih denarijev, kovanih med letoma 58 in 32 pr. n. št., ki so jih našli na zunanji strani nasipa naselbine,20 in 77 posamičnih rimskih novcev.21 Uporabo štirinajstih republikanskih asov, ki so bili sicer izdelani v prvi polovici 2. st. pr. n. št., moramo zaradi njihove slabe ohranjenosti in polovičenja datirati šele v pozno republikansko obdobje. V približno isti čas sodijo tudi: preluknjan denarij iz leta 110/109 pr. n. št., polovi-čen semis iz okoli leta 91 pr. n. št. in bronasti novec tipa “Divos Iulius”, kovan leta 38 pr. n. št. Malo poznejših je osem Avgustovih bronastih novcev, ki jim sledijo še trije Tiberijevi in en Neronov as. Iz leta 69 je Otonov denarij, iz obdobja flavijske dinastije (69–96) pa dva denarija in trije bronasti novci. Iz skoraj stoletnega obdobja nervan­sko-antoninske dinastije (96–192) je dokumentiranih 21 novcev, od tega 5 denarijev, 4 sesterciji, 2 dupondija in 10 asov. Iz 1. ali 2. stoletja je tudi 5 asov, vendar jih zaradi obrabljenosti ni bilo mogoče natančneje opre­deliti. Iz časa Severov (193–235) so le en denarij in dva sestercija, nato pa začnejo v obdobju vojaških cesarjev (235–285) v obtoku prevladovati antoninijani (teh je bilo najdenih 8) in sesterciji, najpogostejši bronast denar tega časa (najdeni so bili trije). Najmlajši rimski novec je centenional – iz let 330/333. Popolna odsotnost novcev iz druge polovice 4. stoletja, ki navadno prevladujejo v poznorimskem obtoku, bi lahko kazala vsaj na prekinitev denarnega gospodarstva, če že ne na opustitev naselbine. 19 Laharnar 2015, 32, t. 4: 13–15. 20 FMRSl IV, 39/2. 21 FMRSl III, 53/1; FMRSl IV, 39/1; FMRSl V, 35. ARCHAEOLOGY OF THE ROMAN SETTLEMENT AND ITS IMMEDIATE PROXIMITY SMALL FINDS The recorded stray finds from Gradišče above Knežak consist of objects and fragments of iron, lead, bronze and other copper alloys. They are pieces of jewellery and costume, tools, weapons and military equipment, vessels and different unidentifiable objects. They date to the Late Bronze (e.g. a fragment of a bronze sickle), Early Iron (e.g. fragments of a three-knobbed, a serpentine and several Certosa brooches) and Late Iron Ages (e.g. Notranjska vari­ants of the VIIIf Certosa brooches, fragments of brooches of a La Tene construction), as well as the Roman period. The Roman finds date to the Augustan period and the 1st century AD (e.g. Almgren 67, 70/73 and Aucissa brooches, two keys, a cylindrical bell), as well as the 2nd–4th centuries (e.g. different kräftig profilierte Fibeln, fragments of a knee-shaped brooch, penannular brooch with a knob, two belt fittings and a belt buckle,19 hoe, ploughshare). The coin finds are very revealing. Apart from the above-mentioned hoard of 386 Republican asses, the site also yielded a group find of four denarii minted between 58 and 32 BC and recovered on the exterior side of the rampart,20 as well as 77 individually recovered Roman coins.21 Fourteen Republican asses minted in the first half of the 2nd century BC should be dated to the Late Republican period on the basis of their worn condition and halving (i.e. deliberately cut into two). Roughly the same dating can also be attributed to a pierced denarius minted in 110/109 BC, a halved semis from ca. 91 BC and a bronze Divos Iulius coin minted in 38 BC. Eight bronze coins of Augustus are slightly later, followed by three asses of Tiberius and an as of Nero. A denarius of Otho was minted in 69. Two denarii and three bronze coins were minted under the Flavian dynasty (69–96). Twenty-one coins are recorded from the almost century-long Nerva-Antonine dynasty (96–192), of those five denarii, four sestertii, two dupondii and ten asses. Another five asses also date from the 1st–2nd centuries, but are too worn to be identified more precisely. One denarius and two sestertii date to the Severan period (193–235). Under the barracks emperors (235–285), the antoniniani (eight were found at Gradišče) and sestertii (most common bronze coins of the period, of which three were found at Gradišče) begin to dominate the monetary circulation. The last of the Ro­man coins is a centenionalis from 330/333. A complete absence of coins from the second half of the 4th century, which usually predominate in the Late Roman circulation, would indicate at least an interruption of the monetary economy in the area, if not an abandonment of the site. 19 Laharnar 2015, 32, Pl. 4: 13–15. 20 FMRSl IV, 39/2. 21 FMRSl III, 53/1; FMRSl IV, 39/1; FMRSl V, 35. Sl. 2: Gradišče nad Knežakom. Rimske fibule. Bron ali bakrova zlitina. M. = 1:2. Fig. 2: Gradišče above Knežak. Roman brooches. Bronze or copper alloy. Scale = 1:2. (Oblikovanje / Design: Ida Murgelj, Narodni muzej Slovenije.) INTERPRETACIJA LIDARSKIH POSATKOV Na podlagi lidarskih podatkov rekonstruiran načrt naselbine prikazuje predvsem sledi stanja zadnjega ob-dobja njene poselitve. To ima glede na številne antične drobne najdbe in podatke o odkritju zida z malto22 težišče v rimski dobi, čeprav morata biti njen obseg in arhitekturna zasnova v precejšnji meri podobna obsegu in zasnovi prazgodovinske oziroma železnodobne na­selbine. Jedro železnodobne naselbine zamejujeta strmo severno pobočje in notranji okop, razširitev gradišča s spodnjo naselbinsko teraso in gradnjo zunanjega oko-pa pa je morda, če verjamemo ugotovitvam Battaglie, mlajšeželeznodobni poseg. Ali so redkeje pozidano “delavniško” zahodno območje utrdili sočasno z gradnjo osrednjega dela gradišča ali morda pozneje, ne vemo. Ob načrtu naselbine smo z lidarskimi podatki do-bili tudi podobo krajine v okolici gradišča s številnimi arheološkimi sledovi, kot so linearne groblje, polja gro­belj, škatlasta polja, ograde in ugreznjene poti. Pregled arhivskih podatkov o skromnih predhodnih raziskavah in prvi rezultati sondiranj dokazujejo železnodobno starost vsaj nekaterih sledov,23 vendar o evoluciji gradišč­ne krajine še ne vemo veliko, tako da antični nastanek in uporaba nekaterih sledov nista izključena. Skupina linearnih sledov jugovzhodno pod Gra­diščem (sl. 1: f) odstopa od t. i. linearnih grobelj, za katere domnevamo prazgodovinsko starost (sl. 1: g), in je morda sled antične arhitekture. To pa zagotovo velja za ostanek pravokotnega objekta (stolpa) z dimenzijami približno 8 × 7 m na vzpetini Reber (729 m), jugovzhod-no od Gradišča (sl. 1: h). Urlebova ga je sondirala leta 1957 in odkrila rimsko keramiko, slabo ohranjen zid in novca cesarja Konstantina II.24 Antični stolp je bil umeščen na severnem robu vzpetine, ki jo sicer obdaja okop (sl. 1: i). Ta je verjetno prazgodovinska gradnja, ki pa ni oklepala naselbine, ampak verjetneje občasno pribežališče ali prostor za črede. Gradnja stolpa na Rebri je bila verjetno del obrambnih ukrepov prebivalcev Gradišča nad Knežakom v nemirnem poznorimskem času, ko je bilo Pivško v neposrednem zaledju sistema alpskih zapor. 22 Battaglia 1927, 108. Laharnar, Lozić, Štular 2016; Laharnar, Lozić, Štular 2017; Laharnar, Lozić, Štular 2019. 24 Urleb 1958–1959b (najdišče, imenovano Gradišče na vrhu nad Knežakom); FMRSl I, 78 (najdišče, imenovano Vrh nad Knežakom); Horvat 2005, 231 (po reviziji je bilo ugotov­ljeno, da v objavi naštete detektorske najdbe izvirajo z Gra­dišča nad Knežakom). ALS DATA INTERPRETATION The plan of the hillfort reconstructed with the help of ALS data mainly shows the traces from the last period of occupation. Considering the numerous small finds from the Roman period and the information on the discovery of a mortar-bound wall,22 the main period of occupation presumably dates to the Roman period, although the ex­tent and basic architectural layout must largely be rooted in prehistory, more precisely the Iron Age. The core of the Iron Age settlement is delimited by the precipices in the north and by the internal rampart, while extending the settlement onto the lower terrace and constructing the outer rampart, if Battaglia’s the findings are to be relied upon, may be a Late Iron Age intervention. It is not clear whether the less densely built-up west ‘artisanal’ exten­sion was fortified simultaneously with the main part of the hillfort or possibly later. ALS data also offer a better insight into the hillfort landscape with numerous features such as linear bound­ary earthworks, cairnfields, box fields, enclosures and hollow ways. An examination of the archival data on the few preliminary investigations and the initial results of trial trenching campaigns prove an Iron Age date of at least some of them,23 but we know very little of the de­velopment of the hillfort landscape, hence a Roman date of part of the features cannot be excluded. The group of linear traces SE below the hillfort (Fig. 1: f) differs from the presumably prehistoric linear earthworks (Fig. 1: g) and may represent evidence of Ro­man architecture. This is certainly true of the remains of a roughly 8 x 7 m large rectangular building (tower) located on the hill of Reber (729 m), SE of Gradišče (Fig. 1: h). Urleb’s trial trenching at Reber in 1957 revealed Roman pot­tery, a poorly surviving wall and a coin of Constantine II.24 The Roman tower was constructed at the northern edge of the elevation enclosed within a rampart (Fig. 1: i). The latter is probably a prehistoric earthwork that did not protect a settlement, but rather an occasional refuge or served as an animal enclosure. Constructing a tower at Reber was probably a defensive measure on the part of the inhabitants of Gradišče above Knežak in the tumultuous Late Roman period, when the Pivka area lay in the immediate hinterland of the Claustra Alpium Iuliarum barrier system. 22 Battaglia 1927, 108. 23 Laharnar, Lozić, Štular 2016; Laharnar, Lozić, Štular 2017; Laharnar, Lozić, Štular 2019. 24 Urleb 1958–1959b (site named Gradišče na vrhu nad Knežakom); FMRSl I, 78 (site named Vrh nad Knežakom); Horvat 2005, 231 (a revision revealed that the finds recovered with a metal detector and listed in this publication originate from Gradišče above Knežak). STATUS IN POMEN GRADIŠČA NADKNEŽAKOM V RIMSKI DOBI Številne rimske drobne najdbe in osvetlitev obstoja kompleksne in strnjene pozidave nakazujejo, da je bilo Gradišče nad Knežakom v rimski dobi zelo obljudeno. Načrt in oblika naselbinskega rastra dopuščata domnevo o obstoju naselbine urbanega videza. Prostorska umestitev in obseg naselbine, potek naselbinskih teras in obrambne arhitekture ter komu­nikacije (poti) v bližini gradišča izvirajo že iz pozne prazgodovine oz. železne dobe. Ta hip ni jasen odnos gradišča do morebitne antične arhitekture ob poti nepo­sredno pod njim, njegov pomen in moč v pozni rimski dobi poudarja tudi gradnja obrambnega stolpa na bližnji Rebri. Kontinuiteta in intenzivnost poselitve Gradišča nad Knežakom iz železne v rimsko dobo pričata o gospodarski moči prebivalcev, morda delno povezani s kmetijsko dejavnostjo. Na to nakazujejo številni sledovi intenzivne rabe tal v neposredni okolici. V pravno-administrativnem pogledu smemo položaj antičnih prebivalcev Gradišča nad Knežakom, podobno kot razmišlja Andrej Gaspari na primeru Ulake nad Starim trgom pri Ložu,25 razumeti kot lokalno skup­nost, ki je prešla od statusa civitas foederata ali tributaria iz časa neposredno po rimski osvojitvi do položaja (gens) adtributa z dodelitvijo upravnemu centru po konsoli­daciji rimske oblasti, pri tem pa so za dlje časa ohranili pravni položaj peregrinov (peregrini).26 V primeru Gradišča nad Knežakom je bila rimski upravni center kolonija v Tergestu. Zgodovinarji namreč soglašajo, da je kraško-notranjsko zaledje, torej tudi Pivško, sodilo pod jurisdikcijo kolonije Tergeste.27 Zgodovinske okoliščine in tradicija železnodobne naselbine na Gradišču nad Knežakom se pa vendarle zdijo drugačne od ulaškega primera. Ulaka je najverjetneje sodila v južno obrobje mestnega ozemlja Emone28 in je imela vlogo središčne naselbine. Tudi za Ulako je značilna kontinuiteta pose-litve iz železne v rimsko dobo, a je v tem v primerjavi z drugimi notranjskimi gradišči na območju vzhodno od Javornikov izjema. Drugače je ob zgornji Pivki, v Košanski dolini in na Ilirskobistriškem takih točk več (sl. 3). Na podlagi drobnih najdb in zgovornih podatkov o gosti pozidavi notranjosti naselij, ki jih kaže arheo­ 25 Gaspari 2020. 26 Glej Zaccaria 1992, 156; Zaccaria 2007, 136. 27 Zaccaria 1992, 163–164. Npr. Šašel Kos 2000, 96–97. Odkritje mejnika med Akvilejo in Emono pri Bevkah sproža tudi vprašanja glede obsega emonskega mestnega ozemlja v smeri Notranjske (Šašel Kos 2002, 381–382). Po mnenju Claudia Zaccarie se je mestno ozemlje Akvileje kot klin do Bevk zarilo med te­ritorija mest Forum Iulii (pri tem ostaja problem teritorialna pripadnost Posočja) in Emona, pri tem pa ostaja ločnica med ozemljem Tergesta in Emone na Javornikih (Zaccaria 2007, 137–139). STATUS AND SIGNIFICANCEOF GRADIŠČE ABOVE KNEŽAK IN THE ROMAN PERIOD The numerous Roman small finds, as well as the complex and dense architecture suggest that Gradišče above Knežak was highly populated in the Roman pe­riod. The layout even indicates a settlement of an urban appearance. The location, size, settlement layout, fortifications, as well as the lines of communication around the hillfort all originate in late prehistory, i.e. the Iron Age. The available evidence does not reveal the relationship between the hillfort and the possible Roman architecture along the routeway immediately below it, though the construction of a defensive tower on the nearby hill of Reber certainly tells of the hillfort’s significance and strength in the Late Roman period. The intensity and continuity of habitation at the hillfort from the Iron Age to the Roman period tell of the economic strength of its inhabitants, possibly con­nected with a specific agricultural activity. The latter is in­dicated by intensive land use in the immediate proximity. Similarly as Andrej Gaspari posits for Ulaka above Stari trg pri Ložu,25 we may see the Roman-period inhabitants of the hillfort at Gradišče above Knežak in administrative and legal terms as a local community that went from a civitas foederata or tributaria, in the time immediately after the Roman conquest, to that of (gens) adtributa belonging to an administrative centre after the consolidation of the Roman rule, retaining the legal status of peregrini over a long period.26 Historians concur that the hinterland of the Kras and Notranjska, hence also the Pivka area and the hillfort at Gradišče above Knežak, fell under the jurisdiction of the colony of Tergeste.27 However, the historical circumstances and the tradition of the Iron Age settlement at Gradišče differ from those of Ulaka. The settlement on the latter hill most probably lay in the southern periphery of the territory of Emona28 and functioned as a locally central settlement. It witnessed a long-term occupation from prehistory to the Roman period as well, but in this it differs from other hillforts of Notranjska east of the Javorniki Hills that did not. Along the upper reaches of the River Pivka, in the Košana Valley and in the area of Ilirska Bistrica, 25 Gaspari 2020. 26 See Zaccaria 1992, 156; Zaccaria 2007, 136. 27 Zaccaria 1992, 163–164. 28 E.g. Šašel Kos 2000, 96–97. The location of the bound­ary stone between the territories of Aquileia and Emona, recovered at Bevke, also raises questions as to the extent of Emona’s territory in the direction of Notranjska (Šašel Kos 2002, 377–379). According to Claudio Zaccaria, the territory of Aquileia wedged northwards to Bevke between the ter­ritories of Forum Iulii and Emona (the territorial affiliation of the region of Posočje is unclear), with the dividing line between the territories of Tergeste and Emona leading along the Javorniki Hills (Zaccaria 2007, 137–139). Sl. 3: Geografska lega večine v besedilu omenjenih najdišč. 1 – naselbine (gradišča), ki so bile z rimskimi vojaškimi posegi ali dokončno vzpostavitvijo rimske oblasti opuščene; 2 –naselbine (gradišča) z jasno kontinuiteto v rimsko obdobje. Fig. 3: Location of major sites mentioned in the text. 1– settlements (hillforts) abandoned after Roman military interventions or the final establishment of Roman rule; 2 – settlements (hillforts) with continuous habitation from the Iron Age to the Roman period. loška interpretacija lidarskih podatkov, kontinuiteto v rimsko dobo ugotavljamo tudi na Kerinu nad Pivko,Šilentaborju, Gradišču na Čepni, Gradišču nad Trnovim v Ilirski Bistrici in verjetno na Gradišču nad Gornjo Košano. Podobne naselbine moramo predvidevati tudi v Vipavski dolini.29 Pri tem opozarjamo, da na tej točki raziskav ni jasna dinamika poselitev naštetih naselbin v celotnem rimskem obdobju. Številne drobne najdbe ne nasprotujejo domnevi, da so arhitekturni ostanki, prepoznani z lidarskimi podatki, predvsem poznorim-ska gradnja iz druge polovice 3. in 4. st., vendar so med najdbami tudi predmeti iz celotnega 1. in 2. st.30 Božidar Slapšak ugotavlja izrazito zmanjšanje ali prekinitev pose-litve Ajdovščine nad Rodikom in drugih kontinuiranih naselbin do druge polovice 2. st.31 To povezuje z dode­litvijo latinskega državljanstva peregrinskim skupnostim pod jurisdikcijo Tergesta in razpadom njihove družbene zgradbe, ki se je zgodil z integracijo lokalnih elit v rimsko državljanstvo in v strukture mestne kolonialne uprave. such settlements are common (Fig. 3). The small finds and ALS data on densely built-up interiors of these set­tlements allow us to posit the Iron Age - Roman Agecontinuity at Kerin above Pivka, Šilentabor, Gradišče at Čepna, Gradišče above Trnovo near Ilirska Bistrica and probably also Gradišče above Gornja Košana. Similar settlements are to be expected in the Vipava Valley.29 We should note that the settlement dynamics at these sites throughout the Roman period have not been tackled yet. The multitude of small finds from the Late Roman period allows for the possibility that the architectural remains mainly represent Late Roman constructions from the second half of the 3rd and the 4th centuries, although the recovered artefacts also include those from the whole of the 1st and 2nd centuries.30 Božidar Slapšak observed a significant decrease or abandonment of the settlement at Ajdovščina above Rodik, as well as decreased habitation at other continuously inhabited settlements up to the second half of the 2nd century.31 He associated this with the con­ 29 Prim. Sv Pavel nad Planino (Ciglenečki 2016, 420, sl. 2). 29 Cf. Sv Pavel above Planina (Ciglenečki 2016, 420, Fig. 2). 3030 Laharnar 2012; objava detektorskih najdb in prikazi Laharnar 2012; the publication of the metal detector lidarskih podatkov so v pripravi.finds and ALS data interpretations is in preparation. 31 Slapšak 2003, 250. 31 Slapšak 2003, 256–257. Vendar, če presojamo poselitev skozi prizmo kronolo­gije drobnih najdb (predvsem novejše dokumentacije o novčnih najdbah), omenjena prekinitev oziroma more-bitna zmanjšana intenzivnost izrabe lokacije, skrčenje naselbine ali manjša kupna moč prebivalcev,32 tako za rodiško Ajdovščino33 kot za Gradišče nad Knežakom, ni bila izrazita. Pri tem je pomenljivo, da pomembnejša gradišča južnega obrobja Spodnje Pivke (Baba in Ambrože­vo gradišče pri Slavini) in Notranjske vzhodno od Javornikov (Stari grad nad Uncem, Tržišče pri Dolenji vasi, Žerovnišček pri Bločicah) z izjemo Ulake (sl. 3) izrazite kontinuitete iz železne v rimsko dobo ne kažejo, saj zadnje obdobje njihove poselitve, ki je praviloma povezana z navzočnostjo in dejavnostjo rimske vojske, datiramo v avgustejsko obdobje oziroma najpozneje v prvo polovico 1. st. n. št.34 Nasprotno pa na Gradišču nad Knežakom in v drugih naštetih naselbinah z nakazano kontinuiteto v rimsko dobo rimskih vojaških najdb iz okupacijske faze ne poznamo oz. jih je bistveno manj (npr. svinčenih izstrelkov za pračo in drugega orožja, okovnih žebljičkov vojaških obuval). Sklepamo, da je tako stanje odraz različnih politik domorodnih skupnos-ti do rimske kolonialne in osvajalske politike ter torej s tem povezane kontinuitete ali diskontinuitete njihovih naselbin po rimski osvojitvi. Zahvale Prispevek je bil napisan v okviru raziskovalnega projekta Predmeti in krajine. Strukturiranost notranjskih železnodobnih skupnosti (št. J6-8251), ki ga sofinancira Javna agencija za raziskovalno dejavnost Republike Slovenije iz državnega proračuna. Za pomoč, koristne diskusije in nasvete iskrena hvala Jani Horvat, Benjaminu Štularju, Janki Istenič, Marjeti Šašel Kos in Andreju Gaspariju. 32 Slapšak 2003, 250, op. 55. 33 FMRSl V, 17; FMRSl VI, 28. Laharnar 2015, 14–22. Grad pri Šmihelu pod Nano­som pa je bil opuščen že po rimskem vojaškem posegu sredi 2. st. pr. n. št. (Horvat 2002; Laharnar 2015, 11–14, 24). ferral of Latin citizenship to the peregrine communities under the jurisdiction of Tergeste and the disintegration of their internal social structure, which occurred when the local elites became Roman citizens and integrated the structures of the colonial administration. However, the chronology of the small finds (particularly the recent numismatic evidence) shows that the already mentioned interruption or possibly decreased intensity of site use, the shrinkage of the settlements or a decreased purchasing capacity of the local population32 were not so significant, either at Ajdovščina33 or Gradišče above Knežak. We should also note that the more prominent hillforts on the southern fringes of the Spodnja Pivka Valley (Baba and Ambroževo gradišče near Slavina) and Notranjska east of the Javorniki Hills (Stari gradabove Unec, Tržišče near Dolenja vas, Žerovnišček near Bločice), with the exception of Ulaka (Fig. 3), do not show a conspicuous continuity of habitation from the Iron Age to the Roman period. The last period of the Iron Age oc­cupation on these hillforts is usually associated with the presence and activities of the Roman army and dated to the Augustan period or the first half of the 1st century AD at the latest.34 In contrast, Gradišče above Knežak and other hillforts with a presumed continuity into the Roman period yielded no or considerably less Roman military finds from the occupation phase, for example lead slingshot, other weapons, hobnails of military footwear. This likely reflects the different reactions of the local com­munities to the Roman expansionist policies, resulting in either continuity or abandonment of the settlements after the Roman conquest. Acknowledgements The authors acknowledge the financial support from the Slovenian Research Agency as part of the J6-8251 research project titled ‘Predmeti in krajine. Strukturiranost notranjskih železnodobnih skupnosti / Objects and landscapes. Structure of the Iron Age communities in Notranjska’. We wish to thank Jana Horvat, Benjamin Štular, JankaIstenič, Marjeta Šašel Kos and Andrej Gaspari for their help, useful discussions and professional advice. Translation: Andreja Maver 32 Slapšak 2003, 257, Fn. 55. 33 FMRSl V, 17; FMRSl VI, 28. 34 Laharnar 2015, 14–22. Grad near Šmihel pod Nano­som was already abandoned after the Roman military inter­vention in the mid-2nd century BC (Horvat 2002; Laharnar 2015, 11–14, 24). SEZNAM NOVČNIH NAJDB / LIST OF COIN FINDS 1 LJ 0015182. Preluknjan / Pierced. T. / W.: 3,61 g. F = 19 mm. Pol. peč. / Die axis: 11. Dobro ohranjen / Good condition. FMRSl IV 39/1–11. 2 LJ 0024417. T. / W.: 23,42 g. Pol. peč. / Die axis: 10. Izrabljen / Worn. FMRSl V 35–1. 3 LJ 0024416. T. / W.: 23,35 g. Pol. peč. / Die axis: 6. Izrabljen / Worn. FMRSl V 35–2. 4 LJ 0015207. Polovičen / Halved. T. / W.: 19,24 g. Dim. / Size: 35 × 17 mm. Pol. peč. / Die axis: 9. Izrabljen / Worn. FMRSl IV 39/1–1. LJ 0014701. Polovičen / Halved. T. / W.: 13,89 g. Pol. peč. / Die axis: 7. Izrabljen / Worn. FMRSl IV 39/1–2. 6 LJ 0015206. Polovičen / Halved. T. / W.: 13,67 g. Dim. / Size: 31 × 15 mm. Pol. peč. / Die axis: 6. Izrabljen / Worn. FMRSl IV 39/1–3. 7 LJ 0014713. Polovičen / Halved. T. / W.: 11,38 g. F = 30,5 mm. Pol. peč. / Die axis: 6. Izrabljen / Worn. FMRSl IV 39/1–4. 8 LJ 0015208. 3/4-tinjen / Cut to three quaters. T. / W.: 10,05 g. F = 28 mm. Pol. peč. / Die axis: 6. Izrabljen / Worn. FMRSl IV 39/1–5. 9 LJ 0014716. Polovičen / Halved. T. / W.: 8,48 g. Pol. peč. / Die axis: 6. Izrabljen / Worn. FMRSl IV 39/1–6. LJ 0014714. Četrtinjen / Cut to a quater. T. / W.: 6,26 g. Pol. peč. / Die axis: 8. Izrabljen / Worn. FMRSl IV 39/1–7. 11 LJ 0024418. Polovičen / Halved. T. W.: 6,24 g. Pol. peč. / Die axis: 7. Izrabljen / Worn. FMRSl V 35–3. 12 LJ 0014717. Preluknjan, polovičen / Pierced, halved. T. / W.: 5,96 g. Pol. peč. / Die axis: 9. Izrabljen / Worn. FMRSl IV 39/1–8. 13 LJ 0014715. Polovičen / Halved. T. / W: 5,74 g. Pol. peč. / Die axis: 7. Izrabljen / Worn. FMRSl IV 39/1–9. 14 LJ 0024419. Polovičen / Halved. T. / W.: 5,61 g. Pol. peč. / Die axis: 8. Izrabljen / Worn. FMRSl V 35–4. LJ 0014699. Preluknjan, polovičen / Pierced, halved. T. / W.: 5,3 g. Izrabljen / Worn. FMRSl IV 39/1–10. 16 LJ 0014700. Polovičen / halved. T. / W.: 4,34 g. Pol. peč. / Die axis: 9. Dobro ohranjen / Good condition. FMRSl IV 39/1–12. 17 LJ 0015221. T. / W.: 11,57 g. F = 26,5 mm. Pol. peč. / Die axis: 7. Odlično ohranjen / Excellently preserved. FMRSl IV 39/1–14. 18 LJ 0024796. Av.: Kopf r. / head r. T. / W.: 8,36 g. Pol. peč. / Die axis: 6. Izrabljen / Worn. FMRSl V 35–5. 19 LJ 0014658. T. / W.: 8,05 g. F = 26 mm. Pol. peč. / Die axis: 6. Izrabljen. / Worn. FMRSl IV 39/1–15. LJ 0014698. Polovičen / Halved. T. / W.: 4,4 g. F = 25 mm. Pol. peč. / Die axis: 6. Izrabljen / Worn. FMRSl IV 39/1–16. 21 LJ 0014659. Polovičen / Halved. T. / W.: 4,21 g. Pol. peč. / Die axis: 12. Izrabljen / Worn. FMRSl IV 39/1–17. 22 LJ 0014660. Preluknjan, fragmentiran / Pierced, fragmented. T. / W.: 1,41 g. Pol. peč. / Die axis: 6. Izrabljen / Worn. FMRSl IV 39/1–18. 23 LJ 0024795. Av.: Glava levo / Head left. T. / W.: 7,13 g. Pol. peč. / Die axis: 6. Izrabljen / Worn. FMRSl V 35–6. 24 LJ 0015222. Določitev po portretu / Portrait identification. T. / W.: 7,21 g. Dim. / Size : 26 × 27 mm. Izrabljen / Worn. FMRSl IV 39/1–19. LJ 0014704. T. / W.: 8,01 g. F = 26 mm. Pol. peč. / Die axis: 12. Dobro ohranjen / Good condition. FMRSl IV 39/1–20. 26 LJ 0014703. T. / W.: 6,16 g. F = 25 mm. Pol. peč. / Die axis: 12. Izrabljen / Worn. FMRSl IV 39/1–21. 27 LJ 0015220. Določitev po portretu / Portrait identification. T. / W.: 9,83 g. Dim. / Size: 26 × 27 mm. Izrabljen / Worn. FMRSl IV 39/1–22. 28 LJ 0015187. Suberat / Plated coin. T. / W.: 2,06 g. Dim. / Size: 16 × 18 mm. Pol. peč. / Die axis: 8. Dobro ohranjen / Good condition. FMRSl IV 39/1–23. 29 LJ 0015188. T. / W.: 2,39 g. Dim./ Size: 17,5 × 18 mm. Pol. peč. / Die axis: 6. Dobro ohranjen / Good condition. FMRSl IV 39/1–24. LJ 0015191. T. / W.: 2,94 g. Dim. / Size: 17,5 × 18,5 mm. Pol. peč. / Die axis: 7. Odlično ohranjen / Excellently preserved. FMRSl IV 39/1–25. 31 LJ 0005789. Določitev po portretu / Portrait identification. T. / W.: 8,02 g. F = 26,5 mm. Izrabljen /Worn. FMRSl III 53/1–1. 32 LJ 0014718. Določitev po portretu / Portrait identification. T. / W.: 5,9 g. Izrabljen / Worn. FMRSl IV 39/1–26. 33 LJ 0015205. T. / W.: 9,79 g. Dim. / Size: 26 × 28 mm. Pol. peč. / Die axis: 6. Zelo lepo ohranjen / Excellently preserved. FMRSl IV 39/1–27. 34 LJ 0015192. T. / W: 2,93 g. Dim. / Size: 17 × 18,5 mm. Pol. peč. / Die axis: 6. Zelo lepo ohranjen / Excellently preserved. FMRSl IV 39/1–28. LJ 0015201. T. / W.: 21,72 g. Dim. / Size: 32,5 × 33 mm. Izrabljen / Worn. FMRSl IV 39/1–29. 36 LJ 0015193. T. / W.: 3,11 g. Dim. / Size: 17,5 × 18,5 mm. Pol. peč. / Die axis: 6. Odlično ohranjen / Excellently preserved. FMRSl IV 39/1–30. 37 LJ 0005172. Določitev po portretu / Portrait identification. T. / W.: 9,2 g. Dim. / Size: 25 × 26 mm. Pol. peč. / Die axis: 6. Dobro ohranjen / Good condition. FMRSl III 53/1–2. 38 LJ 0014711. Določitev po portretu / Portrait identification. T. / W.: 9,56 g. F = 24 mm. Izrabljen / Worn. FMRSl IV 39/1–31. 39 LJ 0014712. Fragmentiran, določitev po portretu / Fragmented, portrait identification. T. / W.: 6,52 g. F = 25 mm. Izrabl­ jen / Worn. FMRSl IV 39/1–32. LJ 0024784. T. / W: 14,06 g. Pol. peč. / Die axis: 5. Dobro ohranjen / Good condition. FMRSl V 35–7. 41 LJ 0015212. T. / W.: 9,48 g. Dim. / Size: 26 × 26,5 mm. Pol. peč. / Die axis: 6. Dobro ohranjen / Good condition. FMRSl IV 39/1–33. 42 LJ 0015189. Suberat / Plated coin. T. / W.: 2,2 g. Dim. / Size: 17 × 18 mm. Pol. peč. / Die axis: 6. Dobro ohranjen / Good condition. FMRSl IV 39/1–34. 43 LJ 0015194. T. / W.: 3,09 g. F = 18 mm. Pol. peč. / Die axis: 6. Odlično ohranjen / Excellently preserved. FMRSl IV 39/1–35. 44 LJ 0015202. T. / W.: 19,71 g. Dim. / Size: 28 × 29 mm. Izrabljen / Worn. FMRSl IV 39/1–36. 45 LJ 0015215. T. / W.: 9,97 g. Dim. / Size: 26 × 28 mm. Pol. peč. / Die axis: 12. Dobro ohranjen / Good condition. FMRSl IV 39/1–37. 46 LJ 0015216. T. / W.: 8,23 g. Dim. / Size: 26 × 26,5 mm. Pol. peč. / Die axis: 7. Dobro ohranjen / Good condition. FMRSl IV 39/1–38. 47 LJ 0015218. T. / W.: 12,52 g. Dim. / Size: 25 × 25,5 mm. Pol. peč. / Die axis: 1. Dobro ohranjen / Good condition. FMRSl IV 39/1–39. 48 LJ 0015214. T. / W.: 10,17 g. Dim. / Size: 24 × 25 mm. Pol. peč. / Die axis: 12. Dobro ohranjen / Good condition. FMRSl IV 39/1–40. 49 LJ 0015190. T. / W.: 3,02 g. Dim. / Size: 18 × 19 mm. Pol. peč. / Die axis: 6. Dobro ohranjen / Good condition. FMRSl IV 39/1–41. 50 LJ 0014707. Določitev po portretu / Portrait identification. T. / W.: 7,33 g. F = 24 mm. Izrabljen / Worn. FMRSl IV 39/1–42. 51 LJ 0014710. Določitev po portretu / Portrait identification. T. / W.: 6,77 g. F = 23 mm. Izrabljen / Worn. FMRSl IV 39/1–43. 52 LJ 0015217. T. / W.: 8,84 g. F = 24 mm. Pol. peč. / Die axis: 6. Izrabljen / Worn. FMRSl IV 39/1–44. 53 LJ 0031019. T. / W.: 25,84 g. F = 30,5 mm. Pol. peč. / Die axis: 12. Dobro ohranjen / Good condition. FMRSl V 35–8. 54 LJ 0015219. Rv.: sedeča figura d. / seated figure r. T. / W.: 16,65 g. Dim. / Size: 26 × 28 mm. Pol. peč. / Die axis: 6. Izrabljen / Worn. FMRSl IV 39/1–45. 55 LJ 0015195. T. / W.: 2,63 g. Dim. / Size: 19 × 20 mm. Pol. peč. / Die axis: 6. Zelo lepo ohranjen / Excelently preserved. FMRSl IV 39/1–46. 56 LJ 0015203. T. / W.: 18,53 g. Dim. / Size: 31 × 30 mm. Pol. peč. / Die axis: 12. Dobro ohranjen / Good condition. FMRSl IV 39/1–47. 57 LJ 0015204. T. / W.: 22 g. Dim. / Size: 31,5 × 30 mm. Pol. peč. / Die axis: 12. Dobro ohranjen / Good condition. FMRSl IV 39/1–48. 58 LJ 0015196. Suberat / Plated coin. T. / W.: 3,57 g. Dim. / Size: 21 × 22 mm. Zelo lepo ohranjen / Excellently preserved. FMRSl IV 39/1–49. 59 LJ 0015197. T. / W.: 3,3 g. Dim. / Size: 20,5 × 22 mm. Pol. peč. / Die axis: 6. Dobro ohranjen / Good condition. FMRSl IV 39/1–50. 60 LJ 0015211. T. / W.: 18,96 g. F = 31 mm. Pol. peč. / Die axis: 12. Zelo lepo ohranjen / Excellently preserved. FMRSl IV 39/1–51. 61 LJ 0015210. T. / W.: 17,65 g. Dim. / Size: 30 × 29 mm. Pol. peč. / Die axis: 12. Dobro ohranjen / Good condition. FMRSl IV 39/1–52. 62 LJ 0015198. T. / W.: 3,02 g. Dim. / Size: 20 × 24 mm. Pol. peč. / Die axis: 12. Dobro ohranjen / Good condition. FMRSl IV 39/1–53. 63 LJ 0015209. T. / W.: 13,83 g. F = 29 mm. Pol. peč. / Die axis: 1. Dobro ohranjen / Good condition. FMRSl IV 39/1–54. 64 LJ 0015199. T. / W.: 2,81 g. Dim. / Size: 22 × 19 mm. Pol. peč. / Die axis: 12. Dobro ohranjen / Good condition. FMRSl IV 39/1–55. 65 LJ 0024787. T. / W.: 2,34 g. Pol. peč. / Die axis: 1. Dobro ohranjen / Good condition. FMRSl V 35–9. 66 LJ 0014705. T. / W.: 2,34 g. F = 17 mm. Izrabljen / Worn. FMRSl IV 39/1–57. 67 LJ 0014702. T. / W.: 1,54 g. F = 18 mm. Izrabljen / Worn. FMRSl IV 39/1–56. 68 LJ 0015213. T. / W.: 2,56 g. Dim. / Size: 21 × 21,5 mm. Pol. peč. / Die axis: 6. Dobro ohranjen / Good condition. FMRSl IV 39/1–58. 69 LJ 0008309. T. / W.: 2,06 g. Dim. / Size: 16 × 17 mm. Pol. peč. / Die axis: 12. Dobro ohranjen / Good condition. FMRSl IV 39/1–59. 70 LJ 0014670. Ožgan / Burnt. T. / W.: 9,94 g. F = 23 mm. Zelo močno izrabljen / Heavily worn. FMRSl IV 39/1–60. 71 LJ 0014709. T. / W.: 9,08 g. F = 25,5 mm. Zelo močno izrabljen / Heavily worn. FMRSl IV 39/1–61. 72 LJ 0014706. T. / W.: 8,24 g. F = 25 mm. Zelo močno izrabljen / Heavily worn. FMRSl IV 39/1–62. 73 LJ 0014708. T. / W.: 5,55 g. F = 24,5 mm. Zelo močno izrabljen / Heavily worn. FMRSl IV 39/1–63. 74 LJ 0022982. Fragmentiran / Fragmented. T. / W.: 4,96 g. Zelo močno izrabljen / Heavily worn. FMRSl V 35–10. 75 LJ 0040379. T. / W.: 0,7 g. Dim. / Size: 15 × 17 mm. Zelo močno izrabljen / Heavily worn. FMRSl VI 54–1. 76 LJ 0030522. Četrtinjen / Cut to a quater. T. / W.: 3,07 g. Izrabljen / Worn. FMRSl V 35–11. 77 LJ 0015200. T. / W.: 13,73 g. Dim. / Size: 28 × 30 mm. Pol. peč. / Die axis: 9. Dobro ohranjen / Good condition. FMRSl IV 39/1–13. BATTAGLIA, R. 1927, Necropoli e Castellieri dell' eta del ferro del Carnaro. – Bulletino di paletnologia italiana 47, 93–115. BRATINA, P. 2000–2004, Knežak. Knežak – gradišče Obroba. – Varstvo spomenikov 39–41. Poročila 2000–2004 (2006), 72. BURŠIĆ-MATIJAŠIĆ, K. 1988–1989, Gradina Vrčin u okviru brončanog doba Istre (Der Ringwall Vrčin im Rahmen der Bronzezeit Istriens). – Arheološki vestnik 39-40, 475–494. CIGLENEČKI, S. 1985, Potek alternativne ceste Siscija-Akvile­ja na prostoru zahodne Dolenjske in Notranjske v času 4. do 6. stoletja (Der Verlauf der Alternativstrasse Siscia- Aquileia im Raum von Westdolenjsko und Notranjsko in der Zeit­spanne vom 4. bis 6. Jh.). – Arheološki vestnik 36, 255–284. CIGLENEČKI, S. 2016, Claustra Alpium Iuliarum, tractus Italiae circa Alpes and the defence of Italy in the final part of the Late Roman period. – Arheološki vestnik 67, 409–424. FMRSl I: P. Kos, Die Fundmünzen der römischen Zeit in Slowe­nien I (Berlin 1988). FMRSl III: P. Kos, A. Šemrov, Die Fündmunzen der römischen Zeit in Slowenien III (Berlin 1995). FMRSl IV: A. Šemrov, P. Kos, Die Fundmünzen der römischen Zeit in Slowenien IV (Berlin 1998). FMRSl V: A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien V (Mainz am Rhein 2004). FMRSl VI: A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien VI (Wetteren 2010). GASPARI, A. 2020, Ulaka. - V / In: J. Horvat, I. Lazar, A. Gaspari (ur. / eds.), Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 141-171. GUŠTIN, M. 1978, Gradišča železne dobe v Sloveniji (Typo­logie der eisenzeitlichen Ringwälle in Slowenien). – Arhe­ološki vestnik 29, 100–121. HORVAT, J. 2002, The Hoard of Roman Republican Weapons from Grad near Šmihel / Zaklad rimskega republikanskegaorožja z Gradu pri Šmihelu pod Nanosom. – Arheološki vestnik 53, 117–192. HORVAT, J. 2005, Poselitev na Pivškem in ob zgornjem toku Reke od pozne bronaste dobe do pozne antike / Settlement in the Pivka Area and along the upper Course of the Reka River from the Late Bronze Age to the Late Antique Period. – V / In: Kras. Voda in življenje v kamniti pokrajini / Kras. Water and Life in a rocky Landscape, 220–248, Ljubljana. ISTENIČ, J. 1994–1995, Knežak – Gradišče nad Knežakom. – Varstvo spomenikov 36 (1997), 250–251. LAHARNAR, B. 2012, Notranjska med prazgodovino in antiko. – Disertacija / PHD thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). LAHARNAR, B. 2015, The Roman army in the Notranjska region / Rimska vojska na Notranjskem. – V / In: J. Istenič, B. Laharnar, J. Horvat (ur. / eds.), Evidence of the Roman army in Slovenia / Sledovi rimske vojske na Slovenskem, Katalogi in monografije 41, 9–41. LAHARNAR, B., E. LOZIĆ, B. ŠTULAR 2016, Prazgodovinska krajina Knežaka in Ilirske Bistrice. – V /In: P. Stipančić, B. Djurić, M. Črešnar (ur./ eds.), Arheologija v letu 2015, dediščina za javnost. Zbornik povzetkov, 30, Ljubljana. LAHARNAR, B., E. LOZIĆ, B. ŠTULAR 2017, Sondiranje arheoloških znakov pred Gradiščem nad Knežakom. – V / In: B. Djurić, P. Stipančić (ur. / eds.), Arheologija v letu 2016, dediščina za javnost. Zbornik povzetkov, 27. LAHARNAR, B., E. LOZIĆ, B. ŠTULAR 2019, Structured Iron Age landscape in the hinterland of Knežak (SW Slovenia). – V/In: D. C. Cowley, M. Fernández-Götz, T. Romankiewicz, H. Wendling (ur. / eds.), Rural Settlement. Relating buildings, landscape, and people in the European Iron Age, 247–255. MÜLLNER, A. 1880, Archäologische Excurse nach Süd-Steiermark und Krain. – Mittheilungen der K.K. Central- Commission zur Erforschung und Erhaltung der Kunst- und historischen Denkmale NF 6, 21–26. PLENIČAR, M. 1959: Tektonski okni pri Knežaku (Two tec­tonic windows at Knežak). – Geologija 5, 5–10. SLAPŠAK, B. 2003, O koncu prazgodovinskih skupnosti na Krasu / The end of prehistoric communities in the Karst region. – Arheološki vestnik 54, 243–257. ŠAŠEL, J. 1975, Rimske ceste v Sloveniji. – V / In: Arheološka najdišča Slovenije, 74–104, Ljubljana.ŠAŠEL KOS, M. 2000, Rimski napisi iz Šmarate (Römische Inschriften von Šmarata). – Kronika 48, 95–101. ŠAŠEL KOS, M. 2002, The boundary stone between Aquileia and Emona / Mejnik med Akvilejo in Emono. – Arheološki vestnik 53, 373–382. ŠEBELA, S. 2005, Tektonske zanimivosti Pivške kotline / Tectonic sights of the Pivka basin. – Acta carsologica 34/3, 566–581. URLEB, M. 1958–1959a, Gradišče nad Knežakom. – Varstvo spomenikov 7 (1960), 285. URLEB, M. 1958–1959b, Gradišče na vrhu nad Knežakom. – Varstvo spomenikov 7 (1960), 285. URLEB, M. 1975, Gradišča v Pivški kotlini. – V / In: Ljudje in kraji ob Pivki 1, 62–71, Postojna. URLEB, M. 1990, Grad pri Šmihelu pod Nanosom. Rezultati zaščitnih izkopavanj (Grad bei Šmihel unter dem Nanos. Resultate der Schutzgrabungen). – Arheološki vestnik 41, 89–104. ZACCARIA, C. 1992, Regio X. Venetia et Histria. Tergeste - Ager Tergestinus et Tergesti adtributus. – V / In: Supple-menta Italica n. s. 10, 139–283, Roma. ZACCARIA, C. 2007, Tra Natisone e Isonzo. Aspetti ammi­nistrativi in eta romana. – V / In: M. Chiaba, P. Maggi, C. Magrini (ur. / eds.), Le Valli del Natisone e dell’Isonzo tra Centroeuropa e Adriatico, Atti del convegno internazionale di studi, San Pietro do Natisone (UD) 15–16 settembre 2006, Studi e ricerche sulla Gallia Cisalpina 20, 129–144. Boštjan Laharnar Narodni muzej Slovenije Prešernova cesta 20 SI-1000 Ljubljana bostjan-laharnar@nms.si Edisa Lozić Universität Graz Institut für Antike Universitätsplatz 3/II A-8010 Graz edisalozic@gmail.com Alenka Miškec Narodni muzej Slovenije Prešernova cesta 20 SI-1000 Ljubljana alenka.miskec@nms.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 141–171 ULAKA Andrej GASPARI Izvleček Bistvene podatke o naselbini na Ulaki so dala izkopavanja W. Schmida med letoma 1936 in 1940 ter zračno lasersko skeni­ranje površja po letu 2010. Naselbina je bila bolj ali manj kontinuirano poseljena od začetka starejše železne dobe do 5. st. n. št. Gradišču s slabo ohranjenim obzidjem in s tremi prepoznavnimi vhodi pripadata plano grobišče severno od vršnega platoja in gomilna nekropola na južnih pobočjih. Pridatki iz grobov kažejo na razcvet v razvitem in mlajšem halštatu (od 7. do 5./4. st. pr. n. št.). Predmeti iz 2./1. st. pr. n. št. z območja gradišča so značilni za mlajšeželeznodobno notranjsko-kraško skupino, odražajo pa tudi stike s Tavriski, Japodi in z italskim prostorom. Rimski vojaški tabor na sosednjem Nadleškem hribu ter okop severozahodno od gradišča na Ulaki s sledovi spopada pričajo o rimskem obleganju v Cezarjevem ali Oktavijanovem času. Druga faza tabora na Nadleškem hribu sodi v avgustejsko obdobje in je verjetno povezana z nadzorom naselbine na Ulaki. Najpozneje v sredini 1. st. n. št. se Ulaka razvije v urejeno naselje z nizi stavb, ki se koncentrično širijo okoli osrednjega nepozidanega prostora. Prevladujejo (pol)vkopane stavbe s kamnitimi temelji in leseno nadgradnjo. Gospodarsko osnovo predstavljajo kovaške in lončarske delavnice. Izrazit višek doživi naselbina med drugo tretjino in koncem 2. oz. začetkom 3. st. Ponoven vzpon v sredini in drugi polovici 4. st. nakazuje sodelovanje prebivalcev oziroma posadke v logistični podpori delovanja obrambnega sistema Claustra Alpium Iuliarum. Ključne besede: Italija (10. regija), Ulaka, Nadleški hrib, mlajša železna doba, rimsko obdobje, naselbina, gradišče, vojaški tabor, obleganje, železarstvo, kovačnice, predelava brona, lončarstvo, orodje Abstract The essential data about the settlement on Ulaka comes from the excavation conducted by Schmid in 1936-1940, and from airborne laser scanning performed after the year 2010. The settlement was more or less continuously occupied from the begin­ning of the Early Iron Age to the 5th century AD. The hillfort with a poorly preserved rampart and three identifiable entrances also had a flat cemetery north of the top plateau and a barrow necropolis on the southern slopes. Grave goods indicate a period of flourishing in the developed and late Hallstatt periods (from the 7th to the 5th/4th centuries BC). Second- and first-century artefacts from the area of the hillfort are typical of the Late Iron Age Notranjska-Kras Group, suggesting contacts with the tribes of Taurisci and Iapodes, as well as with the Italic area. A Roman military camp on the neighbouring hill of Nadleški hrib and a rampart northwest of the hillfort Ulaka with traces of armed conflict attest to a Roman siege in the time of Caesar or Octavianus. The second phase of the fort on Nadleški hrib belongs to the Augustan period and is probably related to the control of the settlement on Ulaka. No later than in the middle of the 1st century, Ulaka became an orderly settlement with concentric rows of buildings around the central unbuilt area. The prevailing type of buildings are (semi-) sunken houses with stone foundations and wooden superstructures. The economy was based on smithies and pottery workshops. The settlement reached a substantial peak between the second third of the 2nd century and the end of the 2nd / beginning of the 3rd century. A resurgence in the middle and second half of the 4th century suggests that the inhabitants or the garrison participated in the logistic support of the Claustra Alpium Iuliarum defence system. Keywords: Italy (Regio X), Ulaka, Nadleški hrib, Late Iron Age, Roman period, settlement, hillfort, military camp, siege, iron working, smithy, bronze working, pottery, tools https://doi.org/10.3986/9789610502586_08 Sl. 1: Geografsko-prometna lega Ulake v rimskem obdobju (podlaga: ©ARSO, 2018). Fig. 1: Geographic-traffic position of Ulaka in the Roman period (base layer: © Slovenian Environment Agency, 2018). GEOGRAFSKO-PROMETNA LEGA Kopasta vzpetina z zanimivim imenom Ulaka, na kateri ležijo ostanki prazgodovinske in rimske naselbine, se dviga nad naseljem Stari trg pri Ložu na severozahod­nem robu Loške doline, z vseh strani zaprtega kraškega polja s precej ravnim, deloma poplavnim dnom, ki ga obdajajo visoke planote od Blok in Racne gore na severu in vzhodu do masiva Snežnika in Javornikov na jugu in zahodu.1 Loška dolina je nekoliko odmaknjena od glavnih poznoprazgodovinskih in antičnih prometnih koridor­jev v zaledju severnega Jadrana, zato sta kontinuirana poselitev in vitalnost naselbine na Ulaki, ki kažeta na njen središčni značaj, verjetno povezana z lego na stičiš-ču večih poti. Med njimi sta pomembnejši t. i. Japodska cesta, ki je iz smeri Tergesta prek Divače, doline Pivke in Javornikov čez Loško dolino, Babno polje in Prezid vodila proti povirju Kolpe, Gorskemu kotarju in Liki, ter pravokotno nanjo potekajoča pot iz smeri Ljubljane GEOGRAPHIC AND TRAFFIC POSITION The domed elevation of Ulaka, with the remains of a prehistoric and Roman settlement rises above the present-day settlement of Stari trg pri Ložu at the northwestern edge of the valley of Loška dolina, a karst polje with a relatively flat, partially flooded bottom, surrounded on all sides by high plateaus: from the Bloke plateau and Racna gora on the north and east, to the Snežnik massive and the Javorniki on the south and west.1 Loška dolina is slightly remote from the main late Pre­historic and Roman transport corridors in the hinterland of the northern Adriatic, and the continuous occupation and vitality of the settlement on Ulaka, which suggest its central character, are probably related to its position at the crossroads of several routes. The two major ones are the so-called Iapodic Route, which led from the direction of Tergeste across Divača, the valley of Pivka, and the moun­tain range of Javorniki to Loška dolina, and then across Babno polje and Prezid towards the upper stream of the 1 Habič 1977. 1 Habič 1977. Sl. 2: Jugovzhodni del Notranjskega podolja s Cerkniškim poljem in Loško dolino ter označenimi najdišči, omenjenimi v besedilu (podlaga: ©ARSO, 2018). Fig. 2: Southeastern part of Notranjsko podolje with Cerkniško polje and Loška dolina, with marked sites, mentioned in the text (base layer: © Slovenian Environment Agency, 2018). čez Golo, Rob, Lužarje, Metulje, Stari trg, Dane in preval Stare ogljenice (997 m) na Javornikih proti Kvarner­skemu zalivu (sl. 1; 2). Ta trasa je le nekaj kilometrov severno od Loške doline prečkala povezavo, ki se je od transkontinentalne komunikacije med Apeninskim polotokom in Podonavjem čez t. i. Iliro-italska vrata odcepila v osrednjem delu Notranjskega podolja pri Uncu in nato ob severnem robu Cerkniškega polja in čez Bloke vodila v dolino Krke ter naprej proti južno­panonskemu prostoru.2 TOPOGRAFSKI ORIS Središčno vlogo je naselbini zagotavljala lega na grebenu, ki se od Devina (792 m) oziroma Velikega (785 m) in Malega grebena (762 m) prek položnejšega hrbta Ulake (675 m) in nadaljevanja v nekoliko nižjem Kolpa river, Gorski kotar, and Lika; and the perpendicular route from the direction of Ljubljana across Golo, Rob, Lužarji, Metulje, Stari trg, Dane, and the Stare ogljenice pass (997 m) in the Javorniki towards the Kvarner Gulf (Figs. 1; 2). Just a few kilometres north of Loška dolina, the latter route crossed the route that branched off from the trans­continental artery between the Appeninne Peninsula and the Danube region across the so-called Illyrian-Italic Gate and turned near Unec in the central part of the Notranj­sko podolje plain, then led along the northern edge of Cerkniško polje and across the Bloke plateau to the valley of the Krka and towards the southern Pannonian territory.2 TOPOGRAPHICAL DESCRIPTION The central role of the settlement was due to its position on the ridge that runs from Devin (792 m), 2 Premerstein, Rutar 1899; Šašel 1975a, 75, 96; Cigleneč-2 Premerstein, Rutar 1899; Šašel 1975a, 75, 96; Cigleneč­ki 1985; Laharnar 2013, sl. 12. ki 1985; Laharnar 2013, Fig. 12. Sl. 3: Ulaka, leto 1998. Pogled proti severu. V ospredju Podcerkev s sv. Martinom, v ozadju severni del Loškega polja in naselji Stari trg in Lož. Fig. 3: Ulaka, in 1998. View towards the north. In the foreground the village of Podcerkev with the church of St Martin; in the background the northern part of the Loško polje and the settlements of Stari trg and Lož. Nadleškem hribu (642 m) zajeda v ravnino Loškegapolja (okoli 570 m) ter ga z nasproti ležečim Šmaraškim vrhom (646 m) deli na dva dela (sl. 2). Položen in raven vrh Ulake je približno 100 m višji kot okoliška ravnina pri Starem trgu (sl. 3). Površina vrhnjega dela okroglo-ovalne oblike po najdaljši osi Veliki greben (785 m), Mali greben (762 m), across the gentler part of the ridge with Ulaka (675 m), and towards the lower Nadleški hrib (642 m), encroaching into the plain of Loško polje (about 570 m) and, together withŠmaraški vrh (646 m) on the opposite side, dividing it in two parts (Fig. 2). The less steep and flat top of Ulaka is about 100 m higher than the surrounding plain at Stari trg (Fig. 3). The circular/oval upper part measures 260 m along the longest axis and up to 250 m in width. The circumference of the plateau that was suitable for habitation is about 980 m, and its surface area about 5 ha. To the north of the top plateau, there is typical karst terrain with exposed rock ridges and dolinas, while other slopes of Ulaka are less steep. The bedrock in the area of the settlement is composed of horizontal layers of dolomite and Jurassic limestone, which rarely reach the surface.3 A vertical aerial photograph of Ulaka from 1966 (Fig. 4) and a lidar image (Fig. 5) reveal a concentric grid of more or less rectangular anomalies, which can without doubt be associated with the layout of the Roman settle­ment. The heaps of debris and hollows, both natural and man-made, create a raster, which appears to be relatively equitably distributed over the top plateau and survives well, except in the western and partly southern parts of the settlement, which have been largely terraced and transformed by later (agrarian) land use. In the northern and eastern parts of the area, there are several promi­nent depressions with steep walls and regular outlines. 3 Gaspari 2000. Sl. 5: Ulaka in Nadleški hrib na podatkih lidarskega (ALS) snemanja (po Laharnar, Lozić 2017, sl. 13; obdelava: E. Lozić). Fig. 5: Ulaka and Nadleški hrib on the airborne laser scanning image (after Laharnar, Lozić 2017, fig. 13; processing: E. Lozić). 1 – Ulaka (prazgodovinska in rimska naselbina / Prehistoric and Roman settlement); 2 – Ulaka (zgodnjerimski tabor/okop / early Roman military camp/rampart); 3 – Nadleški hrib (zgodnjerimska tabora / early Roman military camps). meri 260 m, v širino pa do 250 m. Obseg za poselitev primernega platoja znaša okoli 980 m, njegova površina pa približno 5 hektarjev. Severno od vršnega platoja se širi izrazit kraški teren z razkritimi skalnimi grebeni in vrtačami, ostala pobočja Ulake pa so položnejša. Talno osnovo na območju naselbine sestavljajo vodoravni skladi dolomita in jurskega apnenca, vendar le malo kje segajo na površje.3 Navpična aerofotografija Ulake iz leta 1966 (sl. 4) in lidarski posnetek (sl. 5) razkrivata koncentrično mrežo bolj ali manj pravokotnih anomalij, ki jo lahko brez dvoma povežemo z zasnovo antične naselbine. Raster, ki preseva iz ruševinskih grobelj, naravnih in umetnih poglobitev, daje relativno enovit videz razporeditve okrog vršnega platoja in je nekoliko slabše ohranjen le na zahodnem in deloma južnem delu naselbine, kjer je v veliki meri terasiran in preoblikovan z mlajšo (agrarno) rabo zemljišč. V severnem in vzhodnem delu območja izstopa več izrazitih poglobitev s strmimi stenami in pravilnimi obrisi, ki so glede na topografsko ujemanje z načrti Walterja Schmida, verjetno ostanek izkopavanj v tridesetih letih prejšnjega stoletja (sl. 6).4 ZGODOVINA RAZISKAV Zgodovinska poročila o Ulaki sodijo na sam zače­tek zanimanja za arheološke spomenike na Slovenskem. Ostanki naselbine so pritegnili že pozornost polihis­torjev Janeza Ludvika Schönlebna (1681) in njegovega sodobnika Janeza Vajkarda Valvasorja (1689), ki sta v njej prepoznala oppidum Terponus (........), mesto Transalpinskih Japodov na poti pred znamenitim Metulumom iz Apijanove pripovedi o Oktavijanovih vojnah v Iiriku.5 Tezo o umestitvi Metuluma na bloške Metulje in Terpona na Ulako je oživil zgodovinar Anton Tomaž Linhart (1788), o tem pa so pisali tudi profesor Ivan Anton Župančič (1801), postojnski župnik Peter Hicinger (1854, 1855), zgodovinar in časnikar Peter Radics (1864), jezikoslovec Davorin Trstenjak (1858), zgodovinar Avgust Dimitz (1874) in drugi.6 Starejša odkritja predmetov z Ulake so zabeležene v seznamih pridobitev Deželnega muzeja v Ljubljani, ki jih je objavljal Illyrisches Blatt (1819–1849), medtem ko so poročila o najdbah, ki jih je zbiralo Historično društvo za Kranjsko, izhajala v Mitteilungen des historischen Ve-reins für Krain (1846–1868). Predmete, med katerimi po številčnosti izstopajo rimski novci, so prinašali predvsem duhovniki iz župnije Stari trg.7 3 Gaspari 2000. 4 Schmid 1937; 1944. 5 Ill., 18–19. 6 Kritično k t. i. kranjskim hipotezam (nem. “Krainer hy­pothesen”) Veith 1914, 28, 38–50, sl. 4. O zgodovini raziskav glej Šašel 1975b; Gaspari 2000; Laharnar 2012, 122–129. Topographically matching Walter Schmid’s plans, they are probably the remains of the 1930s excavations (Fig. 6).4 RESEARCH HISTORY Historical reports on Ulaka date to the very begin­ning of the interest for archaeological monuments in Slovenia. The remains of the settlement attracted the curiosity of polyhistors Janez Ludvik Schönleben (1681) and his contemporary Janez Vajkard Valvasor (1689), who identified it as the oppidum of Terponus (........), a town of Transalpine Iapodes, situated on the route towards the famous Metulum from Appian’s report of the wars of Augustus in Illyricum.5 The hypothesis that Metulum should be identified with the village of Metulje on the Bloke plateau, and Terponus with Ulaka, was revived by historian Anton Tomaž Linhart (1788) and discussedby Professor Ivan Anton Župančič (1801), parish priest in Postojna Peter Hicinger (1854, 1855), historian and journalist Peter Radics (1864), linguist Davorin Trstenjak (1858), historian Avgust Dimitz (1874), and others.6 Early discoveries of artefacts from Ulaka are re­corded in the acquisition lists of the Provincial Museum of Carniola, which were published in the Illyrisches Blatt newspaper (1819–1849). Reports on the finds collected by the Historical Society for Carniola were published in Mit­teilungen des historischen Vereins für Krain (1846–1868). The artefacts, among which Roman coins are the most numerous, were largely contributed by priests from the Stari trg parish.7 The first to attempt more serious research on Ulaka was digger of antiquities Jernej Pečnik. He described Ulaka as one of the largest prehistoric settlements of Car-niola (1890, 1894, 1904), with a massive stone rampart of more than 3000 m in length. He mentioned a distinctly high – compared to other hillforts – number and richness of the finds discovered by farmers ploughing their fields. He distinguished between prehistoric buildings – pits dug in the ground and lined with stones, and Roman-period walls with mortar.8 Alfons Müllner, curator of the Provincial Museum of Carniola, conducted a thorough inspection and measur­ing of the settlement. In his report he mentioned many clusters of stones from former buildings, different types of stones – even from far away; brick and mortar fragments, quernstones, pieces of slag and slag-coated stones, and coins.9 His study Emona included the text of two Roman 4 Schmid 1937; 1944. 5 Ill., 18–19. 6 A critical look on the so-called Carniolan hypotheses (German: “Krainer hypothesen”): Veith 1914, 28, 38–50, Fig. 4. 7 For research history, see: Šašel 1975b; Gaspari 2000; La- harnar 2012, 122–129. 8 Pečnik 1890, 382; 1894, 10; 1904, 187. 9 Müllner 1878, 88–89. Sl. 6: Schmidov “Tloris japodske Ulake” (po Schmid 1939 [geodetski posnetek: inž. Janez Černjač] in dopolnjeno po Schmid 1944). Ob originalnih oznakah objektov so dodane rdeče izpisane konkordančne številke objektov iz Schmidovega poročila leta 1944 (glej op. 40 in tab. 1). Fig. 6: Schmid‘s “Groundplan of Japodic Ulaka” (after Schmid 1939 [land survey: engineer Janez Černjač] and supplemented after Schmid 1944). Red concordance numbering of buildings from the Schmid‘s report (1944) are added to the original markings of the objects (see note 40 and tab. 1). Leto objave/poročila Year of publication/ report Oznaka objekta / Numbering of the buildings 1944 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1937, 1939 1 2 3 4 A B F C 5 D H 6 E - - Tab. 1: Ulaka. Konkordančni prikaz številčenja objektov v poročilu (Schmid 1944) ter oznakami v objavi (Schmid 1937) in na geodetskem načrtu (Schmid 1939). Prim. sl. 6. Tab. 1: Ulaka. Concordance table of the numbering of the buildings in the report (Schmid 1944), markings in the publication (Schmid 1937) and on the survey map (Schmid 1939). See Fig. 6. Prvi, ki se je resneje lotil raziskovanja Ulake, je bil starinokop Jernej Pečnik. Ulako opisuje kot eno največjih prazgodovinskih naselbin Kranjske (1890, 1894, 1904), z več kot 3000 m dolgim, velikim kamni­tim nasipom. V primerjavi z drugimi gradišči omenja izjemno visoko število in bogastvo najdb, na katere so pri oranju naleteli kmetje. Loči med prazgodovinskimi objekti-jamami, vkopanimi v zemljo in obloženimi s kamni, in rimskodobnimi, z malto zidanimi zidovi.8 Kustos deželnega muzeja Alfons Müllner v poro-čilu, ki ga je napisal po natančnem ogledu in merjenju naselbine, omenja številne skupine kamenja iz nekdanjih zgradb, različne, tudi od daleč stran izvirajoče vrste kamnin, kose opeke in malte, kamne ročnih mlinov, kose žlindre in z žlindro oblite kamne ter novčne najd­be.9 Njegova študija Emona prinaša tudi vsebino dveh rimskih napisnih kamnov, vzidanih v oltar cerkve Sv. Martina v Podcerkvi, ki jih je zabeležil že Schönleben.10 Leta 1906 je Ulako obiskal konservator dunajske Centralne komisije Jakob Žmavc, ki v dnevniku opi­suje, da je mestoma zasledil nasip, ki pa po njegovem mnenju ni tako velik, kot navaja Pečnik. Med drugim navaja osebe, ki naj bi hranile gradivo z naselbine, in omenja, da je knez Herman Schönburg-Waldenburg, lastnik gradu Snežnik, nameraval vso planoto Ulake sistematično preiskati.11 Osrednjo točko arheoloških raziskav Ulake pome­nijo sistematična izkopavanja Walterja Schmida v letih pred drugo svetovno vojno.12 Schmid, ki je na območju naselbine sondiral že avgusta 1916 v okviru raziskav poznorimskih zapor pri Prezidu,13 je velikopotezne raz­iskave, ki so potekale v jesenskih mesecih let 1936, 1937, 1939 in 1940, izpeljal kot predstojnik prazgodovinskega in zgodnjezgodovinskega oddelka ter numizmatičnega kabineta v graškem muzeju Joanneum v času svojega dopusta. Pri izvedbi terenskih del, ki sta jih omogočila 8 Pečnik 1890, 382; 1894, 10; 1904, 187. 9 Müllner 1878, 88–89. 10 Müllner 1879, Nr. 167 in 168. 11 J. Žmavc, Potni zapiski, zvezek 4, 16, 24–26; Arhiv Ar-heološkega oddelka Narodnega muzeja Slovenije (Az 86 - 22: št. 95). 12 Markantna osebnost vsestransko izobraženega znan­stvenika, sistematičnega in temeljitega muzejskega delavca, etnografa, požrtvovalnega terenskega arheologa in topogra-fa (r. kot Franc Šmid 1875 na Gašteju pri Kranju, u. 1951 v Gradcu) ter po pričevanjih sodobnikov poštenega in priljub­ljenega človeka vedrega in pokončnega značaja, ki ni bil brez vpliva na njegovo znanstveno presojo, morda najbolje preseva skozi izrek Vir sapiens dominabitur astris, s katerim je Franjo Baš pospremil nekrolog uglednemu znanstveniku. O življenju in delu W. Schmida glej Klemenc 1950; Kastelic 1951; Baš 1951; Modrijan 1953; bibliografija pri Sutter 1953. Schmid je bil povabljen k vključitvi v krog znanstvenih so-delavcev novoustanovljene Slovenske akademije znanosti in umetnosti, vendar se vabilu ni več uspel odzvati. 13 Pick, Schmid 1922–1924, 303–306; Schmid 1944, op. 1; Šašel, Petru 1971, 51, 64. inscription stones, built into the altar of the church of St Martin in Podcerkev, which had been recorded already by Schönleben.10 In 1906, Ulaka was visited by Jakob Žmavc, conser­vator of the Viennese Royal and Imperial Central Com­mission for the Research and Preservation of Artistic and Historical Monuments. He wrote in his journal that he had identified the rampart in some locations, but he did not consider it as large as Pečnik had claimed. He listed the people who kept some of the material from the set­tlement and mentioned that Prince Herman Schönburg-Waldenburg, the owner of Castle Snežnik, was planning a systematic investigation of the entire Ulaka plateau.11 The central point of the archaeological research of Ulaka were the systematic excavations conducted by Walter Schmid in the years before World War II.12 As early as August 1916, Schmid excavated some test trenches in the area of the settlement within the framework of the research of Late Roman fortifications near Prezid.13 This was followed by large-scale investigations in the autumns of 1936, 1937, 1939, and 1940, which he conducted as the head of the Department of Prehistory and Early History and the Numismatic Cabinet in the Joanneum Museum in Graz during his holidays. Schmid’s fieldwork was made possible by the support of the administration of the Drava banate and the Chamber of Commerce, Craft, and Industry in Ljubljana. In addition, he had the help of Jernej Hafner, parish priest in Stari trg and amateur historian; and engineer Janez Černjač, lecturer at the Technical Faculty of the University of Ljubljana. The latter created a land survey plan for Ulaka, which included the buildings that had been identified by then (a reduced version was published in Schmid 1939; see Fig. 6). Schmid continued his research on Ulaka, encouraged by the promising dis­coveries by Hafner and perhaps also by the test trenching conducted in 1935 by the Archaeological Seminar of the University of Ljubljana on the hill of Nadleški hrib 10 Müllner 1879, Nr. 167 and 168. 11 J. Žmavc, Potni zapiski, Volume 4, 16, 24–26; Archives of the Archaeological Department of the National Museum of Slovenia (Az 86 - 22: no. 95). 12 Born as Franc Šmid in 1875 in Gaštej near Kranj, died 1951 in Graz. The striking personality of this universally ed­ucated scientist, systematic and thorough museum curator, etnographer, selfless field archaeologist and topographer, and – according to his contemporaries – honest and likeable man of a cheerful and upright character, which was not without some influence over his scientific judgement, is perhaps best described by the words Vir sapiens dominabitur astris, used by Franjo Baš in the obituary for this respected scientist. For the life and work of W. Schmid, see: Klemenc 1950; Kastelic 1951; Baš 1951; Modrijan 1953; bibliography in Sutter 1953. Schmid was invited to join the circle of scientists of the newly founded Slovenian Academy of Sciences and Arts, but could no longer answer the call. 13 Pick, Schmid 1922–1924, 303–306; Schmid 1944, note 1; Šašel, Petru 1971, 51, 64. podpora kr. banske uprave Dravske banovine ter Zborni­ce za trgovino, obrt in industrijo v Ljubljani, se je Schmid oprl na pomoč starotrškega župnika in ljubiteljskega raziskovalca lokalne zgodovine Jerneja Hafnerja ter inženirja Janeza Černjača, docenta Tehnične fakultete ljubljanske univerze, ki je izdelal geodetski načrt Ulake z vrisanimi ugotovljenimi objekti (pomanjšan je objavljen v Schmid 1939; glej sl. 6). Schmida je k nadaljevanju raziskav na Ulaki poleg obetavnih odkritij župnika Haf­nerja morda dodatno vzpodbudilo sondiranje Arheo­loškega seminarja ljubljanske univerze na Nadleškem hribu leta 1935, ki ga je vodil Balduin Saria in s katerim sta se zapletla v nesporazum o ubikaciji Metuluma.14 Bistveno ostrejšo polemiko, v kateri je sledil tradiciji t. i. kranjskih hipotez in s precej rahlimi argumenti v Gradupri Šmihelu prepoznaval Metulum, v gradišču Javor pri Dolnjem Zemonu pa Terponus, je Schmid že desetletje prej vodil s c.-k. artilerijskim častnikom in strokovnja­kom za antično vojaško zgodovino Georgom Veithom.15 Schmid razen prvega poročila o raziskovanjih v letu 1936 izsledkov izkopavanj na Ulaki ni uspel publi­cirati, saj so načrtovano objavo pripravljenega besedila s slikovnimi prilogami v Mitteilungen der Prähistori­schen Kommission preprečili vojni dogodki. Schmidov terenski dnevnik, fotografije in skice ter tipkan rokopis s posameznimi načrti, ki jih hrani Universalmuseum Joanneum v Gradcu, časopisna poročila in geodetski posnetek izkopavanj iz leta 1939, objavljen v časopisu Slovenec,16 ter drobne najdbe iz izkopavanj v hrambi Narodnega muzeja Slovenije (NMS), so predstavljale osnovo revizijske študije rezultatov Schmidovih razis­kav.17 Slednja prinaša tudi pregled podatkov o drobnih najdbah z območja naselbine iz starejše literature, ki jih med gradivom iz depoja NMS razen redkih izjem ni mogoče prepoznati oziroma so izgubljene. Med izkopavanji Ulake leta 1939 je Schmid izvedel tudi sondiranje18 ob koncu 18. stoletja opuščene cerkve Sv. Petra, ki je stala v jugovzhodnem delu naselbine v bližini današnjega spomenika NOB.19 Septembra is-tega leta je raziskoval še na območju ledine Gradišče v severovzhodnem delu Starega trga, ki v franciscejskem katastru nosi ime Naglis Thurn. V treh dneh je izkopal ostanke z jarkom obdanega pravokotnega objekta z dimenzijami 27,40 × 15,80 m, ki ga sestavljata obzidje s tremi stolpi in samostojna, večprostorska stavba veli­kosti 13,10 × 9,20 m v njegovi notranjosti. Na podlagi spremljajočih keramičnih in kovinskih drobnih najdb 14 Glej Schmid 1937, 29–30; Saria 1937, 59. Glej diskusijo v Jahreshefte des Österreichischen Ar-chäologischen Institutes in Wien 21–22, 1922–1924, Beiblatt, str. 277–308, 479–494, 495–508, 507–510; Schmid 1923– 1924, 178–182. 16 Schmid 1939. 17 Gaspari 2000. 18 Glej Petru 1969, 116, op. 4. 19 Kebe 1996, 184–187, 208–209, 310. under the direction of Balduin Saria, with whom he had a disagreement about the localization of Metulum.14 A decade earlier, Schmid had a much sharper debate with Georg Veith, artillery officer and expert on the military history of classical antiquity, in which Schmid followed the tradition of the so-called Carniolan hypotheses, and, with fragile arguments, identified Metulum with thehillfort Grad near Šmihel, and Terponus with the hillfort Javor near Dolnji Zemon.15 With the exception of his first research report in 1936, Schmid never managed to publish his results of the excavations on Ulaka: his intended publication of the text with pictures in the appendices in Mitteilungen der Prähis­torischen Kommission was prevented by the war. The basis for the review study of the results of Schmid’s research were Schmid’s field journal, photographs, sketches, and typewritten manuscript with individual plans, which are all stored in the Universalmuseum Joanneum in Graz; newspaper reports and the land survey plan from the 1939 excavations, which were published in the Slovenec16 newspaper; and small finds from the excavations, which are stored in the National Museum of Slovenia (hence­forward: NMS).17 The study (2000) furthermore presents an overview of the data from earlier publications on the small finds from the area of the settlement, which can, with rare exceptions, no longer be identified among the material in the storage of the NMS, or are lost. During his 1939 excavations of Ulaka, Schmid also conducted test trenching18 in the church of St Peter, which had been abandoned at the end of the 18th century and stood in the southeastern part of the settlement, in the vicinity of the present-day World War II monument.19 In September of the same year, he investigated in the north­eastern part of Stari trg, in an area called Gradišče, or Na-glis Thurn in the Franciscean Cadastre. Within the course of three days, he excavated the remains of a rectangular complex, surrounded by a moat. The complex with the dimensions of 27.40 × 15.80 m included defensive walls with three towers, and within the walls an independent multi-roomed building measuring 13.10 × 9.20 m. Based on the accompanying ceramic and metal small finds, Schmid identified the site as an early medieval fortified manor (so-called Hausberg).20 Peter Petru’s opinion that the complex was a Late Antique fortified villa was not supported by convincing arguments and did not prevail.21 14 See: Schmid 1937, 29–30; Saria 1937, 59. 15 See the discussion in Jahreshefte des Österreichischen Archäologischen Institutes in Wien 21–22, 1922–1924, Bei­blatt, pp. 277–308, 479–494, 495–508, 507–510; Schmid 1923–1924, 178–182. 16 Schmid 1939. 17 Gaspari 2000. 18 See: Petru 1969, 116, note 4. 19 Kebe 1996, 184–187, 208–209, 310. 20 Schmid 1922, 39. 21 Petru 1969; on the so-called mottes and similar me­dieval fortified structures in Slovenia, see: Predovnik 2008, je Schmid najdišče opredelil kot srednjeveški utrjen dvorec (t. i. Hausberg).20 Mnenje Petra Petruja, da gre za poznoantično utrjeno vilo, ni podprto s prepričljivimi argumenti in se ni uveljavilo.21 Razen manjših sondiranj ter dokumentiranja po­segov v letih 1959, 1983 in 2003, ki so bili omejeni na jugovzhodni del naselbine in niso prinesli pomembnejših rezultatov, na Ulaki po Schmidovih izkopavanjih ni bilo večjih terenskih raziskav. V 1990-ih je Ulaka postala pogost cilj nepooblaščenih iskalcev z detektorji kovin. Znaten del zelo številnih najdb, ki so bile izkopane na tak način, je za svoje zbirke pridobil NMS. Integralno objavo pomembnejših mlajšeželeznodobnih in antičnih predmetov z Ulake in bližnjih najdišč, ki so bili že delo-ma zajeti v novejših specialističnih študijah in objavah, pripravlja B. Laharnar.22 Isti je tudi prvopodpisani avtor prostorsko-prometne analize vloge Ulake in Nadleškega hriba ter študije podatkov lidarskega snemanja, ki je prinesla povsem nov uvid v prostorsko organiziranost obeh najdišč. Slednja je nakazala obstoj doslej nepoznanih arheoloških struktur, potrjenih s testnimi sondiranji in terenskim pregledom leta 2017.23 POZNOPRAZGODOVINSKA POSELITEV IN SLEDOVI ZGODNJIH STIKOV Z ITALSKIM SVETOM DO RIMSKE OSVOJITVE Vedenje o naselbinskih in obrambnih strukturah, povezanih s poznoprazgodovinsko poselitvijo Ulake, je ob izostanku modernih invazivnih raziskav omejeno na domnevan potek nasipa oziroma obzidja, ki je prepozna­ven v ostrem prehodu med ravno teraso v notranjosti in pobočjem, ter lego in obliko vhodov. Argumentacija izza Schmidove domneve, da na vrhu Ulake ni bilo starejše železnodobne naselbine – predvideval jo je na jugozahodnih in zahodnih pobočjih,24 ni prepričljiva, pri čemer navidezno odsotnost zaključenih naselbinskih plasti in obrambnih naprav, povsem neprimerljivo z okoliškimi gradišči z dobro prepoznavnimi kamnitimi obzidji,25 povezujemo s plitvo ležečo, pretežno dolomit-no podlago in posledično skromno stratifikacijo, zlasti pa s posegi in uničenji v rimskem obdobju ter rabo prostora po koncu poselitve. V začetku 1930-ih odkrita grobna celota izvira iz “ene od gomil na zahodnem pobočju” Ulake, neprever­jena pa ostajajo starejša poročila o pripadajočih gomilah 20 Schmid 1922, 39. 21 Petru 1969; k t. i. motam in sorodnim srednjeveškim utr­jenim objektom v Sloveniji glej Predovnik 2008, 377–379, sl. 5. 22 Laharnar 2012; 2015. 23 Laharnar, Lozič 2015–2016; B. Laharnar, osebna in-formacija, 2017. 24 Schmid 1937, 17; 1944, 2. 25 Urleb 1977; Schein 1987. After Schmid’s excavations, there have been no major investigations on Ulaka, except for small test ex­cavations and documentation of building interventions in the years 1959, 1983, and 2003, which were limited to the southwestern part of the settlement and did not give more significant results. In the 1990s, Ulaka became a popular destination for unauthorised metal detectorists. A considerable part of the many finds excavated in such a way has been acquired by the NMS for its collections. An integral publication of some of the more important Late Iron Age and Roman artefacts from Ulaka and other sites in the vicinity is being prepared by Boštjan Laharnar, and some of these artefacts have been featured in recent specialised studies.22 Laharnar is also the first author of a spatial, movement and transport analysis of the roles of Ulaka and Nadleški hrib, and a study of the data from lidar scanning, which provided a brand new insight into the spatial organization of the two sites. It implied the existence of previously unknown archaeological structures, which were confirmed by archaeological test excavations and field survey in 2017.23 LATE PREHISTORIC OCCUPATION AND TRACES OF EARLY CONTACTS WITH THE ITALIC WORLD BEFORE THE ROMAN CONQUEST Due to the absence of modern invasive investiga­tions, our knowledge of the settlement and the defence structures associated with the late prehistoric occupation of Ulaka is limited to the assumed course of the rampart, recognisable by the sharp transition between the flat ter­race in the interior and the slopes, and by the position and shape of entrances. The argumentation behind Schmid’s assumption that there was no earlier Iron Age settlement on the top of Ulaka – he expected it on the southwestern and western slopes24 – is unconvincing; the apparent absence of defensive structures and settlement layers as whole, unprecedented in the surrounding hillforts with easily-recognizable stone ramparts,25 is now seen mostly as a consequence of spatial interventions in the Roman period, the use of space after the end of human habitation, and also the fact that the predominantly dolomite bedrock lies at relatively shallow depths beneath the surface, which results in poor stratification. A burial was discovered at the beginning of the 1930s in “one of the barrows on the western slope” of Ulaka. Earlier reports on the barrows and flat cremation graves 377–379, Fig. 5.22 Laharnar 2012; 2015. 23 Laharnar, Lozič 2015–2016; Boštjan Laharnar, perso­ nal information, 2017. 24 Schmid 1937, 17; 1944, 2. 25 Urleb 1977; Schein 1987. in planih žganih grobovih v okolici cerkve Sv. Martina v Podcerkvi, ter najdbah iz Tekavče jame na vzhodnem pobočju nad Starim trgom. O dinamiki predrimske poselitve Ulake tako za zdaj pričajo zlasti posamične in skupne najdbe, odkrite z detektorji kovin. Eno pomembnejših skupin najdb v hrambi NMS sestavlja večja količina halštatskega gradiva (čolničaste, kačaste in certoške fibule, gumbi, večji faleri, obesek, zapestnice ter stekleni jagodi) z območja prevala severno od Ulake, ki verjetno izvira iz več grobov.26 Naj­dbe s širšega območja naselbine sicer kažejo dinamično sliko, saj izpričujejo vsaj občasno prisotnost človeka že v bakreni dobi (kamnita puščična ost s trnom) in morda bronasti dobi (kosi kovine in del uhate sekire), gradivo iz sporočenih in domnevnih grobnih celot ter skromne keramične in kovinske najdbe z naselbine (npr. kačasta fibula) pa nakazujejo intenzivnejšo poselitev skozi ce­lotno starejšo železno dobo. Pomemben del prazgodovinskega gradiva sodi v (srednji in) pozni laten ter kaže za notranjska gra­dišča običajno podobo. V to obdobje lahko uvrstimo dele notranjsko-kraške in severnojadranske noše (npr. obeski sklepancev s človeško glavo, kakršne so v izteku mlajše železne dobe nosili izključno na Notranjskem,27 primerek notranjske različice certoških fibul vrste VIIf, fibula vrste Kastav različice Ulaka, fibule z rombasto razširjenim lokom in okrog loka ovito tetivo, bronaste pletene ovratnice s tremi vozli, polkrožni rešetkasti okovi vrste Kastav) ter keltski denar (velika srebrnika noriške in tavriške skupine ter zahodnokeltski potin-kov iz zli-tine bakra, kositra in svinca).28 Latensko gradivo odraža tako stike s keltskim Tavriskom pripisanim prostorom mokronoške skupine, na katero je notranjsko-kraška skupina mejila nekje zahodno od Navporta in doline zgornje Krke, kot vplivi viniške (kolapijanske) in ja­podske skupine iz Pokolpja oziroma Like in sosednjih območij.29 Zgodnjo vpetost v sredozemsko trgovino in monetarno gospodarstvo dokazujejo redka najdba grš­kega novca iz 3. st. pr. n. št.,30 viktoriati in drug rimski republikanski denar iz 2. st. pr. n. št.,31 splošne stike s severnoitalsko-rimskim svetom pa odlomka fibule z okvirjasto nogo in čaše vrste Idrija. Fibuli vrste Alezija in Gorica ter pomemben del od več kot 22 svinčenih izstrelkov za pračo gre verjetno povezati z aktivnostmi rimske vojske v predavgustej­skem času, prisotnost rimskih vojakov v avgustejsko- tiberijskem obdobju pa poleg dela omenjenih želodov 26 Gaspari 2000. 27 Božič 1999, 202–203; Laharnar 2012, 180, 217, pril. 1: 8–10. 28 Horvat 1995, 191; Laharnar 2012, 122–129, 217–225, 240. 29 Božič 1999, 202–203; Horvat, Bavdek 2009, 19–20, 131–132; Laharnar 2012, 239. 30 FMRSl V 39-1; Miškec 2008, 290. 31 Miškec 2003, 373. in the vicinity of the church of St Martin in Podcerkev remain unverified and the same goes for the finds from the Tekavča jama cave on the eastern slope above Stari trg. The dynamics of the pre-Roman occupation of Ulaka is thus attested mostly by the finds – both individual and assemblages of artefacts – discovered with metal detectors. One of the more significant assemblages in the storage of the NMS included a considerable amount of Hallstatt material (boat, serpentine, and Certosa fibulae, knobs, larger phalerae, a pendant, bracelets, two glass beads) from the area of the pass north of Ulaka, which probably came from several graves.26 Finds from the wider area of the settlement show a dynamic image and attest to at least occasional presence of humans as early as the Eneolithic (stone arrowhead with a tang) and perhaps the Bronze Age (metal fragments and a piece of a shaft-hole axe), while the material from the reported and assumed burials and the scarce ceramic and metal finds from the settlement (e.g. a serpentine fibula) indicate more intense occupation during the entire Early Iron Age. An important segment of the prehistoric material belongs to the (Middle and) Late La Tene period and does not deviate from what is common in the hillforts of Notranjska. To this period we can date parts of the Notranjska-Kras and North Adriatic dress (e.g. pendants with a human head on metal belts, worn exclusively in the Notranjska region at the end of the Late Iron Age27, a specimen of the Notranjska variant of the Certosa VIIf type fibula, a Kastav type fibula of the Ulaka variant, fibulae with a wire spun around a rhombically taper­ing bow, bronze twisted wire torques with three knots, semi-circular lattice fittings of the Kastav type) and Celtic money (two large silver coins of the Norican and Tauriscan groups and a west Celtic potin coin – minted of copper, tin, and lead alloy).28 The La Tene material reflects contact with the territory of the Mokronog Group, which is attributed to the Celtic tribe of Taurisci. The border between the Mokronog and the Notranjska-Kras Groups was somewhere west of Nauportus and the upper Krka valley. Noticeable is also the influence of the Vinica (Co-lapian) and Iapodic Groups from the Kolpa region, Lika, and the neighbouring areas.29 An early inclusion in the Mediterranean trade and monetary economy is attested by the rare find of a Greek coin from the 3rd century BC,30 and by victoriati and other republican Roman coins from the 2nd century BC.31 General contacts with the north Italian and Roman world are attested by a fragment of 26 Gaspari 2000. 27 Božič 1999, 202–203; Laharnar 2012, 180, 217, App. 1: 8–10. 28 Horvat 1995, 191; Laharnar 2012, 122–129, 217–225, 240. 29 Božič 1999, 202–203; Horvat, Bavdek 2009, 19–20, 131–132; Laharnar 2012, 239. 30 FMRSl V 39-1; Miškec 2008, 290. 31 Miškec 2003, 373. nakazujejo obeski konjske opreme in morda tudi šotor-ski klin.32 Zgodnejša skupina rimskih vojaških najdb je verjetno povezana z obleganjem, ki se je očitno končalo s porazom domorodne skupnosti in morda prekinitvijo življenja v naselbini, morda v povezavi z utrjevanjem vzhodne meje Galije Cisalpine v Cezarjevem času ali – najpozneje – v času Oktavijanovih vojn v Iliriku. Logis­tiko in taktiko napada dokazujejo – poleg lokacij najdb izstrelkov, ki se poleg notranjosti pojavljajo na pobočjih pod severnim delom naselbine33 – zlasti zgodnja faza utrdbe na Nadleškem hribu (sl. 5: 3) in ostanki nekoliko manjšega tabora ali oblegovalnega nasipa dobrih 200 m severozahodno od Ulake (sl. 5: 2). Ob sondiranju in terenskem pregledu jeseni 2017 je bila na delu nasipa, obrnjenem proti gradišču, odkrita precejšnja količina uporabljenih izstrelkov in rimske vojaške opreme, ki brez dvoma pričajo o srditem spopadu med domačini in rimsko vojsko v predavgustejskem obdobju.34 Mlajša skupina najdb morda pomeni sledove zadrževanja rimskih vojakov ali celo manjše posadke v drugi fazi oporišča na Nadleškem hribu v avgustejskem obdobju, ki ga Laharnar povezuje z nadzorom poti do vojnih žarišč v zaledju Kvarnerja in preprečevanjem uporov že pokorjenih skupnosti.35 Še ne dovolj razum­ljeni najdišči z zgodnjerimskim gradivom sta razgledni vrh Križne gore (857 m), od koder izvirajo narebrena skodelica iz stekla jantarne barve z belimi lisami in po­samezne novčne najdbe,36 ter greben Devina z manjšo skupino republikanskih asov in mlajših novcev (sl. 2).37 Prebivalce območja notranjsko-kraške skupine se po uveljavljenem mnenju v najširšem smislu prišteva med Karne, ki so po antičnih literarnih in epigrafskih virih v 2./1. st. pr. n. št. poseljevali hribovito zaledje Akvileje in Tergesta, vključno z območjem prelaza Razdrto (antične Okre). Pri poskusu etnične oprede­litve staroselskih skupnosti v južnem delu Notranjskega podolja kljub temu ni mogoče mimo Strabonovega podatka, da Karni pri Okri mejijo na Japode, in drugo literarno evidenco, ki širše območje naravnih prehodov iz Gorskega kotarja in zaledja Kvarnerja proti Krasu in Tržaškemu zalivu umešča v gospodarsko in vojno-stra­teško interesno sfero Japodov.38 Loška dolina z Ulako in bližnjo naselbino v Šmarati (sl. 2) je po ustalitvi vzhodne meje 10. regije rimske Italije administrativno najverjetneje pripadla agru Emone, ki ga je masiv Snežnika in Javornikov ločil od ozemlja pod jurisdikcijo Tergesta.39 Nedaleč na jugu je potekala meja 32 Laharnar 2012, 240–241; 2015, 24–25, 29–30, t. 3: 1–14. 33 Laharnar 2011, 348–349, t. 2: 14–20. 34 B. Laharnar, osebna informacija, 17. 11. 2017. 35 Laharnar 2015, 94; Laharnar, Lozić 2015–2016, 65–68. 36 Urleb 1968, t. 1: 3; FMRSl I 80. 37 FMRSl V 38, 1–5. 38 Slapšak 2003, 246. 39 Zaccaria 2007, sl. 7. a fibula with a frame-shaped foot and a fragment of an Idrija type beaker. Two fibulae of the Alesia and Gorica types and a significant percentage of more than 22 lead slingshots can probably be associated with the activities of the Roman army in the pre-Augustan period. The presence of Roman soldiers in the Augustan-Tiberian period is indicated by some of the above-mentioned slingshots, by pendants that belong to horse tack, and perhaps by a tent peg.32 An earlier group of Roman military finds is probably related to a siege, which apparently ended with the defeat of the native community and perhaps also with a suspension of life in the settlement, possibly in connection with the fortification of the eastern border of Gallia Cisalpina in the time of Caesar or – at the latest – in the time of the wars of Octavianus in Illyricum. The logistical and tactical aspects of the attack are attested – in addition to the locations of the discovered slingshots, which occur not only in the interior but also on the slopes below the northern part of the settlement33 – by the early phase of the stronghold on Nadleški hrib (Fig. 5: 3) and by the remains of a somehow smaller camp or siege rampart a little more than 200 m northwest of Ulaka (Fig. 5: 2). Test trenching and field survey in autumn 2017 revealed a considerable number of used projectiles and items of Roman military gear on the part of the rampart that faces the hillfort – a doubt­less indication of a skirmish between the natives and the Roman army in the pre-Augustan period.34 A group of later finds might indicate the presence of Roman soldiers or even a small garrison in the second phase of the Nadleški hrib stronghold in the Augustan period. Laharnar associates the stronghold with the control of the routes to war zones in the hinterland of the Kvarner Gulf and the prevention of revolts in the already subdued communities.35 Two sites with early Roman material that are not yet well enough understood are the top of Križna gora (857 m), which offers a good view of the surroundings, and the ridge of Devin. A ribbed cup made of amber glass with white spots and some isolated finds of coins originate from the former,36 and a smaller assemblage of republican asses and later coins from the latter (Fig. 2).37 The prevailing opinion is that the inhabitants of the territory of the Notranjska-Kras Group can be in the broadest possible sense identified as the Carni, who, according to Roman literary and epigraphic sources, inhabited the hilly hinterlands of Aquileia and Tergeste, including the area of Razdrto (Roman Ocra), in the 2nd-1st centuries BC. An attempt at an ethnic identification of the 32 Laharnar 2012, 240–241; 2015, 24–25, 29–30, Pl. 3: 1–14. 33 Laharnar 2011, 348–349, Pl. 2: 14–20. 34 Laharnar, personal information, 17th November 2017. 35 Laharnar 2015, 94; Laharnar, Lozić 2015–2016, 65–68. 36 Urleb 1968, Pl. 1: 3; FMRSl I 80. 37 FMRSl V 38, 1–5. med Italijo ter provincialnima ozemljema Dalmacije (Liburnije) in Panonije (sl. 1); tri glavne administrativne enote so se stikale v globokem zaledju Tarsatike - morda nekje na širšem prostoru Prezida oziroma Gorskega kotarja proti dolini Kolpe. ARHEOLOGIJA ANTIČNE NASELBINE ZASNOVA NASELBINE IN GRADBENE ZNAČILNOSTI STAVB, ODKRI­TIH V OKVIRU SCHMIDOVIH RAZISKAV Med Schmidovimi izkopavanji na Ulaki, ki so verjetno potekala tudi v njegovi odsotnosti in v slabem vremenu, je bilo delno ali v celoti raziskanih 15 objektov (sl. 6; tab. 1), ki so v poročilu opredeljeni kot hiše (1, 2, 3, 4, 9 [prodajalna], 12, 15) in kovačnice (5 - A, 6 - B, 7 - F, 8 - C, 10 - D, 11 - H, 13 - E, 14),40 pri čemer izrecno nava­ja, da večjih stanovanjskih objektov ni raziskoval zaradi zaraščenosti terena in omejenosti finančnih sredstev. Schmidova funkcionalna delitev posameznih objektov je v podrobnostih nejasna, problematična pa je tudi njihova kronološka umestitev, saj je bila pozor­nost izkopavalcev usmerjena predvsem ali izključno v odkrivanje struktur in drobnega gradiva, medtem ko je dokumentiranje višin in stratigrafske provenience najdb očitno v celoti izostalo. Schmidova kronološka opre­delitev posameznih objektov se tako domnevno opira na časovno homogene skupke najdb iz spodnjih delov izkopanih depozitov v bližini prepoznanih struktur, čeprav je razpon predmetov iz posameznim “objektom” pripisanih plasti pogosto bistveno širši. Iz sicer skromne in pomanjkljive dokumentacije Schmidovih izkopavanj izhaja, da so imeli objekti na Ulaki pretežno pravokoten ali kvadraten tloris z zunan­jimi dimenzijami od 4,5 × 4,3 do 16,3 × 5 m. Zasnova stavb se je prilagajala oblikovanosti podlage, pri čemer je bil del objektov postavljen v naravne poglobitve, ostali pa so bili delno ali v celoti vkopani oziroma vsekani v skalno osnovo med 1,2 in 1,66 m globoko (sl. 7; 8). Temelji so bili zgrajeni iz lomljencev, vezanih z malto, nekatere stene pa je tvorila prirejena skalna osnova. Tla v objektih naj bi bila večinoma pokresana in izravnana z glino, v enem primeru pa prekrita z maltnim tlakom. Objekti so imeli en ali dva prostora. V poglobljene dele 40 Številke se nanašajo na Schmidovo številčenje objektov v neobjavljenem poročilu (Schmid 1944), ki se do vključno številke 4 ujema s številčenjem v objavi (Schmid 1937) in na geodetskem načrtu (id. 1939), velike črke pa na oznake kovačnic iz objave in geodetskega načrta, ki jih v poročilu l. 1944 Schmid opusti. Izkopavanja so potekala v naslednjem zaporedju: leto 1936: objekti 1–5; leto 1937: 6, 12 (?), 13; leto 1939: 8–10; leto 1940: 7, 11, 14 (?), 15 (?). Objektov 11 in 14 ni bilo mogoče locirati, identifikacija objekta 7 (kovačnica F) pa je nezanesljiva. native communities in the southern part of the Notran­jsko podolje plain, nevertheless cannot disregard neither Strabo’s information that near Ocra, the Carni bordered on the Iapodes, nor another literary evidence, according to which the wider area of natural passages from Gorski kotar and the hinterland of the Kvarner Gulf towards Kras and the Gulf of Trieste was within the economic and military-strategic interest sphere of the Iapodes.38 After the stabilization of the eastern border of Regio X of Italy, it is most likely that Loška dolina with Ulaka andthe nearby settlement in Šmarata (Fig. 2) administratively belonged to the territory of Emona, which was separated from the area under the jurisdiction of Tergeste by the massifs of Snežnik and Javorniki.39 Not much further to the south ran the border between Italy and the provincial territories of Dalmatia (Liburnia) and Pannonia (Fig. 1); the three major administrative units converging in the deep hinterland of Tarsatica – possibly somewhere in the wider area of Prezid or Gorski kotar, towards the Kolpa valley. ARCHAEOLOGY OF THE ROMAN SETTLEMENT LAYOUT OF THE SETTLEMENT AND CONSTRUCTIONAL PROPERTIES OF THE BUILDINGS DISCOVERED WITHIN THE FRAMEWORK OF SCHMID’S INVESTIGATIONS During Schmid’s excavations on Ulaka, which probably continued uninterrupted even in his absences or in bad weather, 15 buildings were partly or entirely investigated (Fig. 6; Tab. 1). In the report, these buildings are identified as houses (1, 2, 3, 4, 9 [shop], 12, 15) and smithies (5 - A, 6 - B, 7 - F, 8 - C, 10 - D, 11 - H, 13 - E, 14).40 It is explicitly stated that larger residential build­ings were not investigated due to overgrown terrain and limited funds. Schmid’s functional classification of buildings is unclear in details; similarly problematic is their dating, since the attention of the excavators was focused predomi­nantly or exclusively on discovering structures and small finds, while there seems to have been no documentation 38 Slapšak 2003, 246. 39 Zaccaria 2007, Fig. 7. 40 The numbers refer to Schmid’s numbering of buildings in his unpublished report (Schmid 1944). Up to and includ­ing the number 4, they match the numbers in the publication (Schmid 1937) and in the survey map (id. 1939). Capital let­ters refer to the smithies in the publication and in the map, but Schmid no longer used them in his 1944 report. The buildings were excavated in the following order: Buildings 1–5 in 1936: Buildings 6, 12 (?), 13 in 1937; Buildings 8–10 in 1939; Buildings 7, 11, 14 (?), 15 (?) in 1940. Buildings 11 and 14 could not be located, and the identification of Building 7 (Smithy F) is unreliable. Sl. 7: Ulaka. Schmidova izkopavanja. Hiša 1: A – nakovalo in ješa; B – ješa; D – ognjišče; C – hodnik; E – tlak; S – prag (po Schmid 1937, 18–19, sl. 3). Fig. 7: Ulaka. Schmid‘s excavations. House 1: A – anvil and forge; B – forge; D – fireplace; C – hallway; E – pavement; S – doorstep (after Schmid 1937, 18–19, Fig. 3). Sl. 8: Ulaka. Schmidova izkopavanja. Hiša 5/Kovačnica A: A – kovačnica; B –predverje; C – hodnik s stopnicami; H – ognjišče; a – kotlinica z varom; b – odpadki ognjišča in var; d, e – stopnica; f – prizidek; h – mala amfora; l, k, n – var (po Schmid 1937, 20–22, sl. 7). Fig. 8: Ulaka. Schmid’s excavations. House 5/Smithy A: A – smithy; B – foyer; C – hallway with stairs; H – fireplace; a – hollow with forge sand; b – fireplace waste and forge sand; d, e – stair; f – annex; h – small amphora; l, k, n – forge sand (after Schmid 1937, 20–22, Fig. 7). objektov so vodile kamnite ali lesene stopnice. Nad­gradnja stavb, ki jih je Schmid interpretiral kot delavnice, naj bi bila izdelana v kladni tehniki, o čemer je sklepal na podlagi kosov glinastega premaza z odtisi lesenih leg in palic. Ohranjeni kosi maltnega ometa z odtisi trsja, šibja in desk bi lahko podpirali izkopavalčeve navedbe o poslikanih stenah “prodajaln” in “manjših stanovanjskih hiš” (2, 3, 4, 9, in 15). Tegule naj bi prekrivale samo vodni zbiralnik na jugovzhodnem delu naselbine, glavnina stavb pa je bila domnevno krita s skodlami ali s slamo. Tloris in način gradnje objektov na Ulaki ne od­stopata od značilnosti železnodobnega stavbarstva na Notranjskem41 in v Posočju, ki se ohranijo še v pozno­rimski čas in mlajša obdobja. Natančnejša podoba (pol) vkopanih, eno- ali dvonadstropnih stavb s suhozidno drenažo oz. temelji iz lomljencev, glinenimi, kamni­timi ali lesenimi tlemi ter v kombinaciji vodoravnega temeljnega bruna, pokončnih lesenih stebrov in dvoj­nega opaža grajenimi stenami, ki kažejo sorodnosti s t. i. retijskim tipom hiše iz vzhodnega dela Centralnih Alp in južnoalpskih dolin, je v širšem prostoru znana npr. z višinske naselbine Castelraimondo v Furlaniji.42 V skalna tla vklesani objekti z Ulake imajo na Notranjskem primerjavo na železnodobni naselbini in zgodnjerimski postojanki na vzpetini Sovič nad Postojno, kjer je bil oktobra 2017 odkrit podoben, v karbonatno podlago poglobljen objekt v izmeri približno 3 × 2 m, prelimi­narno datiran v 2./1. st. pr. n. št.43 Apneno maltno vezivo za zidove in stenski omet ter opečna strešna kritina se na nekaterih naselbinah s kontinuiteto poselitve iz pra­zgodovine kot novost rimskega obdobja pojavita že na začetku principata, na drugih pa šele v srednjecesarskem obdobju ali pozneje. Tlorisna zasnova rimske naselbine, ki jo je mogoče razbrati na lidarskem posnetku (sl. 5), prikazuje bolj kompleksno situacijo od poenostavljenega Schmidovega načrta z objekti, ki se vrstijo ob osrednji osi oz. ulici (sl. 6). Razvidni so tako samostojni objekti kot gruče in vrste stavb, ki mestoma ustvarjajo videz rastra tesno skupaj postavljenih prostorov v radialnih nizih oziroma blokih z opazno tendenco koncentričnega širjenja od osrednjega, očitno nepozidanega (trškega) dela naselbi­ne, proti zunanjemu obodu. Bloke in gruče stavb ločujejo bolj ali manj široki presledki (morda v funkciji prehodov in protipožarnih razmikov) in nepozidana območja. Po analogiji z naselbino na Mostu na Soči bi lahko nize stavb opredelili kot obrtne delavnice, samostojno stoječi objekti pa bi lahko imeli stanovanjski značaj. Mreža ulic oziroma komunikacij v smeri vhodov, od katerih se dva zarisujeta na zahodnem delu, eden pa na jugovzhodnem delu naselbine, je slabo čitljiva. Najbližji znani regionalni primerjavi za višinski tip antične naselbine s podobno 41 Horvat 1995, 185–186. 42 Santoro Bianchi 1992. 43 Za posredovane podatke se zahvaljujem Manci Oma-hen, vodji izkopavanj (Avgusta d. o. o.). of elevations or the stratigraphic origin of the finds. Schmid’s dating of individual buildings is thus supported by chronologically homogeneous assemblages of finds from the lower parts of the deposits excavated in the vicin­ity of the identified structures, although the time span of artefacts from layers attributed to individual “buildings” is often much greater. The scarce and insufficient documentation of Schmid’s excavations reveals that most of the buildings on Ulaka had a rectangular or square floor plan with outside dimensions from 4.5 m × 4.3 m to 16.3 m × 5 m. The layout of the buildings was adapted to the configura­tion of the ground with some of the buildings placed in natural hollows and others either partially or fully dug into the bedrock between 1.2 and 1.66 m deep (Fig. 7; 8). The foundations were made of roughly cut stones bound with mortar; some of the walls were formed out of modified bedrock. The floors of the buildings were mostly chiselled and levelled with clay; in one case it was furnished with a mortar pavement. The buildings had one or two rooms. Stone or wooden stairs led to the underground parts of buildings. The superstructures of the buildings interpreted as workshops were supposedly baulk wall constructions; Schmid concluded this on the basis of fragments of clay coating with impressions of wooden posts and sticks. Preserved fragments of mortar coating with impressions of reeds, wicker, and planks could support the excavator’s claims that “shops” and “small residential houses” (2, 3, 4, 9, 15) had painted walls. Only the cistern in the south­eastern part of the settlement was covered by tegulae, and most of the buildings are assumed to have been covered with wooden shingles or thatch. Floor plans and mode of construction do not deviate from the characteristic features of Iron Age construction in the Notranjska41 and Soča regions, which survived even to the Late Roman period and beyond. These (half) dug one- or two storey buildings with dry stone drainage and foundations made of roughly cut stones; with clay, stone, or wooden floor; and with walls constructed with the combination of a horizontal foundation beam, vertical wooden posts, and double panelling, show similarities to the so-called Raetian-type houses from the eastern part of the Central Alps and from Southern Alpine val­leys. In the wider area, a clearer image of them is known e.g. from the hilltop settlement of Castelraimondo in Friuli.42 In Notranjska, the buildings on Ulaka, carved into the rocky ground, can be compared to the Iron Age settlement and Early Roman outpost on the hill of Sovič above Postojna, where a similar building, dug into the carbonate bedrock, measuring approximately 3 × 2 m and preliminarily dated to the 2nd-1st century BC, was discovered in October 2017.43 In some of the settlements 41 Horvat 1995, 185–186. 42 Santoro Bianchi 1992. 43 Information courtesy of Manca Omahen, who direct­ed the excavation (Avgusta d. o. o.). strukturirano zazidavo območja nekdanjega prazgo­dovinskega gradišča so Ajdovščina nad Rodikom,44 Gradišče pri Knežaku,45 Kerin nad Pivko, Gradišče nad Trnovim v Ilirski Bistrici46 in Sv. Pavel nad Vrtovinom.47 Lokacija grobišča prebivalcev naselbine ni zane­sljivo ugotovljena. Nagrobnika, vzidana v cerkvi sv. Martina v Podcerkvi, naj bi bila po Danskem (1857) odkrita na območju nekropole na Ulaki. Plane žgane grobove severno in južno od Podcerkve sicer omenja Žmavc, vendar ne poda njihove časovne opredelitve.48 KRONOLOGIJA V okviru celotnega korpusa najdb dinamiko rim-skodobne poselitve Ulake najbolje odraža več kot 500 evidentiranih rimskih novcev, ki naselbino uvrščajo med podeželska središča z izdatneje dokumentiranim denarnim obtokom. V skupini zgodnejših določljivih numizmatičnih najdb (sl. 9)49 izstopata precejšnje število rimskega republikanskega denarja iz 2. st. pr. n. št. (skup-no 10 novcev, med njimi denarij, kvinarij, trije viktoriati in pet asov) in skromnejša prisotnost srebrnih novcev iz 1. st. pr. n. št., zastopanih s kovom iz leta 68 pr. n. št. in morda še s štirimi denariji, kovanimi v letih 49/48, 48 in 32/31 pr. n. št.50 Sledita zmeren dotok denarja v 1. st. n. št. in močan porast v drugi tretjini 2. st., ki ustrezata glavnini kronološko oprijemljivega gradiva iz Schmidovih izkopavanj. Med objekti s prepričljivejšimi sklopi kronološko homogenih najdb izstopajo hiša 5 (kovačnica A) (sl. 8) s kompletom kovaškega orodja (sl. 13: 2–4), ki bi jo bilo mogoče na podlagi Klavdijevega novca, amfor in dvoročajnih skodel s precej pridržka umestiti v sredino ali drugo polovico 1. st., ter hiša 13 (kovačnica E), ki jo rozetna fibula ter zakladna najdba novcev, deponirana po 173/174,51 postavljajo v zadnjo tretjino 2. stoletja. Sočasni ali morda nekoliko mlajši predmeti iz istega objekta vključujejo ožgano dvokril-no pasno spono, gumb in peltast okov (sl. 10: 1,2,4), ki imajo dobre analogije v sklopih iz druge polovice 2. in prve polovice 3. st.52 Ponoven in izrazit višek je v novčnem obtoku do-kumentiran šele po sredini 3. st., ki mu sledita stabilen 44 Slapšak 1997; Kokalj, Hesse 2017, 14. 45 Laharnar 2012, 66–67; Laharnar, Lozić, Miškec 2020, v tej knjigi; Kokalj, Hesse 2017, 56–57. 46 Laharnar, Lozić, Miškec 2020. 47 Ciglenečki 2016, 420–421, sl. 2. 48 Urleb 1968, 479. 49 FMRSl I 84/1; FMRSl III 56; FMRSl IV 42; FMRSl V 39; FMRSl VI 57. Za posredovane podatke se zahvaljujem Alenki Miškec iz Numizmatičnega kabineta Narodnega muzeja Slo­venije. 50 Kos 1986, 29 (evidentirani pod najdiščem Stari trg pri Ložu (FMRSl I 83/1, 1–4). 51 FMRSl I 84/2; Kos 1986, 85. 52 Gaspari 2000, 52; Laharnar 2015, 22–23, t. 3: 7–10. with continuous occupation from prehistory, lime mortar binder for walls, wall plaster, and brick roof tiling appear as a novelty of the Roman period as early as the begin­ning of the principate, and in others only in the Middle Imperial period or later. The layout of the Roman settlement, as it can be detected from the lidar image (Fig. 5), shows a more complex situation than Schmid’s simplified plan with buildings lined along the central axis/street (Fig. 6). Inde­pendent buildings and clusters of them can be identified, sometimes creating the impression of a raster of rooms closely against each other in radial rows or blocks and with a noticeable tendency to spread concentrically from the unbuilt central (commercial) part of the settlement towards the outer edge. Blocks and clusters of buildings are separated by more or less wide gaps (perhaps func­tioning as passageways or fire safety gaps) and unbuilt areas. Judging by the analogy with the settlement in Most na Soči, the rows of buildings could be identified as craft workshops, while the independently standing buildings could have been of residential character. The poorly visible grid of streets/passages is oriented towards the entrances, two of which can be seen in the western part, and one in the southeastern part of the settlement. The nearest known regional comparisons for a Roman hilltop settlement with a similarly structured built-up area that used to be a prehistoric hillfort are Ajdovščina above Rodik,44 Gradišče near Knežak,45 Kerin above Pivka, Gradišče above Trnovo in Ilirska Bistrica,46 and Sv. Pavel above Vrtovin.47 The location of the cemetery that belonged to the settlement has not been identified with certainty. Ac­cording to Danski (1857), the two tombstones built into the church of St Martin in Podcerkev were discovered in the area of the necropolis on Ulaka. Žmavc mentions flat cremation graves north and south of Podcerkev, but gives no dating for them.48 CHRONOLOGY Within the framework of the entire corpus of finds, the dynamics of the Roman settlement on Ulaka is best reflected by more than 500 documented Roman coins. This means the settlement on Ulaka is among the countryside centres with better-documented monetary circulation. The most prominent among the identifiable early numismatic finds (Fig. 9)49 are a considerable num­ 44 Slapšak 1997; Kokalj, Hesse 2017, 14. 45 Laharnar 2012, 66–67; Laharnar, Lozić, Miškec 2020, in this book; Kokalj, Hesse 2017, 56–57. 46 Laharnar, Lozić, Miškec 2020. 47 Ciglenečki 2016, 420–421, Fig. 2. 48 Urleb 1968, 479. 49 FMRSl I 84/1; FMRSl III 56; FMRSl IV 42; FMRSl V 39; FMRSl VI 57. The author would like to thank Alenka Miškec Sl. 9: Ulaka. Strukturiranost posamičnih najdb (do­ločljivih) novcev (podatki: Numizmatični kabinet Narodnega muzeja Slovenije). Fig. 9: Ulaka. Composition of the attributable single coin finds (data: the Numismatic Cabinet of the National museum of Slovenia). Sl. 10: Ulaka. Schmidova izkopavanja. Bronasti deli vojaške opreme. Hiša 8 / Kovačnica C: 1 – gumb (pr. 3,6 cm). – Hiša 13 / Kovačnica E: 2 – okov v obliki pelte (d. 3,4 cm); 3 – gumb; 4 – spona (d. 3,5 cm). Vse predmete hrani NMS. Fig. 10: Ulaka. Schmid‘s excavations. Bronze elements of milita­ry attire. House 8 / Smithy C: 1 – knob (diameter 3.6 cm). – Ho­use 13 / Smithy E: 2 – peltate fitting (lenght 3.4 cm); 3 – knob; 4 – buckle (lenght 3.5 cm). All artefacts are kept by the NMS. dotok v prvi polovici 4. st. ter povečano število denarja iz sredine in druge polovice 4. st. Obtok poznorimskega obdobja, ki se zaključuje z dvema Teodozijevima novce-ma in dvema kovoma iz let 388–403 in ki kaže podob­nost z dinamiko najdišč iz okvira obrambnega sistema Claustra Alpium Iuliarum,53 dopolnjujejo posamično odkrita jermenski zaključek amforičaste oblike in okov v obliki propelerja iz okvira vojaške opreme54 ter 22,3 cm visok bronast zvonec z ovalnim ušesom iz hiše 5 (kovačnica A) (sl. 11). Slednjega lahko glede na najdbe podobnih večjih zvoncev z višinskih naselbin (Unec, Polhograjska Gora)55 in drugih najdišč (npr. založna najdba 6 zvoncev iz bližnjega Gorenjega Jezera; sl. 2)56 previdno povežemo z obstojem sistema signalnega obveščanja. Najdiščne okoliščine tremisa iz obdobja druge vlade Zenona (480–491), o kateri prvi poroča P. Hitzinger, niso sporočene,57 morebitne druge najdbe, ki 53 Glej Kos 2012. 54 Laharnar 2015, 24, t. 3: 13–14. 55 Božič 2005, 317–318. Podatek Mija Topličanec, Zavod za varstvo kulturne dediščine Slovenije, Območna enota Ljubljana; hrani No-tranjski muzej Postojna. 57 Božič, Ciglenečki 1995, 262; FMRSl IV 42, 132. ber of Roman republican coins from the 2nd century BC (10 coins altogether, among them a denarius, a quinarius, three victoriati, and five asses) and a smaller number of silver coins from the 1st century BC, represented by a coin minted in 68 BC, and perhaps also by four denarii, minted in the years 49/48, 48, and 32/31 BC.50 There followed a moderate monetary influx in the 1st century AD, and a strong increase in the second third of the 2nd century, which includes the majority of the chronologically at­tributable material from Schmid’s excavations. Among the buildings with more convincing assemblages of chronologically homogeneous finds are House 5 (Smithy A) (Fig. 8), which contained a set of blacksmith’s tools (Fig. 13: 2–4), and can be, with much reserve, dated to the mid­dle or second half of the 1st century, based on a Claudius’ coin, amphorae, and two-handled bowls; and House 13 (Smithy E), which is dated to the last third of the 2nd cen­tury, based on a rosette fibula and a coin hoard, deposited after 173/174.51 Contemporary or possibly slightly later objects from the same building include a burned bilobate belt buckle, a knob, and a peltate fitting (Fig. 10: 1,2,4), which have good analogies in the assemblages from the second half of the 2nd and first half of the 3rd century.52 The next prominent peak in coin circulation is documented only after the middle of the 3rd century, fol­lowed by a stable influx in the first half of the 4th century from the Numismatic Cabinet of the National Museum of Slovenia for this information. 50 Kos 1986, 29. Documented below the site of Stari trg pri Ložu (FMRSl I 83/1, 1–4). 51 FMRSl I 84/2; Kos 1986, 85. 52 Gaspari 2000, 52; Laharnar 2015, 22–23, Pl. 3: 7–10. bi nakazovale rabo območja naselbine ob koncu 5. ali v začetku 6. stoletja, pa niso znane. PREDELAVA KOVIN IN DRUGE GOSPODARSKE DEJAVNOSTI Schmid, zanesljivo eden največjih poznavalcev prazgodovinskega in antičnega železarstva svojega časa, je sledove predelave kovin prepoznal skoraj v vseh izkopanih objektih na Ulaki, naselbina pa naj bi izstopala ravno po velikem številu kovačnic. Odkrito kovaško orodje, ognjišča, talilniki, strjevalne posode, bronasta in železna žlindra ter najdbe več istovrstnih železnih orodij v resnici govorijo za razvito predelavo kovin in podpirajo tezo o Ulaki kot trškem središču lokalne skupnosti. Železova žlindra, ki jo je na območju naselbine že pred Schmidom dokumentiral A. Müllner, in omembe na površini ležeče železove rude v ravnini pod hribom, ki jo je bilo po imenovanem raziskovalcu v sredini 18. stoletja dovolj za predelavo v cesarski železarni v Čabru,58 ne dopuščajo dokončne presoje, ali so se na Ulaki oziroma v njeni neposredni okolici odvijale vse faze predelave železa, ali pa samo izdelovanje predmetov iz drugod pripravljene surovine in občasno taljenje rude. Arheološki sledovi železopredelovalne dejavnosti v agru Emone59 in v zaledju Tergesta v splošnem ne odstopajo od širše dokumentiranega vzorca iz severnih delov im­perija, v okviru katerega so sledovi železarstva v sekun­darnih aglomeracijah in mestnih naselbinah pretežno omejeni na ostanke predelave surovega (kovnega) železa v končne izdelke, intenzivno primarno pridobivanje že­leza oziroma taljenje železove rude z večjimi količinami žlindre in talilnimi pečmi pa je potekalo predvsem v podeželskih obratih.60 Obe glavni fazi predelave ali samo kovaške delavnice, praviloma namenjene popravilom, so dokumentirane zlasti v rudarskih vikusih,61 naselbinah ob glavnih komunikacijah62 in vilah rustikah. Taljenje rude je zaradi lažje logistike (dostava rude in kuriva), protipožarne varnosti in sanitarnih razlogov potekalo na robu ali izven strnjenih naselbin.63 58 Müllner 1909, 54. 59 Glej Inkret 2013, 61–69. 60 Polfer 2000, 74–77. 61 Rothenhöfer 2005, 65–68; Burnham, Wacher 1990, 41. 62 Na območju Notranjske je v tem oziru zanimiva večja količina železnega orodja, vključno z nakovalom, delov vo­zov in konjske opreme ter drugih predmetov, datiranih v čas od 1. do 4. st., s širšega območja najdišča Vodice pri Kalcah (Pflaum 2007), kjer bi smeli domnevati vsaj postajo in pre­pregališče (mutatio) v okviru poštnega sistema (cursus publi­cus) na glavni cesti Emona–Aquileia. Npr. večje količine železove žlindre v kulturni plasti s tremi novci iz 4. st. na Cesti 5. maja v Ajdovščini, pripisa­ne obrtniškim delavnicam jugozahodno od trdnjave Castra (Osmuk 1977, 198, sl. 50); mineraloška analiza vzorcev 25 kg and an increased amount of money in the middle and second half of the 4th century. The circulation of the Late Roman period, which ends with two Theodosius’ coins and two coins from the years 388–403, and which shows similarity to the dynamics of the sites of the Claustra Alpium Iuliarum defence system,53 is complemented by a separately discovered amphora-shaped strap end and a propeller-shaped fitting that belong to military gear,54 as well as a 22.3 cm high bronze bell with an oval ear from House 5 (Smithy A) (Fig. 11). The latter could, judging by similar finds of larger bells from hilltop settlements (Unec, Polhograjska Gora)55 and other sites (e.g. a hoard of 6 bells from the nearby Gorenje Jezero; Fig. 2),56 with some reservation relate to the existence of a signalling system. The discovery circumstances of a tremissis from the second reign of Zeno (480–491), first reported by Hitzinger, were not documented.57 Other hypothetical finds that could indicate that the area of the settlement was in use at the end of the 5th or the beginning of the 6th century, are not known. METALWORKING AND OTHER ECONOMIC ACTIVITIES Schmid, certainly one of the greatest connoisseurs of prehistoric and Roman iron working of his time, identi­fied traces of metal working in almost all the excavated buildings in Ulaka and the settlement supposedly had a particularly high number of smithies. Blacksmith­ing tools, fireplaces, melting and consolidation vessels, bronze and iron slag, and also several iron tools of the same type, indeed indicate developed metal working and support the thesis that Ulaka was a trade centre of the local community. Even before Schmid, iron slag was documented in the area of the settlement by Müllner. According to him, there was enough surface iron ore in the plain below the hill to be processed in the imperial iron works in Čabar in the middle of the 18th century.58 This, however, still provides no definite answer of whether all stages of iron working were present on Ulaka (or in its immediate sur­roundings), or was Ulaka just the place where objects were manufactured from the materials brought from elsewhere, and a location of occasional ore smelting. Archaeological traces of iron working in the territory of Emona59 and in the hinterland of Tergeste generally do not deviate form 53 See: Kos 2012. 54 Laharnar 2015, 24, Pl. 3: 13–14. 55 Božič 2005, 317–318. 56 Information by Mija Topličanec, Institute for the Pro­tection of Cultural Heritage of Slovenia, Ljubljana Regional Office; kept by Notranjski muzej Postojna. 57 Božič, Ciglenečki 1995, 262; FMRSl IV 42, 132. 58 Müllner 1909, 54. 59 See: Inkret 2013, 61–69. Količina in raznovrstnost sledov predelave in obdelave kovin Ulako odmikata od manjših obratov v okviru hišnih gospodarstev vil64 in manjših naselbin, v katerih so kovaško dejavnost opravljali njihovi prebivalci ali potujoči rokodelci, in podpirata domnevo o obstoju specializiranega metalurškega centra s profesionalnimi kovači (ferrarii oz. fabri).65 Podobno intenzivna kovaška in metalurška dejavnost na rodiški Ajdovščini, izpričana z razmeroma pogostimi najdbami orodja in železove žlindre ter podprta z rezultati geofizikalnih raziskav,66 je domnevno v veliki meri povezana s ponovnim raz­cvetom naselbine v poznorimskem obdobju. Kovaška ognjišča na Ulaki si lahko na podlagi izkopavalčevih podatkov predstavljamo večinoma kot odprta ognjišča na do 0,75 m visokih in 0,42 × 0,56 m do 1,85 × 1,2 m velikih kamnitih podstavkih pravokotne (stavbe 5, 7, 8, 12) ali okrogle (stavbi 6, 10) oblike, ki so bila opremljena s šobami za dovod zraka z mehovi (citati na sl. 7; 8). Posamezna ognjišča naj bi bila obokana, z zaokroženo zadnjo steno iz kamenja in žgane gline ter oblogo iz kapnikov oziroma sige, odporne proti visokim temperaturam. Pri rekonstrukciji oblike in delovanja opisanih ognjišč se je Schmid opiral na ikonografske vire o antičnih kovačnicah.67 Okoli ognjišč, ki so bila opremljena s kamnitimi podstavki in kladami z zglajeno zgornjo površino, inter-pretiranimi kot podstavki za nakovala ali delovna mesta, naj bi navadno ležale saje, var, prežgana ilovica in železna žlindra. Podobo značilnega železopredelovalnega obrata v hiši 5 zaključujeta komplet tipično kovaškega orodja (kladivo,68 klešče69 in pila70; sl. 13: 2–4) ter vdolbina (sl. 8: a), ki jo morda lahko glede na ureditev iz zgodnje­cesarskih delavnic na Štalenski gori (Magdalensberg)71 interpretiramo kot vkopano kovaško ognjišče oz. talilno peč. Bronasta žlindra in drugi ostanki predelave brona dopuščajo domnevo, da imamo na Ulaki opravka z de­ žlindre, odkrite med novejšimi izkopavanji bližnje lokacije na Cesti 5. maja, je pokazala, da gre za sekundarno, kovaško žlindro in pri enem vzorcu za prečiščen surovec (Kramar et al. 2015); plast s kovaškim odpadom je starejša od zidanih ostankov arhitekture, datirane v srednje- in poznocesarsko obdobje (Tratnik, Žerjal 2017, 252, 263, sl. 6 in 7). 64 Npr. sklop železnega orodja in kovinskega odpada, morda namenjenega reciklaži, iz vile Školarice pri Dekanih; podatek Tina Žerjal, avgust 2017. 65 V Porenju so podobni obrati za intenzivno pridobi­vanje surovega železa in njegovo nadaljnjo predelavo značilni za 4. st. (Polfer 2000, 76–77; Rothenhöfer 2005, 68). 66 Glej Murgelj 2000. 67 Gaspari 1998–1999; k antičnim kovačnicam in nji­hovim ostankom v jugovzhodnoalpskem prostoru Dolenz 1997, 37–48. 68 Pietsch 1983, 23–25, sl. 26, t. 6: 93; Gaitzsch 1980, 87– 89; Dolenz 1997, 159–160, t. 49: W1. 69 Gaitzsch 1980, 230–231. 70 Gaitzsch 1980, 49–50; Pietsch 1983, 49–50; Dolenz 1997, 172–173. 71 Dolenz 1997, 47. a more widely documented pattern known from the northern parts of the empire, where traces of iron work­ing in secondary agglomerations and urban settlements are mostly limited to the traces of processing pig iron into final products, while intensive primary processing of iron or smelting of iron ore with larger amounts of slag and in smelting furnaces took place predominantly in countryside establishments.60 Either both main phases of iron processing or just blacksmithing workshops, usu­ally intended for repairs, are most often documented in mining vici,61 in settlements on the main routes,62 and in villae rusticae. For easier logistics (ore and fuel delivery), fire safety, and for sanitary reasons, ore smelting took place at the edges of agglomerations or outside them.63 The number and variety of traces of metal working and processing set Ulaka apart from smaller establish­ments functioning within the framework of domestic economy of villae,64 and from smaller settlements, where blacksmithing was performed by the inhabitants or by travelling craftsmen; and support the hypothesis of the existence of a specialised metallurgical centre with pro­fessional blacksmiths (ferrarii or fabri).65 On Ajdovščina above Rodik, similarly intense blacksmithing and metal­lurgical activities, attested by relatively numerous tools and iron slag and further supported by the results of geophysical surveys,66 are supposedly closely related to the second period of flourishing of the settlement in the Late Roman period. Based on the excavator’s information, blacksmiths’ forges on Ulaka can be envisioned as mostly open forges, 60 Polfer 2000, 74–77. 61 Rothenhöfer 2005, 65–68; Burnham, Wacher 1990, 41. 62 In the territory of Notranjska, a larger amount of iron tools, including an anvil, parts of wagons and horse tack, and other objects, dated to the time between the 1st and the 4th centuries, was discovered in the wider area of the site Vodice near Kalce (Pflaum 2007), where at least a changing station (mutatio) within the postal service system (cursus publicus) on the main road from Emona to Aquileia could be assumed. 63 E.g. larger amounts of iron slag in a cultural layer with three coins from the 4th century in Ajdovščina (Cesta 5. maja), attributed to craft workshops southwest of the Cas­tra fort (Osmuk 1977, 198, Fig. 50); a mineralogical analysis of the samples of 25 kg of slag discovered during the recent excavations in the nearby site of Cesta 5. maja has proved that this is secondary slag and in one sample a purified ingot (Kramar et al. 2015); the layer with blacksmithing waste is earlier than the remains of masonry structures dated to the Middle and Late Imperial periods (Tratnik, Žerjal 2017, 252, 263, Figs. 6 and 7). 64 E.g. the assemblage of iron tools and metal waste, perhaps intended to be recycled, from villa Školarice near Dekani; information by Tina Žerjal, August 2017. 65 In the Rhine region, similar establishments for in­tensive production of pig iron and its further processing are typical of the 4th century (Polfer 2000, 76–77; Rothenhöfer 2005, 68). 66 See: Murgelj 2000. Sl. 12: Ulaka. Schmidova izkopavanja. Odlomki preluknjane stene ognjišča iz Hiše 5 / Kovačnice A in Kovačnice 9. Vse predmete hrani NMS. Fig. 12: Ulaka. Schmid‘s excavations. Fragments of the perfo­rated clay plates of the fireplace from the House 5 / Smithy A and Smithy 9. All artefacts are kept by the NMS. (Foto / Photo: Tomaž Lauko, NMS). lavnicami, v katerih so predelovali tako železo kot bron.72 Predelava različnih kovin se je lahko odvijala na istem ali pa na ločenih ognjiščih. Nepojasnjena ostaja funkcija številnih fragmentov preluknjanih glinenih plošč z 1,5 do 4,5 cm debelim ostenjem in gosto razmeščenimi okroglimi odprtinami premera 2,5 do 4,5 cm ter odtisi lesenih letev notranjega ogrodja (sl. 12). Te naj bi po Schmidu sestavljale pred­vsem ostenja manjših ognjišč na nizkem podstavku, ki so bila odprta le z ene strani, kar pa je vprašljivo, saj tolikšno število zagotovo ni bilo potrebno za napeljavo cevi iz meha, nasprotno, v tem primeru bi onemogočalo doseganje potrebnih temperatur v ješi. Edine konkretne primerjave ponujajo bronastodobne in železnodobne ognjiščne rešetke,73 zato velja v pričakovanju oprijemlji­vejših najdb v kontekstu tudi ostanke plošč s predrtinami 72 Polfer 2000, 76; Rothenhöfer 2005, tab. 1. Za analogije iz severnojadranskega in jugovzhodno­predalpskega prostora glej npr. Lonza 1981, 74–80, t. 44–48; Bolta 1960, 282, t. I: 6; Dular, Križ 2004, t. 11: 16; Vinazza 2016, 14–15. K okroglim dvodelnim prenosnim pečem, na­menjenih predvsem pripravi hrane, in njihovim funkcional­nim značilnostim glej Coulon 2015. up to 0.75 m high, and placed on either rectangular (Buildings 5, 7, 8, 12) or round (Buildings 6, 10) stone bases with the dimensions between 0.42 × 0.56 m and 1.85 × 1.2 m. They were equipped with nozzles for the supply of air from the bellows (citations in Fig. 7; 8). The fireplaces were supposedly vaulted, with a rounded back wall made of stones and baked clay, and a lining made of dripstone or flowstone, resistant to high temperatures. Schmid based his reconstruction of the form and func­tion of the described forges on the iconographic sources on Roman smithies.67 Soot, welded metal, fired clay, and iron slag were usually found around the fireplaces, which were equipped with stone bases and blocks with a smoothed upper sur­face, interpreted as the bases for anvils or for working places. The image of a typical iron working establishment in House 5 is completed by a set of typical blacksmith’s tools (a hammer,68 pincers,69 a file;70 Fig. 13: 2–4) and a hollow in the ground (Fig. 8: a), which might be – by analogy with the Early Imperial workshops at Magdalens­berg71 – interpreted as a blacksmith’s forge of a smelting furnace, dug into the ground. Bronze slag and other remains of bronze working allow for the hypothesis that both iron and bronze were worked in the workshops of Ulaka.72 Different metals could be worked either in the same forge or in separate forges. The function of numerous fragments of 1.5 to 4.5 cm thick perforated clay plates with densely arranged circular openings with diameter between 2.5 and 4.5 cm and imprints of wooden battens of the interior framework (Fig. 12) remains unexplained. According to Schmid, most of these plates were the walls of smaller fireplaces on low bases and open only from one side. This interpretation, however, is questionable, for the number of openings was certainly too high if they were meant for the hoses from the bellow – they would actually prevent reaching the necessary temperatures in the forge. Since the only relevant comparisons are fireplace grates from the Bronze and Iron Ages,73 the remains of perforated plates from Ulaka, which bear absolutely no traces of molten metal or slag, can be – in absence of better identifiable finds in con­ 67 Gaspari 1998–1999; for Roman smithies and their remains in the Southeastern Alpine area, see: Dolenz 1997, 37–48. 68 Pietsch 1983, 23–25, Fig. 26, Pl. 6: 93; Gaitzsch 1980, 87–89; Dolenz 1997, 159–160, Pl. 49: W1. 69 Gaitzsch 1980, 230–231. 70 Gaitzsch 1980, 49–50; Pietsch 1983, 49–50; Dolenz 1997, 172–173. 71 Dolenz 1997, 47. 72 Polfer 2000, 76; Rothenhöfer 2005, Tab. 1. 73 Fort the analogies from the northern Adriatic and Southeastern Alpine areas, see e.g. Lonza 1981, 74–80, Pls. 44–48; Bolta 1960, 282, Pl. I: 6; Dular, Križ 2004, Pl. 11: 16; Vinazza 2016, 14–15. For round double portable kilns, in­tended predominantly for food preparation, and their func­tional properties, see: Coulon 2015. z Ulake, na katerih ni najmanjših sledov taline ali žlindre, obravnavati kot dele premičnih pravokotnih rešetk pri odprtih kuhinjskih ognjiščih ali pečeh za pripravo hrane, izdelanih v prazgodovinski tradiciji.74 Z rabo v kovačnicah je Schmid povezal tudi številne glinene svitke in jih interpretiral kot podstavke za talilne lončke. Na Ulaki je iz njegovih izkopavanj ohranjenih kar 53 svitkov ali njihovih delov, ki jim moramo prišteti še najdbe iz poznejših raziskav in nadzorov posegov na robnih delih naselbine (interpretiranih kot zavrženo gradivo iz starejših izkopavanj). Svitki iz objektov, ki jih je izkopal Schmid, skoraj zagotovo ne morejo biti rezi­dualna najdba in jih je mogoče z nekaj pridržka postaviti v rimski kontekst, sicer pa so značilen in pogost del na­selbinskih inventarjev iz časa med pozno bronasto dobo in koncem latenskega obdobja.75 V rimskem obdobju so svitki dokumentirani v zelo omejenem številu, praviloma v manjših podeželskih naselbinah (zgodnje)cesarskega časa. Prevladujoča razlaga svitke pojasnjuje kot pod-stavke za posode pri kuhi na ognjišču, dokumentirani konteksti iz mlajšehalštatskih faz lončarskega obrata v objektih 15A in 23 na Mostu na Soči, kjer so sestavljali mreže za ločevanje izdelkov v peči,76 pa dokazujejo njihovo rabo tudi v obrtnem ambientu. Zbir železnega orodja sestavljajo izključno dol­gotrajne oblike, ki so bile razširjene predvsem med 1. in 3. stoletjem, nekatere pa se skoraj nespremenjene upo­rabljajo še v poznorimsko obdobje. V starejšo skupino bi lahko poleg omenjenega kovaškega pribora (sl. 13: 2–4) z nekaj pridržka uvrstili vsaj še masivno drvarsko sekiro z okrepljenim čelom (sl. 13: 1) in vejnik s tulastim nasadiš-čem (sl. 13: 5). Izstopa skupina šestih tesel zelo podobnih oblik in dimenzij, ki jih zaznamuje zaobljeno upognjen list ter uho z ovalno luknjo za pritrditev na toporišče in polkrožno zvišanim ostenjem ter kratko kladivasto čelo (sl. 14). Podobna tesla z velikim, včasih skoraj pravim kotom med listom in toporiščem, in kladivastim čelom, ki jih ločuje od sicer zelo podobnih kopač, so običajen sestavni del depojskih najdb na Notranjskem77 in v širši regiji. Kronološko oprijemljivejše primerjave iz taborov renskega limesa kažejo na težišče njihove uporabe v 74 K podstavkom za kuhanje na ognjiščih in tipom zi­danih ognjišč in pečic v Pompejih glej Foss 1994, 18, 78–64. V okviru sorodnih kuhinjskih pripomočkov iz italskega am-bienta velja omeniti npr. podstavek-stojalo za kuhanje v obli­ki perforiranega okroglega pladnja (pr. okoli 25 cm) na treh nizkih nogah iz poznohelenistične faze (pribl. 125–80/75 pr. n. št.) naselbine Tel Anafa v dolini Hula, ki ima 18 okroglih odprtin premera 3 cm (Berlin 1993, 40–41, sl. 3, 4); splošno h kuhinjskim podstavkom glej Scheffer 1981. 75 Npr. Horvat 1995, 188; Horvat, Bavdek 2009, 122–123. 76 Svoljšak 2014, 293; Svoljšak, Dular 2016, 166–174. 77 Gaspari et al. 2000, 198, sl. 10: 25,26; 11: 28,33; Pflaum 2007, 302, t. 3: 21,22. Schmid poroča o najdbi več identičnih orodij, ki jih sicer opredeli za kopače (Hauen), na Šilentaboru (Schmid 1944, 46). Sl. 13: Ulaka. Schmidova izkopavanja, Hiša 5 / Kovačnica A. Železno orodje: 1 – sekira (d. 22,7 cm); 2 – klešče (d. 46,4 cm); 3 – kladivo (d. 17,2 cm); 4 – pila (d. 30,8 cm); 5 – vejnik (d. 26,4 cm). Vse predmete hrani NMS. Fig. 13: Ulaka. Schmid‘s excavations. House 5 / Smithy A. Iron tools: 1 – axe (lenght 22.7 cm); 2 – pincers (lenght 46.4 cm); 3 – hammer (lenght 17.2 cm); 4 – file (lenght 30.8 cm); 5 – billhook (lenght 26.4 cm). All artefacts are kept by the NMS. (Foto / Photo: Tomaž Lauko, NMS). text – understood as parts of movable rectangular grates in open kitchen fireplaces or ovens for food preparation, manufactured in the prehistoric tradition.74 The numerous discovered clay rings were also asso­ciated with smithies and interpreted as stands for smelt­ing pots, according to Schmid. No less than 53 surviving clay rings or their fragments were discovered on Ulaka during his excavations, not counting the finds from later investigations and archaeological monitoring in the areas along the edges of the settlement (interpreted as discarded material from earlier excavations). Clay rings from the buildings excavated by Schmid almost certainly cannot be residual finds and with some reservation they can be 74 For stands for cooking in fireplaces and for types of masonry fireplaces and ovens in Pompeii, see: Foss 1994, 18, 78–64. Within the framework of similar kitchen utensils from the Italic milieu, we should mention e.g. the stand for cooking pots in the form of a perforated round tray (diam­eter about 25 cm) on three low feet from the Late Hellenis­tic phase (approximately 125–80/75 BC) from the Tel Anafa settlement in the Hula valley, which has 18 round openings with diameter of 3 cm (Berlin 1993, 40–41, Figs. 3, 4); for kitchen stands in general, see: Scheffer 1981. Sl. 14: Ulaka. Schmidova izkopavanja. Železne tesle: 1 – Hiša 8 / Kovačnica C (d. 17,3 cm); 2 – Hiša 8 / Kovačnica C (d. 19,2 cm); 3 – Hiša 13 / Kovačnica E (d. 17,5 cm); 4 – Hiša 8 / Kovačnica C (d. 17,4 cm); 5 – Hiša 5 / Kovačnica A (d. 16,2 cm); 6 – Hiša 6 / Kovačnica B (d. 16,3 cm). Vse predmete hrani NMS (foto: Tomaž Lauko, NMS). Fig. 14: Ulaka. Schmid‘s excavations. Iron adzes: 1 – House 8 / Smithy C (lenght 17.3 cm); 2 – House 8 / Smithy C (lenght 19.2 cm); 3 – House 13 / Smithy E (lenght 17.5 cm); 4 – House 8 / Smithy C (lenght 17.4 cm); 5 – House 5 / Smithy A (lenght 16.2 cm); 6 – House 6 / Smithy B (lenght 16.3 cm). All objects are kept by the NMS (photo: Tomaž Lauko, NMS). srednjecesarskem in poznorimskem obdobju,78 ob zelo verjetnem zgodnejšem izvoru te oblike tesla.79 Verjetna serijska produkcija tesel na Ulaki govori o tem, da so bile tamkajšnje kovačnice specializirane za izdelavo orodja za potrebe naselbine in bližnja tržišča, upoštevajoč možne okoliščine dokumentiranih viškov naselbine med drugo polovico 2. in koncem 4. st. pa ni izključena niti njihova vpetost v dobavo materiala za vojsko. Različni tipi sekir in tesel, vključno z manjkajočo, 15 cm dolgo križno sekiro z zvišanima stenama ovalnega ušesa iz hiše 8 (kovačnice C),80 vejniki, žaga in sveder dokazujejo izkoriščanje gozdov (Schmid omenja oglje jelke in bora), o poljedelstvu pa pričajo večje količine različnih poljščin v posameznih prostorih (pšenica,žito, grah, proso, ječmen in čičerka). Živinoreja je arheološko oprijemljiva z najdbami kosti domačih živali (omenjene so kosti goveda, drobnice in svinj), manjšega (zgodnjecesarskega) bronastega zvonca s petkotnim ušesom in železnih verig. Nekatere zvrsti namiznega posodja imajo izrazit lokalni značaj, kot npr. dvoročajne skodele (sl. 15) s številnimi primerjavami 78 Pietsch 1983, 25–28, tip III in IV. 79 Npr. zgodnjerimski grob 3 iz Reke pri Cerknem (glej Guštin 1991, 64, t. 31: 2). 80 Schmid 1944, 32. placed in the Roman context. Clay rings are generally typical common items in settlement inventories between the Late Bronze Age and the end of the La Tene period.75 In the Roman period, clay rings are documented in a very limited number, usually in small countryside settlements from the (Early) Imperial period. According to the pre­vailing interpretation, clay rings were placed under pots while cooking in the fireplace. Documented contexts from the late Hallstatt phases of the pottery workshop in Buildings 15A and 23 in Most na Soči, where clay rings were used in grids for separating products in kilns,76 are evidence for their use in craft workshops as well. Iron tools include, without exception, long lasting forms, most common between the 1st and 3rd centuries. Some of them were still used almost unchanged in the Late Roman period. In addition to the above-mentioned blacksmithing tools (Fig. 13: 2–4), the earlier group could, with some reservations, also include a massive woodcutting axe with a reinforced butt (Fig. 13: 1) and a socketed billhook (Fig. 13: 5). Prominent among them is a group of six adzes of very similar forms and dimensions, characterized by a rounded bent head, an oval socket with raised sides for attaching a wooden haft, and a short hammer-shaped butt (Fig. 14). Similar adzes with a wide 75 E.g. Horvat 1995, 188; Horvat, Bavdek 2009, 122–123. 76 Svoljšak 2014, 293; Svoljšak, Dular 2016, 166–174. Sl. 15: Ulaka. Schmidova izkopavanja, Hiša 2. Odlomki dvoročajnih skodel (pr. ustij: 17–22 cm; P 12682). Predmete hrani NMS. Fig. 15: Ulaka. Schmid‘s excavations, House 2. Fragments of the two handled bowls (diameter of rims: 17–22 cm; P 12682). Artefacts are kept by the NMS. (Foto / Photo: Tomaž Lauko, NMS). iz grobov na Svinji Gorici v Cerknici,81 naselbinskih plasti v Dolenjem Logatcu (mansio Longatico)82 in struge Ljubljanice na Ljubljanskem barju.83 Velika količina najdenih odlomkov nakazuje možen obstoj lončarske delavnice iz druge polovice 1. st. ali 2. st., ki je zalagala zlasti naselbine vzdolž komunikacije proti Emoni. Maloštevilnost okvirno sočasnih najdb uvožene fine namizne keramike in transportnega posodja ne dopušča celovitejše ocene o vpetosti naselbine v trgovske tokove, ohranjene oblike amfor za oljčno olje (Dressel 6B) in kosi terre sigillate (skodele tipa Sarius, krožniki oblike Consp. 20 ali 21 in skodelice oblike Consp. 27 ali 29) pa pričakovano ustrezajo širše dokumentirani navezanosti notranjsko-kraškega območja na istrsko in severnoital­sko produkcijo. 81 Urleb 1983, t. 3: 1; 13: 4; 18: 2; 21: 6; 27: 1. 82 Šinkovec 2020, v tej knjigi. 83 Gaspari 2003, 177. – sometimes almost right – angle between the blade and the haft, and a hammer-shaped butt, which distinguishes them from the very similar hoes, are common items in hoards in Notranjska77 and the wider region. Better dated analogies from the camps of the Rhine limes indicate that the high point of their use were the Middle Imperial and Late Roman periods,78 while this form of a adzes is prob­ably of earlier origin.79 The fact that adzes were probably serially manufactured on Ulaka suggests that the smith­ies of Ulaka were specialized in the manufacture of tools for the needs of the settlement and the nearby markets. Taking into account the possible circumstances of the documented peaks between the second half of the 2nd century and the end of the 4th century, it is not impos­sible that they were included in the supply of materials for the army. Different types of axes and adzes, including the miss­ing 15 cm long pickaxe with an oval shafthole with raised sides from House 8 (Smithy C),80 billhooks, a saw, and a drill are evidence for logging (Schmid mentions fir and pine charcoal). Considerable amounts of different types of crops (wheat, cereals, peas, millet, barley, chickpeas) in some of the rooms attest to agricultural activity. Live­stock breeding is reflected in the archaeological record by bones of domestic animals (cattle, sheep and goats, pigs), a small (Early Imperial) bronze bell with a pentagonal ear, and iron chains. Some types of tableware are of a distinct local character, e.g. two handled bowls (Fig. 15) with numerous analogies in the graves of the Svinja Gorica cemetery in Cerknica,81 in settlement layers in Dolenji Logatec (mansio Longatico),82 and in the riverbed of the Ljubljanica in the Ljubljansko barje.83 A large number of discovered fragments suggests a possible existence of a pottery workshop in the second half of the 1st century or in the 2nd century, which supplied the settlements along the route towards Emona. The low number of roughly contemporary items of imported fine tableware and transport vessels does not allow for a more comprehen­sive estimation of the manner in which the settlement was included in trade currents. The surviving forms of amphorae for olive oil (Dressel 6B) and fragments of terra sigillata (Sarius type bowls, plates of the Consp. 20 or 21 types, the Consp. 27 or 29 types cups) correspond – as expected – to the more widely documented reliance of the Notranjska-Kras region on the Istrian and North Italian production. 77 Gaspari et al. 2000, 198, Figs. 10: 25,26; 11: 28,33; Pflaum 2007, 302, Pl. 3: 21,22. Schmid reported the discovery of sev­eral identical tools, identified as hoes (Hauen), at Šilentabor (Schmid 1944, 46). 78 Pietsch 1983, 25–28, Types III and IV. 79 E.g. the Early Roman Grave 3 from Reka near Cerkno (see: Guštin 1991, 64, Pl. 31: 2).80 Schmid 1944, 32. 81 Urleb 1983, Pls. 3: 1; 13: 4; 18: 2; 21: 6; 27: 1. 82 Šinkovec 2020, in this book. 83 Gaspari 2003, 177. STATUS, PREBIVALCI IN RAZVOJ NASELBINE Središčni značaj prazgodovinske naselbine, ki sta jo Ulaki zagotavljala položaj in velikost za poselitev primernega vršnega dela vzpetine ob vozlišču pokrajin­skih komunikacij, poleg pomembne količine gradiva iz mlajšega in poznega halštata ter (srednjega in) poznega latena potrjujejo tudi sledovi domnevnega rimskega ob-leganja v sredini ali že prvi polovici 1. st. pr. n. št. Vlogo osrednje naselbine v Loški dolini in tem delu Notranjske Ulaka očitno obdrži tudi v cesarskem času, morda s krajšo prekinitvijo v avgustejsko-tiberijskem obdobju, ko je bila na vršnem platoju morda prisotna rimska po­sadka. Najpozneje v sredini 1. st. n. št. se Ulaka razvije v prostorsko urejeno naselje “trškega, skoraj mestnega videza”,84 ki ga zaznamujejo kompleksen raster pozidave ter poudarjeno obrtni značaj z gospodarskimi objekti in posameznimi stanovanjskimi stavbami z elementi rimskega bivalnega habitusa. Kontinuiteta in intenzivnost poselitve Ulake pričata o gospodarski moči prebivalcev, najverjetneje poveza­ni z razvito metalurško dejavnostjo in gozdarstvom, katerih pomen je zelo verjetno presegal lokalne okvire. Gospodarsko rabo obsežnih gozdov hribovitega zaled­ja Tergesta, Foruma Iulii in Emone,85 v okviru katere smemo domnevati pridobivanje lesa za gradnjo stavb in ladjedelništvo (tako za civilne kot vojaške potrebe), kurjave, oglja, smol idr., potrjuje posredna in neposred­na literarna,86 epigrafska87 in arheološka evidenca.88 V pravno-administrativnem smislu je položaj lokalne skupnosti prešel od statusa civitas foederata ali tributaria iz časa neposredno po osvojitvi do položaja (gens) adtributa z dodelitvijo upravnemu centru (najver­jetneje Emoni; res publica Emonensium) po konsolidaciji rimske oblasti, pri čemer naj bi – podobno kot ostale sta­roselske skupnosti v robnih delih nekdanje Cisalpinske Galije – communis opinio,89 verjetno za dlje časa ostala v juridičnem položaju peregrinov (peregrini).90 Središčni značaj naselbine s poudarjeno obrtno (in trgovsko?) funkcijo indicira možnost, da je Ulaka uživala enega od rimskih pravno-administrativnih statusov naselbin 84 Schmid 1944, 45. 85 Razen območij, ki so ležala neposredno ob zasebnih posestih, in drugih izjem, so bila pašna območja in gozdovi po osvojitvi kot nerazdeljeno javno dobro (ager scriptura­rius) še v 1. in 2. st. n. št. bodisi zadržani pod neposrednim nadzorom države ali dodeljeni v upravljanje določeni mestni skupnosti (saltus publicus). V okviru slednjega so bili gozdovi deloma izkoriščani za javne gradbene projekte, zagotavljanje kurjave za ogrevanje term ipd., deloma pa prepuščena v rabo vsem pripadnikom skupnosti. 86 Vitr. 2, 9, 15 in 16 (glej Šašel 1981). 87 Inscr. It. X, 4, 340 (Ajdovščina). 88 Npr. sekira s pečatnikom iz Ljubljanice (Turk et al. 2009a, 258–259, kat. št. 54). 89 Prim. Faoro 2015, 189–191. 90 Glej Zaccaria 1992, 156; 2007, 136. STATUS, INHABITANTS, AND DEVELOPMENT OF THE SETTLEMENT Ulaka’s character of a central prehistoric settlement, a consequence of its location and the size of the top plateau, which was suitable for habitation and lay near the junction of regional communications, is attested by a significant amount of material from the younger and late Hallstat and (middle and) late La Tene periods, and further confirmed by the traces of the supposed Roman siege in the mid­dle of the 1st century BC or perhaps even in its first half. Even in the Imperial period, Ulaka apparently retained its role of the central settlement in Loška dolina and that part of Notranjska. There was a possible minor hiatus in the Augustan-Tiberian period, when a Roman garrison might have been present on the top plateau. No later than in the middle of the 1st century AD, Ulaka developed into a settlement with a regular plan and “the look of a market town, almost a city,”84 characterized by a complex raster pattern of buildings and a distinct air of craft activities, with workshops and individual residential houses with elements of the Roman residential customs. The continuity and intensity of occupation on Ulaka attest to the economic power of its inhabitants, prob­ably related to the developed metallurgical and logging activities, the significance of which very likely exceeded the local context. The use of the vast forests in the hilly hinterland of Tergeste, Forum Iulii, and Emona,85 within the frame of which we may expect logging of wood for the construction of buildings and shipbuilding (for both civil­ian and military purposes), fuel, charcoal, resins, etc., is attested directly and indirectly by literary,86 epigraphic,87 and archaeological evidence.88 In a legal-administrative sense, the status of the local community transformed from civitas foederata or tribu­taria from directly after the conquest, to (gens) adtributa, when it was attributed to an administrative centre (most likely Emona; res publica Emonensium) after the consoli­dation of the Roman rule. Not unlike other autochthonous communities in the outer parts of the former Gallia Cis­alpina, communis opinio,89 the inhabitants probably kept 84 Schmid 1944, 45. 85 Except for the areas directly bordering private prop­erty and some other exceptions, after the conquest, pasture areas and forests were undistributed public land (ager scrip-turarius) and as late as the 1st and 2nd centuries either kept under the direct control of the state or allocated to town communities (saltus publicus). Within the framework of the latter, forests were partly used for public construction proj­ects, for fuel to heat thermae, etc., and partly given in use to all the members of a community. 86 Vitr. 2, 9, 15 and 16 (see: Šašel 1981). 87 Inscr. It. X, 4, 340 (Ajdovščina). 88 E.g. axe with a stamp from the Ljubljanica (Turk et al. 2009a, 258–259, Cat. no. 54). 89 Cf. Faoro 2015, 189–191. nižjega ranga,91 najverjetneje vikusa,92 katerega prebi­valstvo (populus) so sestavljali zlasti (peregrinae) incolae, t.j. člani lokalne neitalske skupnosti na ozemlju Italije,93 pričakovana pa je tudi zgodnja prisotnost državljanov (negotiatores ali mercatores) italskega porekla. Ob tem velja ponoviti Schmidovo opažanje o izstopajoče skrom­ni uporabi fibul,94 ki jo je izkopavalec povezal z bližino Italije,95 medtem ko kulturni pripadnosti domačinov precej jasno pričajo domača osebna imena na nagrob­nikih iz okolice. Ohranjen napis (CIL III 3782 = 10720) v cerkvi Sv. Martina prinaša imena dveh oseb z rimskim državljan­stvom – po vsej verjetnosti pripadnikov iste družine, s psevdogentilnim imenom Caedagonius ([. C]aedagonius [A]ttus in Q(uintus) Caedago/[nius ---]),96 ki je izpeljano iz nekega domačega, morda keltskega(?),97 imena. Na štirih zgodnjecesarskih nagrobnikih iz Šmarate, ki je oddaljena slabe tri kilometre (sl. 2), nastopajo izključno svobodni domačini peregrinega statusa, pri katerih je navedeno enojno osebno ime in ime očeta. Vsa imena98 sodijo v severnojadransko imensko skupino, kamor sodijo tudi venetska, histrijska in liburnijska osebna imena.99 Uporaba uvoženega nabrežinskega apnenca za izdelavo nagrobnika Planija Sekstilija in razvitejšaimenska struktura prebivalcev po M. Šašel Kos kažeta tako na zgodnjo akulturacijo tega prostora kot na večje premoženje in višji socialni status nekaterih tam živečih družin. Podobno kot se to domneva za ostale atribuirane staroselske skupnosti v robnih predelih nekdanje Galije Cisalpine, so imeli tudi lokalni primores zelo verjetno že zelo zgodaj pravno urejeno pravico prenosljive zasebne 100 posesti (ager privatus) in morda tudi ius commercii. Razcvet naselbine proti sredini 2. st. in z viškom v njegovi tretji četrtini, ki presevata skozi novčni obtok, bi morda smeli – tudi glede na najdbe pogojno sočasnih delov vojaške noše – previdno povezovati s prisotnostjo mobilne posadke za nadzor in varovanje prehoda v 91 K antičnim definicijam in označbam za nemestne, po­deželske naselbine glej Strobel 2016, 31–35. 92 Schmid je v naselbini na Ulaki prepoznal središče upravnega okraja (pagus) in ji pripisal status opiduma ali vi-kusa (Schmid 1944, 46). 93 Gagliardi 2006, 102–104. 94 Dokumentirane najdbe fibul iz 1./2. st. vključujejo ne­kaj močno profiliranih fibul tipa Almgren 67 ali 68, izgub­ljen primerek mlajše (?) različice očesnih fibul, manjkajočo okroglo fibulo s predrtino in še eno okroglo emajlirano fi­bulo. 95 Schmid 1944, 46. 96 Prepis: [. C]aedagonius /[A]ttus hic iacet / [an]nor(um) xxv men(sium) X / d(ierum) VIII Q(uintus) Caedago-/[nius ---] / -----­ 97 Dellamare 2007, 52, s.v. Caedagonius. 98 Feucon ali Feuconts, Planius Sextilius Feucontis f., Sex-tilia Tatsoria, Turoius Pletor Feucontis f., Pletor Poteiius Feu­contis f., Volta Lassonia, Pletor Poteiius Plani f., Turoius Nepos. 99 Šašel Kos 2000. 100 Glej Faoro 2015, 184–184. the legal status of peregrini for a long time.90 The central character of the settlement with an emphasis on its craft (and trade?) function suggests the possibility that Ulaka enjoyed one of the Roman legal-administrative statuses for lower-tier settlements,91 probably a vicus,92 the in­habitants (populus) of which were mostly (peregrinae) incolae, i.e. members of a local non-Italic community in the area of Italy.93 We can also expect an early presence of the citizens (negotiatores or mercatores) of Italic origin. Here, Schmid’s observation about the very low use of fibulae94 should be repeated – the excavator attributed it to the vicinity of Italy,95 while the cultural affiliation of the locals is quite clearly attested by native personal names on the tombstones in the vicinity. The inscription (CIL III 3782 = 10720) in the church of St Martin features the names of two people with Roman citizenship, possibly members of the same family, with the pseudogentile name of Caedagonius ([. C]aedagonius [A] ttus in Q(uintus) Caedago/[nius ---]),96 which is derived from some native, perhaps Celtic (?)97 name. Four Early Imperial tombstones from the less than 3 km distant villageof Šmarata (Fig. 2) mention only free native inhabitants of peregrine status, with a single personal name and the name of their father. All the names98 belong to the North Adriatic name group, which includes Venetic, Histrian, and Liburnian personal names.99 According to Šašel Kos, the use of the imported Aurisina limestone for the tombstone of Planius Sextilius and the developed name structure of the inhabitants indicate an early acculturation of the area and large wealth and higher social status of some of the families who lived there. In a similar fashion as it is assumed for other attributed autochthonous communi­ties in the border areas of the former Gallia Cisalpina, it is quite likely that the local primores had a legal right of transferable private property (ager privatus) and perhaps also ius commercii at a very early stage.100 90 See: Zaccaria 1992, 156; 2007, 136. 91 For Roman definitions for non-urban, rural settle­ments, see: Strobel 2016, 31–35. 92 Schmid saw the settlement on Ulaka as the centre of an administrative district (pagus) and assumed it had the status of an oppidum or a vicus (Schmid 1944, 46). 93 Gagliardi 2006, 102–104. 94 The documented finds of fibulae from the 1st/2nd cen­tury include some heavily profiled fibulae of the Almgren 67 or 68 type, a lost later (?) variant of eye fibulae (Augenfibel), a missing round fibula with a perforation, and another round enamelled fibula. 95 Schmid 1944, 46. 96 Transcription: [. C]aedagonius /[A]ttus hic iacet / [an] nor(um) xxv men(sium) X / d(ierum) VIII Q(uintus) Caed­ago-/[nius ---] / -----­ 97 Dellamare 2007, 52, s.v. Caedagonius. 98 Feucon or Feuconts, Planius Sextilius Feucontis f., Sex-tilia Tatsoria, Turoius Pletor Feucontis f., Pletor Poteiius Feu­contis f., Volta Lassonia, Pletor Poteiius Plani f., Turoius Nepos. 99 Šašel Kos 2000. 100 See: Faoro 2015, 184–184. Italijo čez podolji Prezidskega in Babnega polja.101 Strateški pomen širšega območja v zaledju Tarsatike je razviden tudi iz začasne vzpostavitve prokuratorske province Liburnije v izteku 2. st. (184/185), ki velja za zadnji člen v uresničevanju zasnove obrambe Italije102 v okviru kratkotrajne vojaško-administrativne cone pra­ententura Italiae et Alpium iz časa vpadov Markomanov in Kvadov čez Alpes Iuliae v letih 168/169 in 171/172.103 Ta sistem naj bi bil sicer razformiran že v zgodnjih 80-ih letih 2. stoletja, vendar so nekateri sestavni deli obram-be vzhodnoalpskih prehodov obstali, kar nakazujejo Herodianova poročila o dogajanjih ob pohodih vojska Septimija Severa (193) in Maksimina Tračana (238) iz Panonije proti Italiji.104 Schmid je vzroke za domneven propad naselbine v 2. stoletju, ko naj bi bil del objektov uničen v ognju, pripisal Markomanskim vojnam in njihovim posledi-cam, pri čemer se je naslonil na skupno najdbo novcev in sledove požara v kovačnici 13 (E).105 Izrazit upad v novčnem obtoku na naselbini sicer nastopi šele z obdobjem vladavine Karakale in traja vse do sredine 3. st., kar sovpada z okvirno sočasnim upadom števila novcev na nekaterih drugih rimskih višinskih naselbi­nah notranjsko-kraškega območja.106 Upad posamičnih novčnih najdb v Emoni in celo njihovo odsotnost v severozahodnem delu mesta ter druge kazalce izrazitega zmanjšanja števila emonskega prebivalstva v prvi polo-vici 3. st. je Peter Kos hipotetično povezal s posledicami epidemije kuge, ki je v času vladavine Komoda prizadela zlasti Rim in Akvilejo, epigrafsko pa je dokumentirana tudi v Noriku.107 Naselbina na Ulaki se po Schmidu ponovno opo-more v 3. st., del izkopanih kovačnic pa naj bi obratoval v 4. stoletje. Povečan dotok novcev iz druge polovice 3. st. bi lahko pomenil določeno vlogo naselbine v novo organiziranem sistemu obrambe vzhodne meje Italije pred vpadi vedno bolj številčnih barbarskih voj-ska. Ta sistem so v drugi polovici 3. st. poleg utrdb ob pomembnejših komunikacijah v prostoru med Vipav­ 101 Glej Višnjić 2016, 22, 28, sl. 14, 23. 102 Blečić 2001, 78–79. 103 Šašel 1974. 104 Šašel, Petru 1971, 23–24. 105 Schmid 1944, 45; dokumentiran novčni obtok v ju­govzhodnoalpskem prostoru v času med in neposredno po Markomanskih vojnah po P. Kosu ne nakazuje katastrofalne­ga opustošenja in trajnejše prekinitve kulturno-ekonomske­ga razvoja celotnega območja, temveč prej govori za uničenje in krajšo prekinitev v poselitvi (Celeia) ali kratkotrajne stiske (Emona) v življenju mest ob vpadnicah napadov (Kos 1986, 83–91, 98). 106 Podobna cezura v obtoku je dokumentirana za Ja­vor pri Dolnjem Zemonu – grobišče (FMRSl I 76; FMRSl IV 36) in Ajdovščino nad Rodikom (FMRSl I 42; FMRSl III 26; FMRSl IV 16), ni pa zaznavna npr. na Gradišču nad Kneža­kom (FMRSl IV 39/1). 107 Kos 1986, 101–104. The flourishing of the settlement towards the end of the 2nd century with the peak in its third fourth, which is reflected in the coin circulation, might be – this is also supported by the possibly contemporary items of mili­tary dress – carefully associated with a mobile garrison controlling and protecting the passage to Italy across the lowlands of Prezidsko polje and Babno polje.101 The strategic significance of the wider area in the hinterland of Tarsatica is even more evident from the temporary estab­lishment of a procuratorial province of Liburnia towards the end of the 2nd century (184/185), which is considered the last step in the realization of the plan for the defence of Italy102 within the framework of a short-term military-administrative zone praententura Italiae et Alpium in the time of the incursions of the Marcomanni and Quadi across Alpes Iuliae in 168/169 and in 171/172.103 While this system was supposedly abandoned as early as the early 180s, some constituent parts of the defence of the Eastern Alpine territories remained, as indicated by Herodian’s reports on the events surrounding the marches of the armies of Septimius Severus (193) and Maximinus Thrax (238) from Pannonia towards Italy.104 Schmid attributed the reasons for the supposed co-lapse of the settlement in the 2nd century, when some of the buildings were destroyed in a fire, to the Marcomannic wars and their consequences, based on a hoard of coins and traces of fire in Smithy 13 (E).105 A strong decline in coin circulation in the settlement happened only during the rule of Caracalla and lasted until the middle of the 3rd century, which coincides with the roughly contemporary decline in the number of coins in some other Roman hilltop settlements in the Notranjska-Kras region.106 Ac­cording to Peter Kos, the decline in individual coin finds in Emona and even their absence in the northwestern part of the town, as well as other indicators of a strong decline in the population of Emona in the first half of the 3rd century, might have been related to the aftermath of a plague epidemic in the time of Commodus, which 101 See: Višnjić 2016, 22, 28, Figs. 14, 23. 102 Blečić 2001, 78–79. 103 Šašel 1974. 104 Šašel, Petru 1971, 23–24. 105 Schmid 1944, 45; according to Kos, the documented coin circulation in the Southeastern Alpine area in the time during and directly after the Marcomannic wars does not suggest a catastrophic devastation with a more permanent interruption of the cultural and economic development in the entire region, but rather speaks of a destruction and a short hiatus in occupation (Celeia) or a short-term crisis (Emona) in the life of the towns along the lines of the attacks (Kos 1986, 83–91, 98). 106 A similar hiatus in circulation is documented in Javor near Dolnji Zemon – cemetery (FMRSl I 76; FMRSl IV 36) and in Ajdovščina above Rodik (FMRSl I 42; FMRSl III 26; FMRSl IV 16), but it has not been detected e.g. in Gradišče above Knežak (FMRSl IV 39/1). sko dolino, ilirskobistriškim območjem in Notranjsko sestavljale tudi opazovalne ali signalne postojanke na višinskih lokacijah, dokumentirane z naselbinskimi in grobnimi najdbami.108 Številni poznorimski novci in omenjena kosa moške noše (jermenski zaključek in pasni okov) kažejo na razcvet naselbine v času delovanja obrambnega siste-ma Claustra Alpium Iuliarum, ko je Loška dolina ležala v neposrednem zaledju prve obrambne črte sektorja Snežnik–Babno polje–Benete. Križišče komunikacij na severnem delu omenjenega dela zapor je domnevno obvladovala utrdba na Križni gori (sl. 2), morda štabna točka tega sektorja, z dvojnim obzidjem in drobnimi najdbami iz časa med drugo polovico 4. st. in 6./7. st.109 Gradivo z Ulake in naselbinske najdbe z bližnje Šmarate, med katerimi je po ugotovitvah Verene Vidrih Perko v opaznih količinah zastopano importirano, predvsem afriško transportno posodje, značilno za vojaško oskrbo med koncem 4. in sredino 5. st.,110 morda pričajo tako o sodelovanju tamkajšnjih prebivalcev oziroma posadk v logistični podpori delovanja zapor kot poznejši vklju-čenosti v sistem v globino razprostrte obrambe Italije (tractus Italiae circa Alpes).111 Zahvale Za pomoč in nasvete pri pripravi prispevka ter posredovanje podatkov in slikovnega gradiva se lepo zahvaljujem Janki Istenič, Boštjanu Laharnarju in Alenki Miškec ter Poloni Bitenc, Tomislavu Kajfežu in Tomažu Lavku iz Narodnega muzeja Slovenije, Milanu Lovenjaku iz Filozofske fakultete Univerze v Ljubljani, Miji Topličanec, višji konservatorki Zavoda za varstvo kulturne dediščine Slovenije, Območna enota Ljubljana, in Manci Omahen, arheologinji v podjetju Avgusta. 108 Horvat, Žbona Trkman 2016, 117. 109 FMRSl I 80; Urleb 1968; 1974; Ciglenečki 1985, 268– 269; Vidrih Perko 1994, 159–164; Modrijan 2015, 25, sl. 6–7.110 Perko, Bavdek, Lazar 1998, 276–279. 111 Šašel, Petru 1971, 100; Kos 1986, 204; glej Ciglenečki 2016. hit Rome and Aquileia the hardest, and is attested by epigraphic evidence also in Noricum.107 According to Schmid, the settlement on Ulaka recovered in the 3rd century, and some of the excavated smithies supposedly continued into the 4th century. An increased influx of coins from the second half of the 3rd century could indicate that the settlement had a specific role in the newly organized defence system of the eastern border of Italy against the raids of ever more numerous barbarian armies. In the second half of the 3rd century, this system included fortifications along the major communi­cation routes in the area between the valley of Vipavska dolina, the area of Ilirska Bistrica, and Notranjska; as well as viewing or signalling stations on elevated locations, documented by finds from settlements and burials.108 Numerous Late Roman coins and two pieces of male dress (a strap end and a belt fitting) indicate the flourish­ing of the settlement in the time of the Claustra Alpium Iuliarum defence system, when the valley of Loška dolina was in the direct hinterland of the first defence line of the Snežnik–Babno polje–Benete sector. The crossroads of the routes in the northern part of the described area of the barriers was supposedly controlled by the fort on the hill of Križna gora (Fig. 2). With a double defensive wall and small finds from the time between the second half of the 4th century and the 6th/7th centuries, Križna gora might have been the sector headquarters.109 The material from Ulaka and the finds from the nearby settlement in Šmarata, which contain – according to the findings of Verena Vidrih Perko – considerable quantities of imported, predominantly Af­rican transport vessels, typical of military supply between the end of the 4th and the middle of the 5th century,110 might indicate that the local inhabitants or garrisons were involved in the logistic support of the Claustra, and later included in the system of the deep-reaching defence system of Italy (tractus Italiae circa Alpes).111 Acknowledgements For their help and advice in the writing of this contribu­tion and for sharing of data and images, the author is grateful to Janka Istenič, Boštjan Laharnar, Alenka Miškec, Polona Bitenc, Tomislav Kajfež, and Tomaž Lauko from the National Museum of Slovenia, Milan Lovenjak from the University of Ljubljana’s Faculty of Arts, Mija Topličanec, senior conserva­tor from the Ljubljana Regional Office of the Institute for the Protection of Cultural Heritage of Slovenia, and archaeologist Manca Omahen from company Avgusta d.o.o. Translation: Meta Osredkar 107 Kos 1986, 101–104. 108 Horvat, Žbona Trkman 2016, 117. 109 FMRSl I 80; Urleb 1968; 1974; Ciglenečki 1985, 268– 269; Vidrih Perko 1994, 159–164; Modrijan 2015, 25, Figs. 6–7. 110 Perko, Bavdek, Lazar 1998, 276–279. 111 Šašel, Petru 1971, 100; Kos 1986, 204; cf. Ciglenečki 2016. Okrajšave / Abreviations FMRSl I = P. Kos, Die Fundmünzen der römischen Zeit in Slowenien I (Berlin 1988). FMRSl II = P. Kos, A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien III (Berlin 1995). FMRSl IV = A. Šemrov, P. Kos, Die Fundmünzen der römischen Zeit in Slowenien IV (Berlin 1998). FMRSl V = A. Šemrov, Die Fundmünzen der römischen Zeit in Slowenien V (Mainz am Rhein 2004). FMRSl VI = A. Šemrov, P. Kos, Die Fundmünzen der römischen Zeit in Slowenien VI (Wetteren 2010). Inscr. It. X, 4 = Inscriptiones Italiae, vol. X – regio X, fasciculus IV – Tergeste, curavit Petrus Sticotti, Roma 1951. BAŠ, F. 1951, Rajniki. Walter Schmid. – Slovenski etnograf 20–21, 386–387. BERLIN, A. 1993, Italian Cooking Vessels and Cuisine from Tel Anafa. – Israel Exploration Journal 43, 35–44. BLEČIĆ, M. 2001, Prilog poznavanju antičke Tarsatike (Un appunto di conoscenza di Tarsatica antica). – Vjesnik Ar-heološkog muzeja u Zagrebu 34, 65–122. BOLTA, A. 1960, Rifnik. – Varstvo spomenikov 7, 282. BOŽIČ, D. 1999, Die Erforschung der Latenezeit in Slowenien seit Jahr 1964 / Raziskovanje latenske dobe na Slovenskem po letu 1964. – Arheološki vestnik 50, 189–213. BOŽIČ, D. 2005, Die spätrömischen Hortfunde von der Gora oberhalb von Polhov Gradec. – Arheološki vestnik 56, 293–368. BOŽIČ, D., S. CIGLENEČKI 1995, Zenonov tremis in pozno­antična utrdba Gradec pri Veliki Strmici / Der Tremissis des Kaisers Zeno und die spatantike Befestigung Gradec bei Velika Strmica. – Arheološki vestnik 46, 247–277. BURNHAM, B. C., J. WACHER 1990, The “small towns” of Roman Britain. – Berkeley, Los Angeles. CIGLENEČKI, S. 1985, Potek alternativne ceste Siscija – Ak­vileja na prostoru zahodne Dolenjske in Notranjske v času 4. do 6. stoletja. – Arheološki vestnik 36, 255–284. CIGLENEČKI, S. 2016, Claustra Alpium Iuliarum, tractus Italiae circa Alpes and the defence of Italy in the final part of the Late Roman period. – Arheološki vestnik 67, 409–424. COULON, J. 2015, Les fours dits de «potier“ de type Sévrier Haute-Savoie, France. Les indices d’une hypothese fonctio­nnelle alternative. – V / In: S. Boulud-Gazo, T. Nicolas (ur. / eds.), Artisanats et productions a l’âge du Bronze. (Séances de la Société préhistorique française 4), 109–122. DELAMARRE, X. 2007, Nomina Celtica antiqua selecta in-scriptionum / Noms de personnes celtiques dans l’épigraphie classique. – Paris. DOLENZ, H. 1997, Die Eisenfunde aus dem Stadt auf dem Magdalensberg. – Kärntner Museumsschriften 75. DULAR, J., B. KRIŽ 2004, Železnodobno naselje na Cvingerju pri Dolenjskih Toplicah. – Arheološki vestnik 55, 207–250. FAORO, D. 2015, Gentes e civitates adtributae. Fenomeni con-tributivi della romanita cisalpina. – V / In: L. Criscuolo, G. Geraci, A. Bencivenni (ur. / eds.), Simblos. Scritti di storia antica 6, 155–199. FOSS, P. W. 1994, Kitchens and dining rooms at Pompeii: the spatial and social relationship of cooking to eating in the Roman household 1, 2. – Disertacija / PhD thesis, Univer­sity of Michigan (https://quemdixerechaos.files.wordpress. com/2012/11/kitchensdiningpompeii_foss.pdf). GAGLIARDI, L. 2006, Mobilita e integrazione delle persone nei centri cittadini romani: aspetti giuridici. I. La classifica­zione degli incolae. – Pubblicazioni dell‘Istituto di Diritto Romano 40. GAITZSCH, W. 1980, Eiserne römische Werkzeuge. Studien zur römischen Werkzeugkunde in Italien und den nördlichen Provinzen des Imperium Romanum. – BAR, International Series 78. GASPARI, A. 1998–1999, Römische Schmiedewerkstätten auf dem Hügel Ulaka in Innerkrain, Slowenien. – Archaeologia Austriaca 82–83, 519–523. GASPARI, A. 2000, Ulaka. Prazgodovinska in rimska naselbi­na nad Starim trgom pri Ložu. – Magistrska naloga / MA thesis, Univerza v Ljubljani (neobjavljeno / unpublished). GASPARI, A. 2003, Latenske in zgodnjerimske najdbe iz Lju­bljanice. – Disertacija / PhD thesis, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). GASPARI et al. 2000 = A. GASPARI, M. GUŠTIN, I. LAZAR, B. ŽBONA TRKMAN 2000, Late Roman tool finds from Celje, Gradišče at Zbelovska gora and Sv. Pavel above Vrtovin (Slovenia). – V / In: M. Feugere, M. Guštin (ur. / eds.), Iron, Blacksmiths and Tools. Ancient European Crafts. Monographies Instrumentum 12, 187–203. GUŠTIN, M. 1991, Posočje in der jüngeren Eisenzeit. – Katalogi in monografije 27. HABIČ, P. 1977, Nekaj geografskih značilnosti Loške doline. – V / In: A. Avsec, L. Mlakar, J. Šumrada (ur. / eds.), No-tranjski listi 1. Posvečeno Loški dolini ob petstoletnici mesta Loža 1477–1977, 11–15, Stari trg pri Ložu. HORVAT, J. 1995, Notranjska na začetku rimske dobe: Parti pri Stari Sušici, Ambroževo gradišče in Baba pri Slavini (Notranjska [Inner Carniola] at the Beginning of the Ro­man Period: Parti near Stara Sušica, Ambroževo gradišče and Baba near Slavina). – Arheološki vestnik 46, 177–216. HORVAT, J., A. BAVDEK 2009, Okra. Vrata med Sredozemljem in Srednjo Evropo / Ocra. The gateway between the Mediter­ranean and Central Europe. – Opera Instituti Archaeologici Sloveniae 17. HORVAT, J., B. ŽBONA TRKMAN 2016, The 3rd century military equipment in south-western Slovenia. – V / In: J. Horvat (ur. / ed.), The Roman army between the Alps and the Adriatic, Opera Instituti Archaeologici Sloveniae 31 (Studia Alpium et Adriae I), Ljubljana, 99–120. INKRET, A. 2013, Rimskodobno železarstvo in obseg proizvo­dnje. – Diplomska naloga / BA thesis, Univerza v Ljubljani (neobjavljeno / unpublished). KASTELIC, J. 1951, Walter Schmid. – Zgodovinski časopis 5, 292–299. KEBE, J. 1996, Loška dolina in Babno polje. Zgodovina župnij Stari trg pri Ložu in Babno polje. – Ljubljana. KLEMENC, J. 1950, In memoriam Dr. Walter Šmid. 18. I. 1875–24. III. 1951. – Arheološki vestnik 1 (1951), 242–245. KOKALJ, Ž., R. HESSE 2017, Airborne laser scanning raster data visualization: A Guide to Good Practice. – Prostor, kraj, čas 14. KOS, P. 1986, The Monetary Circulation in the Southeastern Alpine Region ca. 300 B. C.-A. D. 1000. - Situla 24. KOS, P. 2012, The construction and abandonment of the Claustra Alpium Iuliarum defence system in light of the numismatic material / Gradnja in opustitev obrambnega sistema Claustra Alpium Iuliarum v luči numizmatičnega gradiva. – Arheološki vestnik 63, 265–300. KRAMAR et al. 2015 = KRAMAR, S., V. TRATNIK, M. HRO­VATIN, A. MLADENOVIĆ, H. PRISTACZ, N. ROGAN ŠMUC, Mineralogical and Chemical Characterization of Roman Slag from the Archaeological Site of Castra (Ajdo­vščina, Slovenia). – Archaeometry 57/4, 704–719. LAHARNAR, B. 2011, Roman lead slingshots (glandes plumbe­ae) in Slovenia / Rimski svinčeni izstrelki za pračo (glandes plumbeae) iz Slovenije. – Arheološki vestnik 62, 339–374. LAHARNAR, B. 2012, Notranjska med prazgodovino in an-tiko. – Disertacija / PhD thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). LAHARNAR, B. 2013, The Roman stronghold at Nadleški hrib, Notranjska region / Rimska utrdba na Nadleškem hribu na Notranjskem. – Arheološki vestnik 64, 123–147. LAHARNAR, B. 2015, The Roman army in the Notranjska region / Rimska vojska na Notranjskem. – V / In: J. Istenič, B. Laharnar, J. Horvat (ed. / eds.), Evidence of the Roman army in Slovenia / Sledovi rimske vojske na Slovenskem, Katalogi in monografije 41, 9–41. LAHARNAR, B., E. LOZIĆ 2015–2016, Roman battlefield ar­chaeology. Study case Grad near Šmihel pod Nanosom and Nadleški hrib (SW Slovenia). – Schild von Steier 27, 60–71. LAHARNAR, B., E. LOZIĆ 2017, Ulaka and Nadleški hrib (Slovenia). Sites of military conflicts from the last decades BC. – V / In: Fachgespräch “Schlachtfelder: Fundstellen und Denkmale, 23. August 2018, Mauerbach (Niederösterrei-ch), Fundberichte aus Österreich 56, D62–D67. LAHARNAR, B., E. LOZIĆ, A MIŠKEC 2020, Gradišče nad Knežakom. - V / In: J. Horvat, I. Lazar, A. Gaspari (ur. / eds.), Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 123-140. LONZA, B. 1981, La ceramica del castelliere degli Elleri. – Societa per la preistoria e protostoria della Regione Friuli--Venezia Giulia. Quaderno 4. MIŠKEC, A. 2003, The Early Romanization of the Southeastern Alpine Region in the Light of Numismatic Finds / Zgodnja romanizacija jugovzhodnoalpskega prostora v luči numiz­matičnih najdb. – Arheološki vestnik 54, 369–379. MIŠKEC, A. 2008, Denarni obtok v Istri v antičnem obdobju. Od grškega denarja do vzpostavitve rimskega denarnega sistema (Monetary circulation in Istria in ancient times. From Greek money to the introduction of the Roman monetary system). – Annales. Series Historia et Sociologia 18, 289–304. MODRIJAN, W. 1953, Walter Schmid zum Gedenken! – Schild von Steier 2, 1953, 5–8. MODRIJAN, Z. 2015, Amphorae from late antiquity hilltop settlements in Slovenia. – V / In: B. Hebert, N. Hofer (ur. / eds.), Fachgespräch “Spätantikes Fundmaterial aus dem Südostalpenraum”. 7. April 2014, Graz (Steiermark), Fun-dberichte aus Österreich. Tagungsbände 1, 21–32. MURGELJ, I. 2000, Kovinsko gradivo z Ajdovščine nad Rodi-kom. – Diplomska naloga / BA thessis, Univerza v Ljubljani (neobjavljeno / unpublished). MÜLLNER, A. 1878, Archaeologische Excurse durch Süd­steiermark und Krain. VIII. Altenmarkt bei Laas. – Mit­teilungen der k.k. Zentral-Kommissionn für Erhaltung und Erforschung der Kunst- und Historischen Dekmale 4, 83–81. MÜLLNER, A. 1879, Emona. Archäologische Studien aus Krain. – Laibach. MÜLLNER, A. 1909, Geschichte des Eisens in Krain, Görz und Istrien von der Urzeit bis zum Anfange des XIX. Jahrhun­derts. – Wien, Leipzig. PEČNIK, J. 1890, “Pečnikovo predavanje”. – Ljubljanski zvon 10, 381–383. PEČNIK, J. 1894, Pogled na kranjska gradišča. – Izvestja Mu-zejskega društva za Kranjsko 4, 6–12. PEČNIK, J. 1904, Prazgodovinska najdišča na Kranjskem. – Izvestja Muzejskega društva za Kranjsko 14, 185–196. OSMUK, N. 1977, Ajdovščina. – Varstvo spomenikov 21, 198–202. PERKO, V., A. BAVDEK, I. LAZAR 1998, Poznoantično najdi-šče Šmarata (The Late Roman Site of Šmarata). – Arheološki vestnik 49, 271–283. PETRU, P. 1969, Rimska utrjena vila v Starem trgu pri Ložu? – Arheološki vestnik 20, 115–124. PFLAUM, V. 2007, The supposed Late Roman hoard of tools and a steelyard from Vodice near Kalce / Domnevno po­znorimska zakladna najdba orodja in hitre tehtnice z Vodic pri Kalcah. – Arheološki vestnik 58, 285–332. PICK, K., W. SCHMID 1922–1924, Frühgeschichtliche Befe­stigungsanlagen im Bereiche der Isonzofront. – Jahreshefte des Österreichischen Archäologischen Institutes in Wien 21–22, Beibl., 277–308. PIETSCH, M. 1983, Die römischen Eisewerkzeuge von Saalburg, Feldberg und Zugmantel. – Saalburg Jahrbuch 39, 2–132. POLFER, M. 2000, Eisenproduktion und Eisenverarbeitung in Nordgallien und dem Rheinland während der römischen Kaiserzeit. – V / In: M. Feugere, M. Guštin (ur. / eds.), Iron, Blacksmiths and Tools. Ancient European Crafts, Mo-nographies Instrumentum 12, 67–87. PREDOVNIK, K. K. 2008, Kosov gomila v Razvanju in vpra­šanje obstoja mot na slovenskem ozemlju (Kos‘ Barrow in Razvanje and the question of existence of mottes in Slove­nia). – Annales. Series Historia et Sociologia 18, 369–384. PREMERSTEIN, A., S. RUTAR 1899, Römische Straßen und Besfestigungen in Krain. – Wien. ROTHENHÖFER, P. 2005, Strukturen des Handwerks im südlichen Niedergermanien. 1. Metallverarbeitendes Handwerk. – V / In: M. Polfer (ur. / ed.), Artisanat et éco­nomie romaine. Italie et provinces occidentales de l’Empire, Monographies Instrumentum 32, 65–73. SAGADIN, M. 1995, Mengeš v antiki (Mengeš in the Roman Period). – Arheološki vestnik 46, 217–245. SANTORO BIANCHI, S. 1992, I risultati dello scavo. – V / In: S. Santoro Bianchi (ur. / ed.), Castelraimondo. Scavi 1988–1990. 1. Lo scavo, Cataloghi e Monografie dei Musei Civici di Udine 2, 125–229. SARIA, B. 1937, Ulaka - Metulum? – Glasnik Muzejskega društva za Slovenijo 18, 59–60. SCHEFFER, C. 1981, Acquarossa II/1. Cooking and Cooking Stands in Italy 1400–400 B.C.– Stockholm. SCHEIN, T. 1987, Prazgodovinska najdišča v občini Cerkni-ca. – Diplomska naloga / BA thessis, Univerza v Ljubljani (neobjavljeno / unpublished). SCHMID, W. 1922, Beiträge zur Geschichte der frühmitte­lalterlichen Besiedelung der Steiermark. – Zeitschrift des Historischen Vereines für Steiermark 18, 27–45. SCHMID, W. 1923–1924, Römische Forschung in Österreich. Die südliche Ostalpenländer. – Bericht der Römisch-Ger­manischen Kommission 15, 178–241. SCHMID, W. 1937, Ulaka. Japodska naselbina nad Starim trgom pri Ložu. Prvo poročilo o raziskavanjih l. 1936. – Glasnik Muzejskega društva za Slovenijo 18, 17–32. SCHMID, W. 1939, Raziskovanje Ulake, japodske pokrajine nad Starim trgom pri Ložu. – Slovenec, št. 236, letnik 67, 14. 10. 1939, 5. SCHMID, W. 1944, Ulaka. Frührömische japodische Siedlung in Altenmarkt bei Laas (Krain). – Poročilo, tipkopis / Report, Typescript: Universalmuseum Joanneum (neobjavljeno / unpublished). SLAPŠAK, B. 1997, Starejša zgodovina Rodika. – V / In: M. Pregelj (ur. / ed.), Rodik med Brkini in Krasom. Zbornik ob 350. letnici cerkve, 19–64, Koper. SLAPŠAK, B. 1999, Slovenski kras v pozni prazgodovini in rimski dobi. – V / In: A. Kranjc (ur. / ed.), Kras. Pokrajina, življenje, ljudje, 145–163, Ljubljana. SLAPŠAK, B. 2003, O koncu prazgodovinskih skupnosti na Krasu / The end of prehistoric communities in the Karst region. – Arheološki vestnik 54, 243–257. STROBEL, K. 2016, “Römische Vici” – “Militarische Vici” – “Zivile Vici”: Kunstbergriffe der Forschung. – V / In: Römische Vici und der Verkehrsinfrastruktur in Raetien und Noricum, Schriftenreihe des Bayerischen Landesamtes für Denkmalpflege 15, 31–51. SUTTER, B. 1953, Bibliographie von Walter Schmid. – Schild von Steier 2, 143–195. SVOLJŠAK, D. 2014, Lončarjeva delavnica ob “obrtni po-ti” v železnodobni naselbini na Mostu na Soči (Pottery workshop on the ‘artisans’ street’ in the Iron Age settlement at Most na Soči). – V / In: S. Tecco Hvala (ur. / ed.), Studia Praehistorica in Honorem Janez Dular, Opera Instituti Archaeologici Sloveniae 30, 287–295. SVOLJŠAK, D., J. DULAR 2016, Železnodobno naselje Most na Soči. Gradbeni izvidi in najdbe / The Iron Age Settlement at Most na Soči. Settlement Structures and Small Finds. – Opera Instituti Archaeologici Sloveniae 33. ŠAŠEL, J. 1974, Über Umfang und Dauer der Militärzone Praetentura Italiae er Alpium zur Zeit Mark Aurels. – Mu­seum Helveticum 31, 225–233 (= Opera selecta, Situla 30, Ljubljana 1992, 388–396). ŠAŠEL, J. 1975a, Rimske ceste v Sloveniji. – V / In: Arheološka najdišča Slovenije, 74–104, Ljubljana.ŠAŠEL, J. 1975b, Ulaka. – V / In: Arheološka najdišča Slovenije, 155, Ljubljana. ŠAŠEL, J. 1981, Castelvm Larignvm (Vitr. 2,19,15). – Historia 30, 254–256 (= Opera selecta, Situla 30, Ljubljana 1992, 645–647). ŠAŠEL, J., P. PETRU (ur. / eds.) 1971, Claustra Alpium Iuliarum 1. Fontes. – Katalogi in monografije 5.ŠAŠEL KOS, M. 2000, Rimski napisi iz Šmarate (The Roman inscriptions from Šmarata). – Kronika 48, 95–101. ŠINKOVEC, A. 2020, Longaticum – Logatec. - V / In: J. Horvat, I. Lazar, A. Gaspari (ur. / eds.), Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 77-91. TRATNIK, V., T. ŽERJAL 2017, Ajdovščina (Castra) – pose-litev zunaj obzidja (Ajdovščina (Castra) – the extra muros settlement). – Arheološki vestnik 68, 245–294. TURK et al. 2009a = TURK, P., J. ISTENIČ, T. KNIFIC, T. NABERGOJ (ur.) 2009a, Ljubljanica – kulturna dediščina reke. – Ljubljana. TURK et al. 2009b = TURK, P., J. ISTENIČ, T. KNIFIC, T. NABERGOJ (eds.) 2009b, The Ljubljanica – a River and its Past. – Ljubljana. URLEB, M. 1968, Križna gora in okolica v antiki (Il monte Križna gora e dintorni nell’epoca romana). – Arheološki vestnik 19, 473–484. URLEB, M. 1974, Križna gora pri Ložu. – Katalogi in mono-grafije 11. URLEB, M. 1977, Loška dolina in okolica v davnini. – V / In: A. Avsec, L. Mlakar, J. Šumrada (ur. / eds.), Notranjski listi 1, Posvečeno Loški dolini ob petstoletnici mesta Loža 1477–1977, 16–30, Stari trg pri Ložu. URLEB, M. 1983, Antično grobišče v Cerknici (La nécropole romaine a Cerknica). – Arheološki vestnik 34, 298–346. VEITH, G. 1914, Die Feldzüge des C. Iulius Caesar Octavia-nus in Illyrien in den Jahren 35–33 v. Chr. – Schriften der Balkankommission, Antiquarischer Abteilung VII, Wien. VIDRIH PERKO, V. 1994, Poznoantične amfore v Sloveniji. – Disertacija / PhD thesis, Univerza v Ljubljani (neobjav­ljeno / unpublished). VINAZZA, M. 2016, Silosi – posebne keramične oblike. Prispevek k poznavanju gospodinjstev v starejši železni dobi na Krasu (Silos – special ceramic forms. Contribution to the knowledge of the Early Iron Age households in the Karst region). – Arheo 33, 7–23. VIŠNJIĆ, J. 2016, Nove spoznaje o obrambenom sustavu Claustra Alpium Iuliarum: Rezultati istraživanja provedenih u sklopu projekta “Claustra - kameni branici Rimskog Carstva”. – Portal. Godišnjak Hrvatskog restauratorskog zavoda 7, 13–34. ZACCARIA, C. 1992, Regio X. Venetia et Histria. Tergeste - Ager Tergestinus et Tergesti adtributus. – V / In: Supple-menta Italica n. s. 10, 139–283, Roma. ZACCARIA, C. 2007, Tra Natisone e Isonzo. Aspetti ammi­nistrativi in eta romana. – V / In: M. Chiaba, P. Maggi, C. Magrini (ur. / eds.), Le Valli del Natisone e dell’Isonzo tra Centroeuropa e Adriatico, Studi e ricerche sulla Gallia Cisalpina 20, 129–144. Andrej Gaspari Univerza v Ljubljani Filozofska fakulteta Oddelek za arheologijo Aškerčeva 2 SI-1000 Ljubljana andrej.gaspari@ff.uni-lj.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 173–186 IG Lucija GRAHEK, Anja RAGOLIČ Izvleček O rimski naselbini, ki je ležala na območju današnje vasi Ig, nimamo veliko podatkov. Okolica vasi je bila s prekinit­vami poseljena vse od mlajše kamene dobe. Posamične najdbe in obstoj cestne mreže kažeta, da se je rimska naselbina razprostirala od izvira Ižice, središča vasi in župnijske cerkve sv. Martina do pokopališča. V letu 2014 je bilo sistematično raziskano grobišče iz 1. in 2. st., ki se je razprostiralo na območju ledine Marof, domnevno vse do Banije na vznožju grajskega hriba Pungrt. Največ podatkov o antičnem Igu nam sporoča množica kamnitih spomenikov. Iz njih razberemo, da je v naselbini živela peregrina skupnost, najverjetneje organizirana v vicus ali v manjše zaselke, ki so upravno sodili k emonskemu agru. Onomastična analiza epigrafsko izpričanih imen kaže, da so tukajšnji prebivalci govorili avtohtoni, t. i. ižanski jezik. Upodobitve na kamnitih spomenikih prinašajo tudi druge informacije o vsakdanjem življenju rimskodobnih Ižancev. Ugodna lega v osrčju gozdov in nahajališč apnenca je omogočala razvoj kovaštva, lesarstva in kamnoseštva, plovni Ižica in Ljubljanica pa sta bili dobra povezava za prevoz dobrin do bližnje kolonije Emone. Ključne besede: Italija (10. regija), Ig, rimska doba, naselbina, vicus, grobišča, epigrafika, onomastika Abstract There is not much data available on the Roman settlement in the area of the present­day village of Ig. The neighbour­hood of the village was intermittently inhabited since the Neolithic period. Individual finds and the existence of a road network indicate that the Roman settlement extended from the spring of the Ižica creek, across the centre of the village and the church of St Martin to the graveyard. A cemetery from the 1st and 2nd centuries, which extended across the area called Marof, supposedly all the way to Banija at the foot of the castle hill Pungrt, was systematically investigated in 2014. The bulk of information on Roman Ig comes from a number of stone monuments. They suggest that the settlement was inhabited by a community of peregrini, probably organized as a vicus or in small hamlets, which administratively belonged to the ager of Emona. The onomastic analysis of the names recorded on the epigraphic monuments indicates that the inhabitants spoke an autochthonous language, the so­called Ig language. Images on stone monuments also reveal other information about the everyday life of the Roman period Ig people. The favourable location of the settlement, near forests and limestone deposits, led to the development of metalworking, forestry and quarrying, while the navigable rivers of Ižica and Ljubljanica functioned as a means of transport, by which goods were transported to the nearby colony of Emona. Keywords: Italy (Regio X), Ig, Roman period, settlement, vicus, cemeteries, epigraphy, onomastics https://doi.org/10.3986/9789610502586_09 Sl. 1: Karta arheoloških najdišč Iga. Fig. 1: Map of archaeological sites of Ig. (Vir / Source: GURS©2004, TTN 5, pomanjšano, M. = 1 : 20.000 / diminished, scale = 1:20,000) 1 Ig – Kolar; 2 Stara šola; 3 Sv. Martin; 4 Ig 82 (Možek); 5 Grad in Sv. Jurij; 6 Staje (“Stari dedec”); 7 Staje; 8 Zidana gorica; 9 Konjeniški klub Cavallo; 10 Zagorica; 11 Banija (Tovarna KIG); 12 Marof LEGA Naselje Ig leži na južnem robu Ljubljanskega barja, ob severovzhodnem vznožju Krimsko­Mokrškega hri­bovja. Staro vaško jedro je bilo osnovano ob izvirih Ižice, tik pod grajskim hribom (tudi Pungrt/Pungart; 366 m).1 Naselje se je na severni in vzhodni strani širilo globlje v Iški morost (Ljubljansko barje), na zahodni strani pa severno od ceste, ki je ob vznožju Pungrta speljana proti vasi Staje (sl. 1; 2). Po kraških izvirih Ižice je bila vas Ig sprva ime­novana Studenec (Prunne – 1323; Prunn – 1349, 1463; Prun – 1382, 1383, 1421, 1463, 1467, 1496; Prum – 1400; Prun – 1463; Studenicz – 1490). Ižanski grad na grajskem hribu je leta 1436 omenjen kot Sunek, ime pa je prešlo 1 Orožen Adamič 1990, 101. LOCATION The Ig settlement is located on the southern fringes of the Ljubljansko barje basin, at the northeastern foot of the hills of Krim and Mokrc. The old village core was situated at springs of Ižica river, just below the Pungrt hill (366 m).1 To the north and east, the settlement extended further into the Ig fen, while on the western side it was located north of the road at the foot of the Pungrt hill in the direction of the village of Staje (Figs. 1; 2). The village of Ig got its original name, Studenec (“Spring”), after karst springs (Prunne – 1323; Prunn – 1349, 1463; Prun – 1382, 1383, 1421, 1463, 1467, 1496; Prum – 1400; Prun – 1463; Studenicz – 1490). The castle of Ig is mentioned in 1436 as Sunek and the name was 1 Orožen Adamič 1990, 101. 174 Sl. 2: Ig z okolico. Kamnolomi in primarna ali sekundarna najdišča kamnitih spomenikov. Fig. 2: Ig and its surroundings. Quarries and primary or secondary sites of stone monuments. (Vir / Source: GURS©2004, TTN 5, pomanjšano na polovico / diminished to a half) tudi na vas – dorff Sunegk (1457, 1460). V virih od 13. stoletja naprej sta vas in okolica imenovana tudi Ig (Yge – 1249, 1251, 1300; castris … Ighe – 1261; Yg – 1262, 1300; Ig – 1262; Ige – 1271, 1274, 1291, 1296, 1309; Igg – 1293, Hyc oz. Hyco – 1299; Huco – 1300; Hic – 1328, …), red­keje Studenec (kot vikariat oz. fara). Od druge svetovne vojne naprej je v rabi ustaljeno ime Ig.2 Ime antične naselbine približno na tem mestu je neznanka, čeprav je njen zgodovinski pomen nedvomen in so ji zgodovinarji in raziskovalci zaradi številnih na Ižanskem najdenih rimskih spomenikov pripisovali zgo­dovinsko pomembnost.3 Tako kot ime ostajata vprašljiva tudi natančna lokacija in obseg antične naselbine. Bal­duin Saria je na podlagi množice rimskih spomenikov in naključno odkritih ostalin antičnega zidovja sklepal, da je treba to najverjetneje iskati pod takratno vasjo Studenec.4 Jaroslav Šašel je antično naselbino lociral 2 Kos 1975, 210–213 in 585–586; Orožen Adamič 1990, 101; Gestrin 1994, 2; Hostnik 1997, 9; Šašel Kos 1999a, 235; Preinfalk 2002. 3 Kot piše Theodor Mommsen v uvodu poglavja o ižanskih rimskih spomenikih v CIL, str. 484: … at vicus hic fuit nominis hodie ignoti, celeber tamen satis. 4 AIJ, str. 56: … Sie liegt wahrscheinlich unter dem transferred to the village – dorff Sunegk (1457, 1460). From the 13th century onwards, the name Ig is used in the sources for the village and its surroundings (Yge – 1249, 1251, 1300; castris … Ighe – 1261; Yg – 1262, 1300; Ig – 1262; Ige – 1271, 1274, 1291, 1296, 1309; Igg – 1293, Hyc or Hyco – 1299; Huco – 1300; Hic – 1328, … ), and more rarely Studenec (as vicarage/parish). Since World War II, the name Ig has been used consistently.2 The Roman name of the settlement is unknown, despite its undisputed historical significance; due to the large number of Roman monuments discovered in Ig and its surroundings, historians and researchers have emphasized its historical importance.3 Like its name, the exact location and extent of the Roman settlement remain questionable. Based on the large number of Ro­man monuments and accidental finds of Roman walls, Balduin Saria inferred that it was probably situated below 2 Kos 1975, 210–213 and 585–586; Orožen Adamič 1990, 101; Gestrin 1994, 2; Hostnik 1997, 9; Šašel Kos 1999a, 235; Preinfalk 2002. 3 Theodor Mommsen writes in the introduction to the chapter on the Ig Roman monuments in CIL, p. 484: … at vicus hic fuit nominis hodie ignoti, celeber tamen satis. v osrednji in zahodni del “starega Iga”, od izvira Ižice, središča vasi in župne cerkve sv. Martina do pokopališča (sl. 1). Prav tako pa ni zavračal možnosti, da lahko na Ižanskem pričakujemo več zaselkov, saj je zapisal, “da je bil v antiki zaselek ali vsaj kmetija v Mateni; morda tudi v Iški Loki, v Iški vasi, Strahomerju in na Tomišlju” (sl. 2).5 ZGODOVINA RAZISKAV IN VIRI V zgodovinskih virih se vas Ig (Yge) prvič omenja leta 1249, leta 1261 pa je prvič omenjen Ižanski grad (castris … Ighe), ki leži na griču Pungrt (sl. 1).6 Prvi zapisi z Iga so bili posvečeni zbiranju in risanju antičnih spomenikov. Edini pomemben zgodnji prepisovalec rimskih napisov iz Kranjske je bil humanist Avguštin Tyfernus (konec 15.–začetek 16. st.). Kot je razvidno iz uvoda v njegovem rokopisu s popisom ižanskih spomenikov (CVP 3528, fol. 69r), se je na Igu mudil spomladi leta 1507. Tu je prerisal in opisal spomenike, ki so v večini ležali okoli pokopališča župne cerkve.7 Na številnih najdbah antičnih nagrobnikov temeljijo tudi prve razprave o antičnem Igu, ki segajo v čas baročnega zgodovinopisja iz 17. stoletja. Ig tako zasledimo v delu Janeza Ludvika Schönlebna Carniolia antiqua et nova sive inclyti ducatus Carnioliae annales sacro-prophani (Tomus I, Labaci 1681). Tu je epigrafskim virom po­svečeno sedmo poglavje (De antiquis Inscritionibus Carnioliae), ki temelji na Lazijevem delu Commentario-rum reipublicae Romanae illius in exteris provinciis bello acquisitis constitutae libri duodecim (Basileae 1551), in ne na Tyfernu. Za Ižansko je pomemben zlasti četrti paragraf sedmega poglavja, kjer so obravnavani antič­ni napisi iz “Velike vasi” oz. današnjega Iga (Antiquae Inscriptiones Magni Vici, seu hodierni Iggii). Na podlagi napačnega branja gradbenega napisa8 z Vrhnike (Nau­portus) je Schönleben kratico MAG VICI na napisu, ki v bistvu označuje vaška načelnika (magistri vici), zmotno dopolnil kot Magnus vicus – (Veliko) vas – in menil, da se ime nanaša na sosednji Ig.9 heutigen Dorf Studenec, wo bei einem Bau gelegentlich antikes Mauerwerk angetroffen worden ist. 5 Šašel 1959, 118. Prim. Vuga 1979b, 315. 6 Preinfalk 2002. 7 XVIII Kal. May. hasce, quae sequuntur, inscriptiones exscripsi in Igg in coemeterio ecclesiae parrochialis a Laybach II miliario. Detajlno: RINMS, str. 23–28. 8 CIL I, 1467 = I2, 2286 = III, 3777 + str. 1729 = 10719; RINMS, str. 29–35: Q(uintus) Annaius Q(uinti) l(ibertus) / Torravius / M(arcus) Fulginas M(arci) l(ibertus) / Philogenes /5 mag(istri) vici de / vic(i) s(ententia) portic(um) f(aciundam) coir(averunt). Prevod: Kvint Anaj Toravij, Kvintov osvobojenec, (in) Mark Fulginas Filogen, Markov osvobojenec, vaška načelnika, sta po sklepu vaške skupščine oskrbela izgradnjo stebrišča (portika) (prevod po: Šašel Kos 2004, 79). 9 RINMS, str. 29–35. the village of Studenec.4 According to Jaroslav Šašel, the Roman settlement was located in the central and western parts of the “old Ig”, from the spring of the Ižica, across the centre of the village, the church of St Martin and to the graveyard (Fig. 1). He did not rule out the possibility of several settlements in Ig and its surroundings, and wrote that “in the Roman period, there was a settlement or at least a farm in Matena; perhaps also in Iška Loka, in Iška vas, in Strahomer and in Tomišelj” (Fig. 2).5 RESEARCH HISTORY AND SOURCES The village of Ig (Yge) is first mentioned in historical sources in 1249, while the Castle of Ig (castris … Ighe), which is situated on the hill of Pungrt, is first mentioned in 1261 (Fig. 1).6 The earliest notes on Ig were recordings and drawings of Roman monuments. The only significant early collector of Roman inscriptions from Carniola was the humanist Augustinus Tyfernus (end of the 15th – be­ginning of the 16th century). According to the introduc­tion to the manuscript with the inventory of the monu­ments from Ig (CVP 3528, fol. 69r), he visited Ig in the spring of 1507. He drew and described the monuments, which mostly lay scattered around the graveyard of the parish church.7 The many discovered Roman tombstones prompted for the first discussions on Roman Ig, which date to the period of the 17th century Baroque historiog­raphy. Ig can thus be found in Johann Ludwig Schönle­ben’s work Carniolia antiqua et nova sive inclyti ducatus Carnioliae annales sacro-prophani (Tomus I, Labaci 1681). Chapter 7 is dedicated to epigraphic sources (De antiquis Inscritionibus Carnioliae) and is not based on Tyfernus but on Lazius’s work Commentariorum reipublicae Romanae illius in exteris provinciis bello acquisitis constitutae libri duodecim (Basileae 1551). Particularly important for Ig and its surroundings is the fourth paragraph of the sev­enth chapter, which discusses Roman inscriptions from “Velika vas” or the present­day Ig (Antiquae Inscriptiones Magni Vici, seu hodierni Iggii). Schönleben committed an error in reading a construction inscription8 from Vrhnika 4 AIJ, p. 56: … Sie liegt wahrscheinlich unter dem heutigen Dorf Studenec, wo bei einem Bau gelegentlich antikes Mauerwerk angetroffen worden ist. 5 Šašel 1959, 118. Cf. Vuga 1979b, 315. 6 Preinfalk 2002. 7 XVIII Kal. May. hasce, quae sequuntur, inscriptiones exscripsi in Igg in coemeterio ecclesiae parrochialis a Laybach II miliario. RINMS, pp. 23–28. 8 CIL I, 1467 = I2, 2286 = III, 3777 + p. 1729 = 10719; RINMS, pp. 29–35: Q(uintus) Annaius Q(uinti) l(ibertus) / Torravius / M(arcus) Fulginas M(arci) l(ibertus) / Philogenes /5 mag(istri) vici de / vic(i) s(ententia) portic(um) f(aciundam) coir(averunt). Translation: Quintus Annaius Torravius, freedman of Quintus, (and) Marcus Fulginas Philogenes, freedman of Marcus, chiefs of the village, had the portico built in accordance with a decision made by the village (translation Z ižanskega območja je danes evidentiranih več kot 120 kamnitih spomenikov (sl. 2), ki so bili običajno uporabljeni kot gradbeni material ali pa so ležali po okolici in postopoma propadali. Povečini so bili odkriti kot stavbni ali dekorativni element, vzidan v cerkvene objekte (cerkev sv. Martina na Igu, cerkev sv. Mihaela v Iški vasi, porušena cerkev sv. Jurija na Pungrtu (na vzhodnem delu grajskega hriba), cerkev sv. Janeza Krstnika v Podkraju), v ižanski grad in nekatere hiše ter gospodarska poslopja na Igu ter v Stajah (sl. 1). Prvi, ki je v njih videl originalen spomenik, vreden postavitve za javnost, pa je bil Janez Gregor Dolničar (1655–1719). Ob gradnji ljubljanske stolnice in semenišča na začetku 18. stoletja je poskrbel, da je bilo štirinajst rimskih spo­menikov z Ižanskega in Emone vzidanih v novi stavbi.10 V drugi polovici 19. stoletja je Alfons Müllner na podlagi topografskih izsledkov in virov (Herodiana in Zosima), razdalje in lege na itinerarijih ter ne nazadnje zaradi velikega števila najdenih nagrobnikov povsem zgrešeno poskušal na Ig locirati antično Emono.11 V drugi polovici 20. stoletja so se s topografijoižanskega območja ukvarjali Jaroslav Šašel, Davorin Vuga in Andrej Pleterski.12 D. Vuga v svojih topograf­skih izsledkih omenja halštatsko gradišče na Pungrtu in ledino Ajdovšček pri Iški Loki, kjer bi naj po njegovem mnenju kljub časovno nezanesljivim najdbam stalo zgodnjesrednjeveško zatočišče.13 Arheološki podatki razmišljanja o lokaciji in obsegu antičnega Iga le delno potrjujejo (sl. 1), saj na tem območju še niso bila izvedena večja, sistematična raziskovanja. Izvedeni so bili topografski pregledi, ob različnih priložnostih so bile odkrite posamične najdbe, manjša zaščitna izkopavanja so bila opravljena na območju rimskega grobišča v Stajah (sl. 1: 7) in na območju “vodarne Brest” med vasema Brest in Staje (sl. 2).14 Manjši arheološki nadzor je potekal v letu 2012 na območju tovarne KIG, severno od Banije (sl. 1: 11),15 zadnja sistematična arheološka izkopavanja pa so bila izvedena v letu 2014 na območju izgradnje Raziskovalne postaje ZRC SAZU na Marofu (sl. 1: 12)16 pri Igu. Epigrafski spomeniki so vključeni v večje korpuse: Corpus inscriptionum Latinarum,17 Antike Inschriften aus Jugoslavien,18 v treh zvezkih pa sta na novo objav­ljene oz. prej neevidentirane rimske kamne objavila 10 Šašel Kos 1998. 11 Müllner 1879. 12 Šašel 1959; ANSl, str. 180–183 (Šašel in Jesse); Vuga 1977; 1979a; 1979b; 1980a; 1980d; 1981; 1982; 1986; Pleterski, Vuga 1987. 13 Vuga 1980a, 129–131. 14 Vuga 1980b; Vuga 1980c; Vičič 1987, 257; Nadbath 2006, 54; Nadbath, Draksler 2008, 76–77. 15 Tomažinčič, Češarek 2013. 16 Grahek 2014; Ragolič 2016; Grahek, Ragolič 2017. 17 CIL III, str. 484–488. 18 AIJ, str. 56–67. (Nauportus), wrongly interpreting the abbreviation MAG VICI, which actually refers to the heads of the village (magistri vici), as Magnus vicus (“Big Village”, Velika vas in Slovene) and thought it referred to the neighbouring Ig.9 Today, more than 120 stone monuments from Ig and its surroundings have been identified (Fig. 2), most of them reused as construction materials or scattered in the surroundings, gradually decaying. Most of them were discovered either as construction or decorative components of buildings, built in churches (the church of St Martin in Ig, the church of St Michael in Iška vas, the demolished church of St George on Pungrt, the church of St John the Baptist in Podkraj), in the castle of Ig, and in some houses and outbuildings in the villages of Ig and Staje (Fig. 1). The first one who considered them an origi­nal monument, worthy of being presented to the public, was Ioannes Gregorius Thalnitscher (1655–1719). During the construction of the Ljubljana cathedral and seminary at the beginning of the 18th century, he made sure that fourteen Roman monuments from Ig and Emona were built in the new building.10 In the second half of the 19th century, Alfons Müllner wrongly located Emona in the area of Ig. His argumentation was based on topographic findings and sources (Herodian and Zosimus), distances and posi­tions in itineraries, and also the numerous tombstones discovered in Ig.11 In the second half of the 20th century, Jaroslav Šašel, Davorin Vuga and Andrej Pleterski worked on the topog­raphy of Ig and its surroundings.12 In his topographic findings, Vuga mentions an Early Iron Age hillfort on Pungrt, and the so­called Ajdovšček area near Iška Loka, where he expects, despite the chronologically uncertain finds, an Early Medieval refuge.13 Archaeological data only partially confirm specu­lations on the location and extent of Roman Ig (Fig. 1), due to the fact that there have been no large systematic archaeological surveys in the area. In addition to topo­graphical overviews and isolated finds discovered at vari­ous occasions, small rescue excavations were carried out in the area of the Roman cemetery in Staje (Fig. 1: 7) and in the area of the “Brest Water plant” between the villages of Brest and Staje (Fig. 2).14 There was a minor archaeo­logical supervision in 2012 on the premises of the KIG company, north of Banija (Fig. 1: 11)15, while the most recent systematic archaeological excavation was carried after: RINMS 1). 9 RINMS, pp. 29–35. 10 Šašel Kos 1998. 11 Müllner 1879. 12 Šašel 1959; ANSl, pp. 180–183 (Šašel and Jesse); Vuga 1977; 1979a; 1979b; 1980a; 1980d; 1981; 1982; 1986; Pleterski, Vuga 1987. 13 Vuga 1980a, 129–131. 14 Vuga 1980b; Vuga 1980c; Vičič 1987, 257; Nadbath 2006, 54; Nadbath, Draksler 2008, 76–77. 15 Tomažinčič, Češarek 2013. Ana in Jaroslav Šašel.19 Spomenike, ki jih hrani Narodnimuzej Slovenije, je za objavo pripravila Marjeta Šašel Kos.20 Prav tako so rimski kamni dosegljivi na vseh ve­čjih spletnih epigrafskih bazah21 in v evropski digitalni knjižnici Europeana. Arheološki izsledki z ižanskega območja so do leta 1965 zbrani v Arheoloških najdiščih Slovenije (1975).22 Zbrani so tudi rezultati nekaterih raziskovanj, ki so potekala v okviru topografije ižanskega območja pod vodstvom D. Vuge.23 PRAZGODOVINA Najstarejše arheološke najdbe na območju da­našnje vasi in bližnje okolice sodijo v čas mlajše kamene dobe, okoliš pa je bil s prekinitvami poseljen skozi vso prazgodovino.24 Na severovzhodnem delu domnevnega območja antičnega Iga so bili odkriti ostanki srednje­ in poznobronastodobnega naselja, kar so potrdila kasnejša arheološka izkopavanja pri domačiji Kolar (sl. 1: 1).25 Gradišče iz železne dobe je bilo verjetno na grajskem hribu oz. Pungrtu (sl. 1), kar potrjujejo tudi občasne površinske najdbe.26 Domnevno je pod Pungrtom ležala manjša naselbina iz mlajše železne dobe, ki pa z najdbami (še) ni nedvomno potrjena.27 RIMSKA NASELBINA IN POTI Antična naselbina je verjetno stala pod današnjo vasjo Ig, od izvira Ižice, središča vasi in župne cerkve sv. Martina do pokopališča, ni pa bila nikoli raziskana. Ostanke rimskega zidu, na katere so naleteli ob gradnjah domačij, omenja že Saria. Poleg ostalin rimskih zidov pri stari šoli (sl. 1: 2) in župni cerkvi na Igu (sl. 1: 3)28 je znanih še nekaj lokacij, ki pa so bile v preteklosti preveč 19 ILJug 297 (Ig); ILJug 298–299 (Strahomer); ILJug 300 (Iška vas); ILJug 301 (Staje). 20 RINMS, str. 255–286; Šašel Kos 2004, 93. Za karto razprostranjenosti najdišč kamnitih spomenikov glej tudi Veranič, Repanšek 2016; za analizo avtohtonih ižanskih imen pa Repanšek 2016. 21 V frankfurtski – EDCS, rimski – EDR in v dunajski bazi Ubi erat lupa. 22 ANSl 1975, 180–183 (Šašel in Jesse). 23 Arheološka zaščitna raziskovanja na Ljubljanskem barju v letu 1979 (Vuga 1980a–c). 24 Vuga 1980d; Velušček 2004, 79; Velušček 2005; Čufar, Velušček, Kromer 2013; Draksler 2014. 25 Nadbath, Draksler 2008; Draksler 2014. 26 Šašel 1975a, 180; Vuga 1980a, 131; Vuga 1980d; Nad­bath, Draksler 2008, 76 s. 27 RINMS, str. 255. Na prehod iz latena v rimsko dobo je preliminarno datirana tudi keramika, ki je bila odkrita pri arheološkem nadzoru ob gradnji plinovodnega omrežja na Baniji (Tomažinčič, Češarek 2013). 28 Prim. Veranič, Repanšek 2016, 310, op. 20. out in 2014 in the construction site of a new research sta­tion of the Research centre of the Slovenian Academy of Sciences and Arts in Marof near Ig (Fig. 1: 12).16 Epigraphic monuments are included in major corpora: Corpus inscriptionum Latinarum,17 Antike In-schriften aus Jugoslavien.18 Newly published and not yet recorded Roman stones were included in a three­volume publication by Ana and Jaroslav Šašel.19 The monuments kept in the National Museum of Slovenia were publishedby Marjeta Šašel Kos.20 The Roman stones are also avail­able in all major internet epigraphic databases21 and in the Europeana, the European digital library. The archaeological findings from Ig and its sur­roundings up to 1965 are collected in the publication Arheološka najdišča Slovenije (1975).22 The results of surveys conducted during the topographical work in Ig and its surroundings under the direction of Vuga have also been collected and published.23 PREHISTORY The earliest archaeological finds from the present­day village and its immediate surroundings belong to the Neolithic period. The area was intermittently inhabited throughout prehistory.24 In the north­eastern part of the supposed area of Roman Ig, Middle and Late Bronze Age settlement remains were discovered and later confirmed by archaeological excavations at household Kolar (Fig. 1: 1).25 An Iron Age hillfort was probably on the hill of Pungrt (i.e. castle hill) (Fig. 1), as indicated by occasional surface finds.26 A small Late Iron Age settlement was sup­posedly situated below Pungrt, but so far it has not been sufficiently confirmed with finds.27 16 Grahek 2014; Ragolič 2016; Grahek, Ragolič 2017. 17 CIL III, pp. 484–488. 18 AIJ, pp. 56–67. 19 ILJug 297 (Ig); ILJug 298–299 (Strahomer); ILJug 300 (Iška vas); ILJug 301 (Staje). 20 RINMS, pp. 255–286; Šašel Kos 2004, 93. For the distribution of sites with Roman stone monuments in the Ig region see also Veranič, Repanšek 2016; for the study of the epihoric anthroponymy of Ig see Repanšek 2016. 21 In the databases of Frankfurt – EDCS, Rome – EDR and Vienna – Ubi erat lupa. 22 ANSl 1975, 180–183 (Šašel and Jesse). 23 Arheološka zaščitna raziskovanja na Ljubljanskem barju v letu 1979 (Vuga 1980a–c). 24 Vuga 1980d; Velušček 2004, 79; Velušček 2005; Čufar, Velušček, Kromer 2013; Draksler 2014. 25 Nadbath, Draksler 2008; Draksler 2014. 26 Šašel 1975a, 180; Vuga 1980a, 131; Vuga 1980d; Nad­bath, Draksler 2008, 76 s. 27 RINMS, p. 255. The pottery discovered during the archaeological supervision of the gas network construction in Banija was also preliminarily dated to the transition between the La Tene and Roman periods (Tomažinčič, Češarek 2013). poškodovane ali uničene, tako da ostaja vsakršna inter­pretacija negotova.29 Na območju domnevne rimskodobne naselbine (Ig 82, domačija Možek) je bila leta 1957 odkrita zakladna novčna najdba antoninijanov in enega denarija, datirana ok. leta 270 (sl. 1: 4).30 Na obseg naselbine kaže tudi obstoj cestne mreže, odkrite pri topografskih pregledih D. Vuge.31 Vzhodno od domnevnega jedra naselja je Vuga lociral arheološko sicer nepotrjeno križišče pomembnih lokalnih prome­tnih povezav: Lavrica–Ig–Podgozd v smeri sever–jug (sl. 1: pot a) in (Pijava Gorica)–Ig–Staje od vzhoda proti zahodu (sl. 1: pot b). Jedro naselja bi potemtakem ležalo zahodno od glavne prometne povezave z Emono, ki jo domnevno predstavlja tudi na lidarskem posnetku pre­poznana trasa ceste Lavrica–Ig, s potekom čez Ljubljan­sko barje zahodno od Kremenice.32 Drugo pomembno križišče je locirano na zahodni rob današnjega naselja. Jugozahodno od osamelca Zidana gorica sta se na pot Ig–Staje, ki je tekla ob vznožju Pungrta v smeri proti Strahomerju (sl. 1: pot b), domnevno priključili cesta Ig–Nauportus (sl. 1: pot c), ki je bila delno raziskana na območju Bresta33 in pot na Golo mimo “Starega Dedca” v Stajah (sl. 1: pot d), ki jo povezujemo z vicinalno cesto proti Blokam in Cerkniški dolini.34 GROBIŠČA Šašel in Vuga sta na podlagi edinega in situ ohra­njenega nagrobnika “Stari dedec”35 v Stajah (sl. 1: 6) domnevala, da se je rimskodobno grobišče razprostiralo zahodno od rimskega naselja, ob sodobni cesti Ig–Iška vas. Vendar je bil na tem območju (v Stajah) odkrit le skeletni grob, datiran v 4. st. (sl. 1: 7).36 Na bližino gro­bišča sta kazali tudi najdba pepelnice na vzpetini Zidana gorica in posamični zlatnik na vznožju (z dvorišča Ko­njeniškega kluba Cavallo) (sl. 1: 8,9); samo grobišče pa je bilo z izkopavanji leta 2014 odkrito na območju ledine Marof (sl. 1: 12) na vznožju iste vzpetine. Natančen obseg nekropole ni znan, po razkropljenosti najdb je obsegalo tudi območje blokovskega naselja Zagorica (sl. 1: 10), 29 AIJ, str. 56; Šašel 1975a, 180; Nadbath, Brenk 2006. 30 Kos 1991. 31 Vuga 1977 (Iška Loka); Vuga 1979b (Staje); Vuga 1980a, 130–133 (Iška Loka, Tomišelj); Vuga 1981 (Iška Loka); Vuga 1982, 208 (Ig). 32 Vuga 1979a, 278; Vuga 1980c, 57; Vuga 1986 (Babna gorica–Ig); Mlekuž 2014, 123. Prim. Pleterski, Vuga 1987, 139, sl. 1. 33 Vuga 1980c. 34 Šašel 1975a, 180; Vuga 1980c; Pleterski, Vuga 1987, 137, sl. 1. 35 CIL III 3804 = 10731 = AIJ 134; Šašel 1959, 118; Vuga 1979b, 314–315. 36 Vuga 1980b. ROMAN SETTLEMENT AND ROUTES The as yet uninvestigated Roman settlement was probably situated below the present­day village of Ig, in the area that extended from the spring of the Ižica creek, across the centre of the village and the church of St Martin to the graveyard. The remains of a Roman wall discovered during the construction of homes are mentioned already by Saria. Aside from the remains of Roman walls near the old school (Fig. 1: 2) and the Ig parish church (Fig. 1: 3),28 there are some other sites, but their interpretation is uncertain due to the damage suffered in the past.29 In 1957, a hoard of coins was found in the area of the supposed Roman settlement (Ig 82, household Možek). It consisted of antoniniani and one denarius, dated to about 270 (Fig. 1: 4).30 The extent of the settlement is indicated also by the existence of a road network, discovered during the topographical surveys by Vuga.31 East of the supposed village core, Vuga located an (archaeologically uncon­firmed) crossroads of major local traffic routes: Lavrica– Ig–Podgozd in a north­south direction (Fig. 1: road a), and (Pijava Gorica)–Ig–Staje in an east­west direction (Fig. 1: road b). This means the core of the settlement was probably west of the main route to Emona, which might be identified as the Lavrica–Ig route, which is visible in the Lidar image and runs across the Ljubljana Marshes west of the village of Kremenica.32 The second significant crossroads is located at the western edge of the present­day settlement. Southwest of Marof and under the isolated hill of Zidana gorica, the Ig–Staje route, which ran along the foot of Pungrt in the Strahomer direction (Fig. 1: road b), was supposedly joined by the Ig–Nauportus route (Fig. 1: road c), which was partly investigated southwest of Zidana Gorica and in the area of Brest,33 and by the route to Golo past “Stari dedec”, the tombstone in the village of Staje (Fig. 1: road d), which can be connected with the vicinal road to Bloke and Cerknica Valley.34 CEMETERIES Based on Stari dedec,35 the only in situ preserved tombstone in Staje (Fig. 1: 6), Šašel and Vuga supposed 28 Cf. Veranič, Repanšek 2016, 310, n. 20. 29 AIJ, p. 56; Šašel 1975a, 180; Nadbath, Brenk 2006. 30 Kos 1991. 31 Vuga 1977 (Iška Loka); Vuga 1979b (Staje); Vuga 1980a, 130–133 (Iška Loka, Tomišelj); Vuga 1981 (Iška Loka); Vuga 1982, 208 (Ig). 32 Vuga 1979a, 278; Vuga 1980c, 57; Vuga 1986 (Babna gorica–Ig); Mlekuž 2014, 123. Cf. Pleterski, Vuga 1987, 139, fig. 1. 33 Vuga 1980c. 34 Šašel 1975a, 180; Vuga 1980c; Pleterski, Vuga 1987, 137, fig. 1. 35 CIL III 3804 = 10731 = AIJ 134; Šašel 1959, 118; Vuga Sl. 3: Ig, Marof, severni del grobišča (izkopavanja 2014; sl. 1: 12). M. = 1 : 350. Fig. 3: Ig, northern part of the cemetery in Marof (the 2014 excavations; Fig. 1: 12). M. = 1:350. vse do Banije (sl. 1: 11) oz. do antične poti Ig–Staje, ob vznožju grajskega hriba (Pungrt).37 Na Marofu je bil raziskan del grobišča z žganimi grobovi, ki so datirani v 1. in 2. st. (sl. 3). V jugozahod­nem delu izkopnega polja je bila dokumentirana cesta (sl. 1: pot e), ki je tekla v smeri od jugovzhoda proti severozahodu. Odkritih je bilo več nivojev poti, ki kažejo na popravila, deloma sta bila ohranjena tudi obcestna jarka za odvajanje meteorne vode in celo sledi kolesnic. Zahodno od grobiščne ceste so ležale ostaline obzidane grobne parcele. V notranjosti parcele je bil dokumen­tiran postament, na katerem je morda stala nagrobna stela (sl. 4), odkrita nedaleč stran v poznoantični jami (sl. 3).38 Z geofizikalnimi raziskavami je bil samo 6 m južno 37 Vičič 1987; Grahek 2014; Grahek, Ragolič 2017. 38 Ragolič 2016. Sl. 4: Ig, Marof. Nagrobna stela Petona (foto: M. Lukić). Fig. 4: Ig, Marof. Funerary stele for Peto (photo: M. Lukić). that the Roman cemetery extended west of the Roman settlement, along the modern Ig–Iška vas road. However, only a single inhumation grave, dated to the 4th century, was discovered there (Fig. 1: 7).36 The vicinity of the cem­etery is further indicated by an ash chest found on Zidana gorica and an isolated gold coin from the courtyard of the Cavallo equestrian club (Fig. 1: 8,9). The cemetery itself was discovered during the 2014 excavation in Marof (Fig. 1: 12) at the foot of the same hill. Its exact extent is unknown. Judging by the distribution of finds, it included the residential area of Zagorica (Fig. 1: 10), all the way to Banija (Fig. 1: 11) and the Ig–Staje route at the foot of the Pungrt hill (castle hill).37 A part of a cemetery with cremation graves, which are dated to the 1st and 2nd centuries, was investigated in Marof (Fig. 3). In the southwestern part of the excavation area, a southeast­northwest running road was identified (Fig. 1: road e). Several levels of the roadway were identi­fied, an indication that repairs had been made to it. There were two partly preserved roadside ditches for storm wa­ter drainage and even wheel tracks. West of the cemetery road there were the remains of a grave plot, surrounded by a wall. In the plot, a pedestal was discovered, which might have supported the funerary stele (Fig. 4) discovered in the nearby Late Antique pit (Fig. 3).38 Just 6 m south of this grave plot, geophysical surveys revealed a structure of similar size (Fig. 3); while a part of what was probably another grave plot was discovered in the vicinity during an archaeological supervision at the end of 2018.39 The graves in Marof typically had simple pits, dug into the geological base and without prominent grave constructions. Cremation burials were mostly in urns (Fig. 5).40 The graves east of the Roman road form four groups (Fig. 3). SETTLEMENT STATUS AND INHABITANTS Epigraphic monuments indirectly testify to the fact that the area was controlled by a community of peregrini. During the Imperial time it was most likely formally organized as a vicus, although the existence of several 1979, 314–315. 36 Vuga 1980b. 37 Vičič 1987, 257; Grahek 2014; Grahek, Ragolič 2017. 38 Ragolič 2016. 39 Grahek 2014; Grahek, Ragolič 2017; Grahek 2019. 40 Grahek 2014; Grahek, Ragolič 2017. Sl. 5: Ig, Marof. Žarni grob 2 (M. = 1:20). 1 steklo; 2–5 keramika; 6 srebro; 7 bron; 8 kost; 9–12 železo. M. 1–5 = 1:3; 6–12 = 1:2. Fig. 5: Ig, Marof. Urn Grave 2 (scale = 1:20). 1 glass; 2–5 pottery; 6 silver; 7 bronze; 8 bone; 9–12 iron. Scale 1–5 = 1:3; 6–12 = 1:2. Sl. 6: Ig, spomenik Gaja Basidija Sekunda (po RINMS 79). Fig. 6: Ig, the monument of Caius Basidius Secundus (after: RINMS 79). od grobne parcele prepoznan objekt podobne velikosti (sl. 3); del verjetno še ene grobne parcele v bližini pa je bil odkrit pri arheološkem nadzoru konec leta 2018.39 Grobovi severovzhodno od antične ceste in v celoti raziskane grobne parcele so grupirani v štiri skupine (sl. 3). Za vse izkopane grobove so značilne preproste, v geološko osnovo vkopane grobne jame brez izrazitih grobnih konstrukcij. Žgani grobovi so bili v veliki večini žarni (sl. 5).40 STATUS NASELBINE IN PREBIVALSTVO Epigrafski spomeniki posredno izpričujejo, da je ta prostor obvladovala peregrina skupnost, ki je bila v času cesarstva najverjetneje formalno organizirana v vicus, čeprav je treba upoštevati tudi možnosti več manjših zaselkov na tem območju (pagi). Nedvomno pa je Ižansko sodilo k emonskemu agru.41 Arheološki viri zaenkrat potrjujejo domnevo o obstoju več manjših vasic (pagi) na območju Ižanskega, dve večji naselbinski jedri pa lahko domnevamo na Igu in v Iški vasi, na kar bi kazali grobovi okrog cerkve sv. Mihaela42 v Iški vasi in nekropola na Marofu kot tudi posamezni grobovi na Igu. Na podlagi epigrafskih spomenikov, ki so bili praviloma najdeni v sekundarni legi, po družbenem položaju na Ižanskem prevladuje srednji sloj lokalnega prebivalstva brez rimskega državljanstva, peregrinov, ki so izpričani na ok. 100 nagrobnikih. Izrazitega suženj­skega sloja ni; osvobojenci so omenjeni le izjemoma.43 Polnopravnih rimskih državljanov s tria/duo nomina je malo.44 Z Iga je znan Gaj Basidij Sekund (Caius Bassidius Secundus), ki je pripadal emonski mestni eliti in opravljal visoke mestne funkcije. Njegova kariera je zabeležena na nagrobniku (sl. 6),45 kar verjetno pomeni, da je nje­gova družina izvirala z Iga ali pa je tam imela posesti. V Emoni je dvakrat opravljal službo edila s sodno oblastjo 39 Grahek 2014; Grahek, Ragolič 2017; Grahek 2019. 40 Grahek 2014; Grahek, Ragolič 2017. 41 Šašel 1959, 117–123; RINMS, str. 255–256. 42 Šašel 1975b; Pleterski, Vuga 1987. 43 ILJug 297 = Šašel 1955; CIL III 10749; CIL III 10739. 44 Šašel 1959, 118 in 122. Prim. RINMS 79; CIL III 3804; CIL III 10740; CIL III 3853. 45 CIL III 10738 = AIJ 127 = RINMS 79. smaller settlements (pagi) is also possible. There is no doubt, however, that the Ig area belonged to the ager of Emona.41 Archaeological data so far confirm the idea of several small villages (pagi) in the surroundings of Ig, while two major settlement cores can be assumed in Ig and in Iška vas, as evidenced by the graves around the church of St Michael42 in Iška vas, and also the isolated graves and the necropolis in Ig. Based on epigraphic monuments, which were as a rule found in secondary position, the prevailing social class in Ig and its surroundings was the middle class of the local population without Roman citizenship – the peregrini, who appear on about 100 tombstones. There is no prominent class of slaves; freedmen are mentioned only exceptionally.43 There are few full Roman citizens 44 with tria/duo nomina. Caius Basidius Secundus, a member of the Emona elite, who held high offices in the town, is known from 41 Šašel 1959, 117–123; RINMS, pp. 255–256. 42 Šašel 1975b; Pleterski, Vuga 1987. 43 ILJug 297= Šašel 1955; CIL III 10749; CIL III 10739. 44 Šašel 1959, 118 and 122. Cf. RINMS 79; CIL III 3804; CIL III 10740; CIL III 3853. Sl. 7: Ig, odlomek nagrobnika z upodobitvijo izdelkov kovaške obrti (po Hostnik 1997, 70). Fig. 7: Ig, tombstone fragment with the depiction of blacksmith products (after: Hostnik 1997, 70). (aedilis iure dicundo bis), bil je upravitelj mestnih financ (quaestor pecuniae publicae), eden izmed dveh županov (II vir iure dicundo), pokrovitelj združenja gozdarjev (patronus collegii dendroforum) in nadzornik združenja gasilcev (praefectus collegii centonariorum), prav tako pa tudi župan s cenzorsko oblastjo (II vir iure dicundo quinqeunnalis). Kot v vsakem mestu sta tudi v Emoni delovali omenjeni združenji, manjši enoti pa gotovo tudi na Igu. Izpričan je tudi sevir, ki je vključen v municipalno kultno življenje v Emoni skrbel za cesarski kult in pri­pravljal slavja za različne praznike.46 Na nižjo lokalno upravo kaže tudi do zdaj štirikrat izpričana formula locus monumenti (dobesedno kraj spomenika oz. nagrobnika).47 Za lastništvo in delitev zemlje so morali skrbeti določeni organi, ki so bili urad­no zadolženi za grobiščni red in vodili evidenco, kdaj je zemlja prešla z državne/kolonialne uprave pod zakon povezan z verskimi zadevami (ius sacrum). Problematiko imenskega fonda, za katerega je mogoče trditi, da je na Ižanskem avtohton, je ob upošte­vanju stare literature na novo ovrednotil Luka Repanšek. Doslej je bil del imenskega zbira z Ižanskega zmotno opredeljen kot keltski (galski), nove analize pa kažejo, da je imena treba obravnavati kot posebno skupino znotraj severnojadranskega imenskega zbira.48 O ekonomiji in gospodarstvu antičnih Ižancev lahko sklepamo na osnovi reliefnih upodobitev na nagrobnikih. Ena izmed panog je bilo železarstvo oz. Ig. His career is recorded on a tombstone (Fig. 6),45 which probably means that his family either originated from Ig or had property there. He twice held the office of aedile with the right to dispense justice in Emona (aedilis iure dicundo bis), he was the treasurer of the public treasury (quaestor pecuniae publicae), joint mayor (II vir iure dicundo), patron of the association of foresters (patronus collegii dendroforum), prefect and patron of the associa­tion of firemen (praefectus collegii centonariorum), and joint mayor in charge of the census (II vir iure dicundo quinqeunnalis). Like in any town, the above­mentioned associations were active in Emona, while two smaller units must have existed also in Ig. There is evidence of a sevir who took part in the ritual life of Emona, was in charge of the Imperial cult and organized celebrations of various holidays.46 Low­level local administration is indicated by the formula locus monumenti (literally: the plot of the (funer­ary) monument), which has been found four times.47 Cer­tain institutions were definitely needed for documenting changes in land ownership and supervising the division of land. These officials kept the cemetery in order and were in charge of documenting when the land was transferred from the state/colonial administration under the religious law (ius sacrum). The issue of the names that might be considered autochthonous in Ig and its surroundings was – by taking into account earlier works – evaluated anew by Luka Repanšek. A part of the collection of names from 46 Alföldy 1958; Šašel Kos 1999b. 45 CIL III 10738 = AIJ 127 = RINMS 79. ILJug 297; RINMS 92; CIL III 3862 = AIJ 186; ILJug46 Alföldy 1958; Šašel Kos 1999b. 301. Glej še Šašel 1955, zlasti str. 380–381; Šašel 1959, 123.47 ILJug 297; RINMS 92; CIL III 3862 = AIJ 186; ILJug 301. 48 Stifter 2012; Repanšek 2016; Ragolič 2016. See also Šašel 1955, especially pp. 380–381; Šašel 1959, 123. kovaštvo (sl. 7).49 Še več rimskih spomenikov kaže na drugo pomembno dejavnost – na kamnoseštvo.50 V neposredni okolici Iga (v Stajah in v Skopačniku ter v nekoliko oddeljeni Podpeči in na Sv. Ani; sl. 2) so bogate plasti apnenca,51 z umetnostnozgodovinsko analizo šte­vilnih spomenikov pa je Edisa Lozić predvidela obstoj več kamnoseških delavnic.52 Gospodarsko pomembna je še plovna Ižica, ki je omogočala razvoj tovorništva, saj so prebivalci antičnega Iga s svojimi proizvodi oskrbovali predvsem Emono.53 49 CIL III 10739; CIL III 10743 = AIJ 137; Šašel 1959, sl. 5; Vuga 1979b, 314. Prim. Šašel 1959, 122–123.50 RINMS, str. 255; Djurić, Rižman 2017. 51 Ramovš 1990, 15–20; RINMS, str. 18–19 sl. 3. 52 Lozić 2009. 53 Šašel 1959, 122–123; RINMS, str. 255. Okrajšave / Abbreviations AIJ = V. Hoffiller, B. Saria, Antike Inschriften aus Jugoslavien 1. Noricum und Pannonia Superior, Zagreb 1938. ANSl = Arheološka najdišča Slovenije, Ljubljana 1975. CIL = Corpus inscriptionum Latinarum. EDCS = Epigraphik-Datenbank Clauss / Slaby (skrbnik: Man­fred Clauss) [http://db.edcs.eu/epigr/epi_de.php]. EDR = Epigraphic Database Roma (skrbnik: DigiLab Centro interdipartimentale di ricerca e servizi, Sapienza Universita di Roma) [http://www.edr­edr.it]. Europeana = Europeana collections [http://www.europeana. eu/portal/en/search?f[COUNTRY][]=slovenia&locale=e n&q=who%3A%28Anja+RAGOLIC%29] (zadnji dostop: 13. 6. 2019). ILJug = A. et J. Šašel, Inscriptiones Latinae quae in Jugoslavia inter annos MCMXL et MCMLX repertae et editae sunt (Situla 5), Ljubljana 1963; iidem, Inscriptiones Latinae quae in Jugoslavia inter annos MCMLX et MCMLXX repertae et editae sunt (Situla 19), 1978; iidem, Inscriptiones Latinae quae in Jugoslavia inter annos MCMII et MCMXL repertae et editae sunt (Situla 25), 1986. lupa = UBI ERAT LUPA, F. und O. Harl, http://lupa.at/ (Bild­datenbank zu antiken Steindenkmälern). RINMS = M. Šašel Kos, The Roman Inscriptions in the National Museum of Slovenia / Lapidarij Narodnega muzeja Slovenije (Situla 35), Ljubljana 1997. Ig and its surroundings had been wrongly identified as Celtic (Gaulish), while recent analyses have shown that they should be interpreted as a special group within the northern Adriatic languages and onomastic areas.48 The economy of the Roman residents of Ig can be inferred on the basis of relief images on tombstones. One of the sectors was metalworking or smithing (Fig. 7).49 Even more Roman monuments point to another significant activity: quarrying.50 There are rich limestone layers in the immediate surroundings of Ig (in Staje and Skopačnik, as well as in the slightly more distant village of Podpeč and Sv. Ana hill) (Fig. 2).51 With the help of an art­history ana­lyiss, Edisa Lozić, speculated about the existence of several stonemasonry workshops.52 Also the navigable creek of Ižica was important for the local economy, since it enabled the development of freighting: the residents of Roman Ig mostly supplied their products to Emona.53 Translation: Meta Osredkar 48 Stifter 2012; Repanšek 2016; Ragolič 2016. 49 CIL III 10739; CIL III 10743 = AIJ 137; Šašel 1959, fig. 5; Vuga 1979b, 314. Cf. Šašel 1959, 122–123.50 RINMS, p. 255; Djurić, Rižman 2017. 51 Ramovš 1990, 15–20; RINMS, pp. 18–19 fig. 3. 52 Lozić 2009. 53 Šašel Kos 1959, 122–123; RINMS, p. 255. ALFÖLDY, G. 1958, Augustalen­ und Sevirkörperschaften in Pannonien. – Acta Antiqua Academiae Scientiarum Hungaricae 6, 433–459. ČUFAR, K., A. VELUŠČEK, B. KROMER 2013, Two decades of dendrochronology in the pile dwellings of the Ljubljansko barje, Slovenia. – V / In: N. Bleicher, H. Schlichtherle, P. Gassmann, N. Martinelli (ur. / eds.), Dendro. - Chronologie - Typologie - Ökologie. Festschrift für André Billamboz zum 65. Geburtstag, 35–40, Freiburg im Breisgau. DJURIĆ, B., I. RIŽMAN 2017, Kamen Emone / The rocks of Emona. – V / In: B. Vičič, B. Županek, (ur. / eds.), Emona MM. Urbanizacija prostora – nastanek mesta / Emona MM. Urbanisation of space – beginning of a town, 121–144, Ljubljana. DRAKSLER, M. 2014, Ig. – V / In: B. Teržan, M. Črešnar (ur. / eds.), Absolutno datiranje bronaste in železne dobe na Slovenskem / Absolute dating of the Bronze and Iron Ages in Slovenia, Katalogi in monografije 40, 417–422. GESTRIN, F. 1994, Ižansko ozemlje v srednjem veku. – Kro­nika 42/3, 1–5. GRAHEK, L. 2014, Prvo strokovno poročilo o arheološkem izko­pavanju IG LC ZRC 2014: najdišče Marof na Igu. – Poročilo / Report, Inštitut za arheologijo, ZRC SAZU, Ljubljana (neobjavljeno / unpublished). GRAHEK, L. 2019, Poročilo o arheološkem nadzoru pri gradnji na najdišču Marof na Igu, parc. št. 1857/20, k.o. Ig; Ig – Rim-skodobna vaška naselbina, EŠD 11406. – Poročilo / Report, Inštitut za arheologijo, ZRC SAZU, Ljubljana (neobjavljeno / unpublished). GRAHEK, L., A. RAGOLIČ 2017, Prvo odkritje rimskega grobišča na Igu / The first discovery of the Roman cemetery at Ig, Slovenia. – V / In: B. Vičič, B. Županek, (ur. / eds.), Emona MM. Urbanizacija prostora – nastanek mesta / Emona MM. Urbanisation of space – beginning of a town, 235–248, Ljubljana. HOSTNIK, M. 1997, Cerkev sv. Mihaela v Iški vasi. – Ljubljana. KOS, M. 1975, Ig. – V / In: Gradivo za historično topografijo Slovenije (za Kranjsko do leta 1500) 1, A–M, 210–213, Ljubljana. KOS, P. 1991, Najdba antoninijanov tretjega stoletja /A Hoard of Third Century Antoniniani. – Situla 29. LOZIĆ, E. 2009, Roman stonemasonry workshops in the Ig area / Rimske klesarske delavnice na Ižanskem. – Arheološki vestnik 60, 207–221. MLEKUŽ, D. 2014, Lidarski posnetek območja med Škofljico in Igom. – V / In: A. Gaspari, Prazgodovinska in rimska Emona. Vodnik skozi arheološko preteklost predhodnice Ljubljane / Prehistoric and Roman Emona. A Guide through the Archaeological Past of Ljubljana's Predecessor, 123, Ljubljana. MÜLLNER, A. 1879, Emona. Archaeologische Studien aus Krain. – Laibach. NADBATH, B. 2006, Ig. – Varstvo spomenikov 42 (2005). Poročila, 54. NADBATH, B., M. BRENK 2006, Ig. – Varstvo spomenikov 39–41(2000–2004). Poročila, 60–61. NADBATH, B., M. DRAKSLER 2008, Ig – rimska vaška na­selbina. – Varstvo spomenikov 44 (2007). Poročila, 76–77. OROŽEN ADAMIČ, M. 1990, Ig. – V / In: Enciklopedija Slovenije 4, 101–102, Ljubljana. PLETERSKI, A., D. VUGA 1987, Rimski grobovi pri Sv. Mi­haelu v Iški vasi (Die römischen Gräber bei der Kirche des Sv. Mihael in Iška vas). – Arheološki vestnik 38, 137–160. PREINFALK, M. 2002, Ig in njegova okolica v srednjem veku. – V / In: M. Preinfalk (ur. / ed.), Krim odmev bo dal. Zbornik občine Ig, 17–30, Ig. RAGOLIČ, A. 2016, Nagrobna stela Petona z Iga / Funerary stele for Peto from Ig (Slovenia). – Arheološki vestnik 67, 277–296. RAMOVŠ, A. 1990, Gliničan od Emone do danes. – Ljubljana. REPANŠEK, L. 2016, Quiemonis and the epichoric anthro­ponymy of Ig (Quiemonis v luči avtohtonih ižanskih osebnih imen). – Arheološki vestnik 67, 321–357. SCHÖNLEBEN, J. L. 1681, Carniolia antiqua et nova sive inclyti ducatus Carnioliae annales sacro-prophani, Tomus I, Labaci 1681. STIFTER, D. 2012, On the linguistic situation of Roman­period Ig. – V / In: T. Meißner (ur. / ed.), Personal names in the Western Roman world. Proceedings of a workshop convened by Torsten Meißner, José Luis García Ramón and Paolo Poccetti, held at Pembroke College, Cambridge, 16–18 September 2011, 247–265, Berlin. ŠAŠEL, J. 1955, A new Roman stele from Ig near Ljubljana. – Živa antika 5, 373–382 (= Opera selecta, Ljubljana 1992, 234–240).ŠAŠEL, J. 1959, Prispevki za zgodovino rimskega Iga. – Kro­nika 7, 117–123. ŠAŠEL, J. 1975a, Ig. – V / In: Arheološka najdišča Slovenije, 180–181, Ljubljana.ŠAŠEL, J. 1975b, Iška vas. – V / In: Arheološka najdišča Slo­venije, 182, Ljubljana. ŠAŠEL KOS, M. 1998, Dolničarjev lapidarij / The Thalnitscher Lapidarium. – Arheološki vestnik 49, 329–353. ŠAŠEL KOS, M. 1999a, Romanizirana staroselska vas na Igu. – V / In: B. Avbelj (ur. / ed.), Zakladi tisočletij. Zgodovina Slovenije od neandertalcev do Slovanov, 235–237, Ljubljana. ŠAŠEL KOS, M. 1999b, Aspects of the Sevirate and Augusta-litas in the Northeastern Adriatic Area. – Histria Antiqua 5, 173–181. ŠAŠEL KOS, M. 2004, Lapidarij Narodnega muzeja Slovenije. Rimski spomeniki. Vodnik. – Ljubljana. TOMAŽINČIČ, Š., D. ČEŠAREK 2013, Poročilo o predhodnih arheoloških raziskavah ob gradnji plinovodnega omrežjana območju tovarne KIG na Igu, parc. Št. 1856/16, 1825/2, 1824/2, 1823/2, 1858/1 in 1819/3 k.o. Ig. – Poročilo / Report, ZVKDS OE Ljubljana (neobjavljeno / unpublished). VELUŠČEK, A. 2004, Past and present lake­dwelling studies in Slovenia. Ljubljansko barje (the Ljubljana Marsh). – V / In: F. Menotti (ur. / ed.), Living on the Lake in prehistoric Europe. 150 years of Lake-dwelling Research, 69–82, Lon­don, New York. VELUŠČEK, A. 2005, Iška Loka ­ bronastodobno naselje na obrobju Ljubljanskega barja (Iška Loka ­ a Bronze Age set­tlement on the edge of the Ljubljansko barje). – Arheološki vestnik 56, 73–89. VERANIČ, D., L. REPANŠEK, Rimski kamniti spomeniki iz cerkve sv. Janeza Krstnika v Podkraju pri Tomišlju / Roman stone monuments in the Church of St. John the Baptist in Podkraj near Tomišelj. – Arheološki vestnik 67, 297–320. VIČIČ, B. 1987, Ig. – Varstvo spomenikov 29, 257. VUGA, D. 1977, Iška Loka. – Varstvo spomenikov 21, 222. VUGA, D. 1979a, Babna Gorica. - Varstvo spomenikov 22, 278. VUGA, D. 1979b, Staje. - Varstvo spomenikov 22, 314–315. VUGA, D. 1980a, Topografski izsledki na Ljubljanskem barju v letu 1979. – V / In: Arheološka zaščitna raziskovanja na Ljubljanskem barju v letu 1979 I, 129–137, Ljubljana. VUGA, D. 1980b, Zaščitno izkopavanje na območju rimskega grobišča v Stajah. – V / In: Arheološka zaščitna raziskovanja na Ljubljanskem barju v letu 1979 I, 21–33, Ljubljana. VUGA, D. 1980c, Zaščitno izkopavanje na območju “Vodarne Brest« med vasema Brest in Staje na Ižanskem. – V / In: Arheološka zaščitna raziskovanja na Ljubljanskem barju v letu 1979 I, 51–58, Ljubljana. VUGA, D. 1980d, Železnodobna najdišča v kotlini Ljubljan­skega barja. – Situla 20/21 (1981), 199–210. VUGA, D. 1981, Iška Loka. – Varstvo spomenikov 23, 238–239. VUGA, D. 1982, Ig. – Varstvo spomenikov 24, 208. VUGA, D. 1986, Ig – Strojanova voda, Varstvo spomenikov 28, 266–267. Lucija Grahek Znanstvenoraziskovalni center SAZU Inštitut za arheologijo Novi trg 2 SI­1000 Ljubljana lucija.grahek@zrc­sazu.si Anja Ragolič Znanstvenoraziskovalni center SAZU Inštitut za arheologijo Novi trg 2 SI­1000 Ljubljana anja.ragolic@zrc­sazu.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 187–200 MENGEŠ Milan SAGADIN Izvleček Prazgodovinsko utrjeno naselje se je od začetka železne dobe razvijalo na Gobavici nad Mengšem in delno na njenem vzhodnem vznožju. Grobišča z žganimi in posameznimi skeletnimi grobovi so ležala na ravnini vzhodno od Gobavice. Gradivo pripada pretežno ljubljanski skupini, v manjši meri tudi dolenjski halštatski skupini. Latensko obdobje je slabo zastopano. Posamične najdbe z Gobavice lahko povežemo z manjšo rimsko vojaško postojanko v času od predavgustejskega do poznoavgustejskega obdobja. Rimska naselbina z zidanimi in lesenimi objekti se je od konca 1. st. pr. Kr. dalje razvila ob vznožju hriba. Najdeni so bili ostanki metalurške dejavnosti. Poselitveno območje se je raztezalo okoli 1,2 km daleč. Na njegovem obrobju so ležali najstarejši rimski grobovi, mlajši grobovi (2. in 3. st.) pa se pojavljajo tudi sredi nekdanje naselbine. Gradivo in način pokopa kažeta sorodnosti z Emono.V pozni antiki se središče naselbine prestavi proti severu. Širša okolica Mengša kaže izjemno intenzivno poselitev, posamezne naselbinske točke se pojavljajo na povprečni medsebojni razdalji 1,4 km. Ključne besede: Italija (10. regija), Mengeš, rimska doba, naselbina, polzemljanka, vodnjak, grobišče, vojska, me-talurgija Abstract Above the town of Mengeš, on the hill of Gobavica and partly at its eastern foot, a prehistoric fortified settlement developed from the beginning of the Iron Age on. Cemeteries with cremation burials and individual inhumation graves lay on the plain east of Gobavica. Most of the material belongs to the Ljubljana Group and a smaller part to the Dolenjska Group. The La Tene period is poorly represented. Isolated finds from Gobavica can be associated with a small Roman military settlement, dated from the Pre-Augustan to the Late Augustan period. From the end of the 1st century BC on, another Roman settlement with masonry and wooden buildings developed at the foot of the hill. Traces of metallurgical activities were recorded. The inhabited area extended to about 1.2 km. While the earliest Roman graves were on its edge, later graves (2nd and 3rd centuries) occur also in the middle of the former settlement. The inventory and the burial ritual show similarities with Emona. In the Late Antique period, the centre of the settlement moved to the north. There is evidence for intensive settlement in the wider surroundings of Mengeš, with the average distance between individual settlement points being 1.4 km. Keywords: Italy (Regio X), Mengeš, Roman period, settlement, pit house, well, cemetery, army, metallurgy https://doi.org/10.3986/9789610502586_10 LEGA Rimsko naselje v Mengšu leži na robu Mengeškega polja, ob rečici Pšati in ob vzhodnem vznožju vzpetine Gobavica. Ne moremo ga povezati z nobenim iz antičnih pisnih virov znanim naseljem. Predlagano lociranje postaje Ad Quartodecimo, ki jo omenja Itinerarium Burdigalense,1 v okolico Mengša2 ni prepričljivo, ker je Mengeš odmaknjen od logičnega poteka ceste Emona– Celeja.3 Identifikacija bi prišla v poštev le, če bi šlo za naselje ob odcepu lokalne ceste proti Gorenjski, kar pa ne gre v koncept omenjenega itinerarija. ZGODOVINA RAZISKAV Vse najstarejše arheološke najdbe iz Mengša so na­ključne.4 Prvo prazgodovinsko najdbo (nož z gumbastim zaključkom na ročaju) je v Deželni muzej v Ljubljani poslal graščak A. Stare leta 1897. A. Müllner leta 1898 omenja najdbe neobdelanega železa na Ogrinovem (sl. 1: 2)5 in odkritje žganega prazgodovinskega groba v Puščavi (sl. 1: 1; danes Pristava 7–8).6 Med najdbami, ki jih je v Narodni muzej v Ljubljani pred 1. svetovno vojno pošiljal Stare in ki po vsej verjetnosti izvirajo z njegovega posestva, je tudi prazgodovinska keramika. Leta 1939 sta bila v tedanji Staretovi drevesnici izkopana dva prazgodovinska žgana grobova (sl. 1: 6) – poleg dveh poznorimskih skeletnih grobov.7 Takoj po 2. svetovni vojni je bilo več prazgodovinskih grobov najdenih in uničenih ob kopanju razbremenilnika za Pšato (sl. 1: 7), severno od podjetja Semesadika.8 Najmanj šest prazgo­dovinskih skeletnih grobov je bilo najdenih ob gradnji blokov na Zadružniški ulici leta 1957 (sl. 1: 11), v bližini teh pa so naleteli na grobove tudi leta 1965 in 1973 pri gradnji družinskih hiš (sl. 1: med 11 in 10).9 V sklop tega grobišča sodijo tudi najdbe (uničenih) grobov na Grobeljski (tedaj Partizanski) cesti iz leta 1948 (sl. 1: 10). V kamnolomu pod Gobavico (sl. 1: 15) je leta 1970 Jože Ogrin našel bronasto iglo,10 leta 1993 pa je lastnik Oranžerije izročil Zavodu za varstvo kulturne dediščine Slovenije OE Kranj prazgodovinsko žaro, najdeno ob poglabljanju kleti (sl. 1: 5).11 1 It. Burd. 560, 8. 2 Müllner 1879, 86. 3 Šašel 1954, 15–16. 4 Gabrovec 1975. 5 Müllner 1898. 6 Gabrovec 1965, 97. 7 Gabrovec 1965, 97. 8 Gabrovec 1965, 98. 9 Gabrovec 1965, 98. 10 Zupančič 1979, 19, t. 1: 9. 11 Sagadin 1997. POSITION The Roman settlement in Mengeš is situated on the edge of the Mengeš Plain, along the creek of Pšata, at the eastern foot of the Gobavica hill. It cannot be identified as any of the settlements that are known from the Roman period written sources. The suggested localization of the Ad Quartodecimo station, which is mentioned in the Itinerarium Burdigalense,in the surroundings of Mengeš2 is not convincing since Mengeš does not lie on the logical course of the Emona–Celeia road.3 The identification would require the settlement to be situated where a local road branches off towards the Gorenjska region, which does not fit the concept of the above-mentioned itinerary. RESEARCH HISTORY The earliest finds from Mengeš are all accidental.4 The first prehistoric find (a knife with a button-shaped end of the hilt) was sent to the Museum in Ljubljana by Stare in 1897. In 1898, Alfons Müllner mentions unworked iron found in Ogrinovo (Fig. 1: 2)5 and the discovery of a prehistoric cremation grave in Puščava (Fig. 1: 1, today Pristava 7–8).6 The finds that were sent to the National Museum in Ljubljana before World War I by the manor owner Stare include prehistoric pottery and most likely originate from his estate. In 1939, two prehistoric cre­mation graves were excavated in addition to two Late Roman inhumation graves in the then Stare tree nursery (Slovenian: Staretova drevesnica) (Fig. 1: 6).7 Soon after World War II, several prehistoric graves were found and destroyed during the excavation of a retention canal for the creek of Pšata north of the Semesadike company (Fig. 1: 7).8 At least six prehistoric inhumation graves were discovered in 1957 during the construction of blocks of flats in Zadružniška ulica (Fig. 1: 11). More graves were found in the immediate vicinity in 1965 and 1973 during the construction of family houses (Fig. 1: between 11 and 10).9 The finds from the (destroyed) graves in Grobeljska cesta (then: Partizanska cesta), discovered in 1948, belong to the same cemetery (Fig. 1: 10). In 1970, Jože Ogrin found a bronze pin in the quarry below Gobavica,10 and in 1993 the Kranj Regional Office of the Institute for the Protection of Cultural Heritage of Slovenia (ZVKDS 1 It. Burd. 560, 8. 2 Müllner 1879, 86. 3 Šašel 1954, 15-16. 4 Gabrovec 1975. 5 Müllner 1898. 6 Gabrovec 1965, 97. 7 Gabrovec 1965, 97. 8 Gabrovec 1965, 98. 9 Gabrovec 1965, 98. 10 Zupančič 1979, 19, Pl. 1: 9. Sl. 1: Mengeš. Prazgodovinska najdišča. Fig. 1: Mengeš. Prehistoric sites. (Vir / Source: TTN 5, pomanjšano / diminished, M. = 1: 10.000 / Scale = 1:10,000; GURS©2004) 1 – Puščava (Pristava 7–8); 2 – Ogrinovo; 3 – Gobavica (sondiranje / trenching 2005); 4 – Osnovna šola; 5 – Oranžerija; 6 – Sta-retova drevesnica; 7 – Razbremenilni kanal Pšate; 8 – Semesadika; 9 – Zavrti; 10 – Grobeljska cesta (nekdanja / former Partizan-ska); 11 – Zadružniška ulica; 12 – Gobavica (sondiranje / trenching 1998); 13 – Rašiška cesta; 14 – Osnovna šola; 2008–2010, 2016); 15 – Kamnolom. Najstarejša omemba antičnih najdb je že iz leta 1834 z območja Staretove graščine (žgan grob; sl. 2: 1), leta 1847 se omenjajo številni ostanki rimskih stavb na Drnovem severno od Mengša (sl. 2: 2), kontinuirano pa so v Narodni muzej v Ljubljani prihajale zlasti rimske grobne najdbe s Staretovega posestva (sl. 2: 1). Leta 1947 so blizu Staretovega drevoreda našli tudi več antičnih zidov, ki jih lahko povezujemo z ostanki stavbnega kom­pleksa na območju osnovne šole (sl. 2: 8);12 ob Kamniški cesti je bil najden žgan grob iz druge polovice 2. st. po Kr. (sl. 2: 6),13 v Slomškovi ulici pa sta bila strokovno odkopana dva rimska žgana grobova iz prve polovice 2. st. po Kr. (sl. 2: 7).14 Vrsta naključnih najdb je bila odkrita ob strojnem izkopu temeljev za osnovno šolo leta 1977 (sl. 2: 8).15 Prvo načrtno zaščitno arheološko izkopavanje v Mengšu je bilo opravljeno leta 1978 na območju podjetja Semesadika (ZVKDS OE Kranj), kjer so bili poleg že omenjenih prazgodovinskih grobov najdeni tudi pozno­antični naselbinski ostanki (sl. 2: 5; 4; 8; 10). Sledila so zaščitna izkopavanja ZVKDS OE Kranj poznoantičnih naselbinskih objektov na gradbišču Vzgojnovarstvenega zavoda Gobica leta 1985-1986 (sl. 2: 9), antične naselbi­ne na gradbišču športne dvorane Fit - top leta 1988 (sl. 2: 10; 5). Leta 1993 in 1997 sledijo še izkopavanja ostankov antične naselbine ob osnovni šoli (sl. 2: 8) ter grobišča na poti med osnovno šolo in športno dvorano (sl. 2: 11; 7).16 Manjše sondažno izkopavanje so leta 1998 na Gobavici izvedli sodelavci Muzeja Mengeš, Inštituta za arheologijo ZRC SAZU in Narodnega muzeja Slovenije (sl. 1: 12; 2: 20; 3).17 Leta 2000 so bila opravljena zaščitna izkopavanja prazgodovinskih (sl. 1: 9) in poznoantičnih grobov na Zavrteh (sl. 2: 12,13) (med hišami št. 5 in 7),18 leta 2001 pa antičnih grobov na Maistrovi in Zadružniški ulici (sl. 2: 14).19 Leta 2005 je bilo na Gobavici opravljeno sondiranje prazgodovinskega obrambnega obzidja (sl. 1: 3),20 dve leti kasneje (2007) pa je bil opravljen zaščitni izkop ostankov rimske stavbe na Jelovškovi ulici št. 2 (sl. 2: 16).21 Leta 2008 je bilo zaradi načrtovane gradnje športne dvorane pri osnovni šoli (sl. 2: 4,8) izvedeno arheološko jedrno vrtanje, ki je bilo dopolnjeno leta 2009 in 2010 in ki je bistveno dopolnilo podatke o obsegu antične in prazgodovinske poselitve.22 Zaradi ponovne širitve osnovne šole je bilo ob vzhodni strani šolskega 12 Šašel 1954, 14; Stražar 1993, 49. 13 Sagadin 1999, sl. 64. 14 Zupančič 1979, 20. 15 Sagadin 1999, 39. 16 Sagadin 1995, 217–245; Sagadin 1997. 17 Turk 1999, 32; Horvat 2015, 185–189. 18 Železnikar 2000; Sagadin 2006, 101. 19 Štibernik 2001; 2006b. 20 Štibernik 2006a. 21 Sagadin, Peče 2007. 22 Novšak, Šinkovec, Verbič 2008; Novšak, Erjavec, Ver­ bič 2009; Novšak, Verbič, Erjavec 2010. OE Kranj) was given a prehistoric urn, discovered while deepening the cellar, by the owner of the Oranžerija teahouse (Fig. 1: 5).11 Roman period finds are first mentioned in 1834 in the area of the Stare manor (a cremation grave; approximately Fig. 2: 1) and then in 1847, when numerous remains of Ro­man buildings in Drnovo north of Mengeš are mentioned (Fig. 2: 2). In addition to that, there was a continuous influx of predominantly Roman funerary finds from the Stare estate to the National Museum in Ljubljana (Fig. 2: 1). In 1947, several Roman walls, which can be associated with the remains of a building complex in the area of the school, were discovered in the vicinity of the Stare Avenue (Fig. 2: 8),12 a cremation grave from the second half of the 2nd century AD was discovered near Kamniška cesta (Fig. 2: 6),13 and two Roman cremation graves from the first half of the 2nd century AD were archaeologically excavated in Slomškova ulica (Fig. 2: 7).14 A series of stray finds was discovered in 1977 during the excavation of the foundations for the school (Fig. 2: 8).15 The first planned archaeological rescue excavation in Mengeš was carried out in 1978 by the ZVKDS OE Kranj in the area of the Semesadike company, where Late Antique settlement finds were discovered in ad­dition to prehistoric graves (Figs. 2: 5; 4; 8; 10). The ZVKDS OE Kranj conducted rescue excavations of Late Antique residential buildings in the construction site of the Gobica kindergarten (Vzgojnovarstveni zavod Go-bica) in 1985/86 (Fig 2: 9), of a Roman settlement in the construction site of the Fit – top sports hall in 1988 (Figs. 2: 10; 5), and near the school (Fig. 2: 8) and in the road between the school and the sports hall in 1993 and 1997 (Fig. 2: 11; 7).16 A smaller test excavation was carried out in 1998 on Gobavica with the collaboration of the Mengeš Museum, the Institute of Archaeology ZRC SAZU and the National Museum of Slovenia (Figs. 1: 12; 2: 20; 3).17 In 2000, a rescue excavation of prehistoric (Fig. 1: 9) and Late Antique graves was carried out in Zavrti (between the houses no. 5 and no. 7; Fig. 2: 12,13),18 and in 2001 in Maistrova ulica and in Zadružniška ulica (Fig. 2: 14).19 In 2005, test trenching of the prehistoric defence wall was conducted on Gobavica (Fig. 1: 3)20 and two years later (2007) a rescue excavation of the remains of a Roman building in Jelovškova ulica no. 2 (Fig. 2: 16) was carried out.21 In 2008, archaeological core drilling was conducted due to the planned construction of a sports hall in the 11 Sagadin 1997. 12 Šašel 1954, 14; Stražar 1993, 49. 13 Sagadin 1999, sl. 64. 14 Zupančič 1979, 20. 15 Sagadin 1999, 39. 16 Sagadin 1995, 217–245; Sagadin 1997. 17 Turk 1999, 32; Horvat 2015, 185–189. 18 Železnikar 2000; Sagadin 2006, 101. 19 Štibernik 2001; 2006b. 20 Štibernik 2006a. 21 Sagadin, Peče 2007. Sl. 2: Mengeš. Rimskodobna najdišča. Fig. 2: Mengeš. Roman period sites. (Vir / Source: TTN 5, pomanjšano / diminished, M. = 1: 10.000 / Scale = 1:10,000; GURS©2004) 1 – Staretovo posestvo; 2 – Drnovo; 3 – Staretova drevesnica; 4 – Osnovna šola (2008–2010, 2016); 5 – Semesadike; 6 – Kamniška cesta; 7 – Slomškova ulica; 8 – Osnovna šola; 9 – Vzgojnovarstveni zavod Gobica; 10 – Športna dvorana Fit - top (1988); 11 – Ob-močje med osnovno šolo in športno dvorano; 12–13 – Zavrti; 14 – Zadružniška ulica; 15 – Kersnikova ulica; 16 – Jelovškova ulica; 17 – Pristava; 18 – Muljava; 19 – Staretovo posestvo (Ravbarjev grad) / Stare estate; 20 – Gobavica (1998). Sl. 3: Mengeš. Gobavica. Posamične najdbe. 1,2,4–7 bakrova litina; 3 železo in kalcedon; 8,9 železo; 10 keramika. M. = 1:2. Fig. 3: Mengeš. Gobavica. Stray finds. 1–2,4–7 copper alloy, 3 iron and chalcedony, 8–9 iron, 10 pottery. Scale = 1:2. (po / after Horvat 2015, sl. / fig. 7) poslopja leta 2016 izvedeno arheološko izkopavanje, ki je potrdilo antično in prazgodovinsko naselbino (sl. 2: 4).23 PRAZGODOVINA Starejše naključne najdbe, posebej pa novejše raz­iskave potrjujejo obstoj obsežnega prazgodovinskega gradišča na Gobavici. Sondiranje obrambnega obzidja je pokazalo vsaj tri faze njegove obnove, v notranjosti naselbine pa ostanke zelo solidno grajenih bivalnih ob-jektov (sl. 1: 3).24 Naselbinske najdbe segajo od začetka starejše železne dobe (konec 8. ali zač. 7. st. pr. Kr.) do latenskega obdobja. Naključna najdba bronaste igle dopušča domnevo o starejšem začetku naselbine na Gobavici (sl. 1: 15).25 Kot kažejo naključne keramične najdbe, pa je prazgodovinska poselitev segla tudi na rav­nino ob vznožju (okolica osnovne šole – sl. 1: 4,5; Rašiška cesta – sl. 1: 13). Sočasni grobovi so bili odkriti na štirih območjih: v Puščavi (jugozahodno pobočje Gobavice, sl. 1: 1), v Staretovi drevesnici, v okolici Oranžerije in med Grobeljsko cesto in Zadružniško ulico (sl. 1: 5,6,7,9,10,11). Grobne najdbe z območja Oranžerije so celo nekoliko starejše kot naselbinske najdbe z Gobavice in kažejo sorodnosti z ljubljansko skupino 9. in 8. st. 23 Urankar, Bešter 2016. 24 Štibernik 2006a. vicinity of the school (Fig. 2: 4,8). Drilling was repeated in 2009 and 2010 and it significantly supplemented the data on the extent of the prehistoric and Roman occupation.22 Due to another enlargement of the school building, an archaeological excavation was conducted in 2016 on the eastern side of the school and it confirmed the Roman and Prehistoric settlements (Fig. 2: 4).23 PREHISTORY Earlier stray finds and even more so the material from more recent surveys confirm the existence of a large prehistoric hillfort on Gobavica. Test trenching in the area of the defence wall revealed at least three phases of its reconstruction, while remains of well-built residential houses were discovered in the interior of the settlement (Fig. 1: 3).24 Settlement finds cover the period from the beginning of the Early Iron Age (end of the 8th /beginning of the 7th century BC) to the La Tene period. An acciden­tally discovered bronze pin allows for the assumption of even earier beginning of the settlement on Gobavica (Fig. 1: 15).25 As indicated by stray pottery finds, the prehistoric 22 Novšak, Šinkovec, Verbič 2008; Novšak, Erjavec, Ver­bič 2009; Novšak, Verbič, Erjavec 2010. 23 Urankar, Bešter 2016. 24 Štibernik 2006a. 25 Turk 1999, 31. 25 Turk 1999, 31. Sl. 4: Mengeš. Semesadika (sl. 2: 5). Fibuli, bakrova litina. M. = 1:2. Fig. 4: Mengeš. Semesadika (Fig. 2: 5). Fibulae, copper alloy. Scale = 1:2. pr. Kr. Žarni grobovi, najdeni v Staretovi drevesnici (sl. 1: 6), dodatno izpričujejo značilnosti ljubljanske skupine. Inventar jih umešča v drugo polovico 8. in začetek 7. st. pr. Kr. Vendar istočasni skeletni grobovi z Zadružniške ulice kažejo vpliv dolenjske skupine.26 Mlajši žgani grobovi 6. st. pr. Kr. iz okolice podjetja Semesadike (sl. 1: 8) pa izpričujejo prevladujoč vpliv posoške skupine.27 Latensko obdobje je slabo zastopano, znanih je le nekaj naključno najdenih predmetov (nož z gumbastim za­ključkom ročaja, nož z zanko na ročaju).28 RIMSKA NASELBINA Ob začetku rimske dobe (od predavgustejskega do poznoavgustejskega obdobja) je vsaj občasno bivala manjša rimska vojaška enota na območju prazgodovin­ske naselbine na Gobavici (sl. 2: 20; 3).29 Ostanki rimske naselbine ležijo ob vzhodnem in severovzhodnem vznožju Gobavice. Velik del zidanih objektov na vznožju je bil uničen ob gradnji osnovne šole (sl. 2: 8). Strnjeno pozidano območje se je širilo do današnje športne dvorane (sl. 2: 10; 5). Arheološke raz­iskave v zadnjih letih so potrdile obstoj poznoantičnih bivalnih objektov (polzemljank) na območjih podjetja Semesadika (sl. 2: 5; 10) in Vzgojnovarstvenega zavoda Gobica (sl. 2: 9).30 Stavbe pri osnovni šoli (sl. 2: 8) so imele 70 cm debele zidove, zidane z lomljenci, kar bi lahko omogo-čalo nadstropno gradnjo. Odkrit je bil tudi 7,5 m globok vodnjak, na suho zložen iz lomljencev, v spodnjem delu 26 Gabrovec 1965, 109. 27 Turk 1999, 36. 28 Železnikar 1999a, 37. 29 Horvat 2015, 185–189. 30 Sagadin 1995, 217–245. settlement extended all the way to the foot of the hill (the surroundings of the school Fig. 1: 4,5; Rašiška cesta Fig. 1: 13). Contemporary graves were discovered in four loca­tions: Puščava (southwestern slope of Gobavica, Fig. 1: 1), the Stare tree nursery, the surroundings of Oranžerija, and between the streets of Zadružniška ulica and Grobeljska cesta (Fig. 1: 5,6,7,9,10,11). Burial finds from the area of Oranžerija, which are even slightly earlier than the set­tlement finds from Gobavica, show similarities with the Ljubljana Group in the 9th and 8th centuries BC. Crema­tion graves from the Stare tree nursery (Fig. 1: 6) also show similarities with the Ljubljana Group. On the basis of their inventory they can be dated to the second half of the 8th and the beginning of the 7th century BC. The contempo­rary inhumation graves from Zadružniška ulica, however, show the influence of the Dolenjska Group.26 Later (6th century BC) cremation graves from the surroundings of the Semesadike company (Fig. 1: 8) show a dominant influence of the Posočje (Sveta Lucija) Group.27 The La Tene period is poorly represented and only some stray finds are known (knife with a button-shaped end of the hilt, knife with a loop on the hilt).28 ROMAN SETTLEMENT At the beginning of the Roman period (Pre-Augus­tan to Late Augustan), a small Roman military unit was stationed, at least occasionally, in the area of the prehis­toric settlement on Gobavica (Figs. 2: 20; 3).29 The remains of the Roman settlement lie below the eastern and northeastern slopes of Gobavica. A large pro­portion of the masonry buildings was destroyed during 26 Gabrovec 1965, 109. 27 Turk 1999, 36. 28 Železnikar 1999a, 37. 29 Horvat 2015, 185–189. (okoli 2 m) pa vsekan v živo skalo (sl. 6). Najdbe v njem segajo od 1. do 4. st. po Kr. Obrtniški del naselja je bil ugotovljen pri gradnji športne dvorane (sl. 2: 10). Tu so ležali ostanki lesene stavbe z glinenim ometom iz 1. in začetka 2. st. Ognjišče v stavbi, ki je bilo grajeno iz podlage oblih kamnov in prevlečeno z ilovico, je bilo vsaj dvakrat obnovljeno. Naj­dene so bile večje količine železove žlindre, kosi pečne obloge, vodni kanal in keramične šobe za vpihovanje zraka v talilne peči.31 Naključne antične najdbe naselbinskega značaja izvirajo tudi s severovzhodnega in vzhodnega vznožja Gobavice (Staretovo posestvo; sl. 2: 19).32 Skrajno južno točko poselitvenega areala predstavljajo ostanki večjega zidanega objekta iz Jelovškove ulice št. 2 (sl. 2: 16). 31 Sagadin 1995, 224–230. 32 Šašel 1954, 14. Sl. 6: Mengeš. Vodnjak pri osnovni šoli (sl. 2: 8). Fig. 6: Mengeš. Well in the vicinity of the school (Fig. 2: 8). the construction of the school (Fig. 2: 8). The agglomera­tion of buildings extended up to the modern sports hall (Fig. 2: 10; 5). Archaeological research in the recent years has confirmed the existence of Late Antique dwellings (pit-houses) around the present-day Semesadike com­pany (Figs. 2: 5; 10) and Gobica kindergarten (Fig. 2: 9).30 The buildings in the area of the school (Fig. 2: 8) had 70 cm thick walls made of quarry stones, which could support a multi-storey construction. A 7.5 m deep well was built of quarry stones in the drywall technique, with the lower part (about 2 m) cut into the bedrock (Fig. 6). The finds from the well are dated from the 1st to the 4th century AD. The area of craftsmanship activities was identified during the construction of the sports hall (Fig. 2: 10). It included the remains of a wooden building with clay plas­ter, dated to the 1st and the beginning of the 2nd century. The fireplace in the building was rebuilt at least twice; it had a base of round stones and was coated with loam. The artefacts include large quantities of iron slag, fragments of wall coating from smelting furnaces, a water canal, and ceramic nozzles for blowing air into smelting furnaces.31 Roman period settlement finds also come from the northeastern and eastern foot of the Gobavica hill (the Stare estate; Fig. 2: 19).32 The southernmost point of the inhabited area is the remains of a large masonry building at Jelovškova ulica 2 (Fig. 2: 16). 30 Sagadin 1995, 217–245. 31 Sagadin 1995, 224–230. 32 Šašel 1954, 14. Sl. 7: Mengeš. Območje med osnovno šolo in športno dvorano (sl. 2: 11). Žgana grobova 2 in 6. – 1 [gr. 6] železo; ostalo keramika. M. 1 [gr. 6] = 1:2; ostalo 1:3. Fig. 7: Mengeš. Area between the school and the sports hall (Fig. 2: 11). Cremation graves 2 and 6. – 1 [Grave 6] iron, the rest pottery. Scale 1 [Grave 6] = 1:2; rest 1:3. Sl. 8: Mengeš. Semesadika (sl. 2: 5). Oljenka. Keramika. M. = 1:2. Fig. 8: Mengeš. Semesadika (Fig. 2: 5). Oil lamp. Ceramics. Scale = 1:2. GROBIŠČA V Mengšu je bil najden en nagrobnik.33 Vklesana imena kažejo bodisi italsko (Rusticus) bodisi staroselsko (Lascontia) poreklo prebivalcev.34 Najstarejša rimska grobišča se razporejajo ob ob-robju doslej ugotovljenega naselbinskega areala, mlajša pa ležijo tudi sredi njega. Na prostoru med osnovno šolo in športno dvorano (sl. 2: 11) je bilo odkritih šest žganih in en skeletni grob, ki so datirani od druge polovice 2. do 4. st. po Kr. (sl. 7). Inventar žganih grobov kaže sorodnosti z emonskimi grobišči.35 Grobišče je nedvomno segalo še do športne dvora­ne (sl. 2: 10), kjer so bile ob gradnji najdene posamezne človeške kosti.36 Sicer pa so bili grobovi odkriti od Muljave (sl. 2: 18) in Pristave (sl. 2: 17) ter Slomškove ulice (sl. 2: 7) na jugu, do Staretove drevesnice na seve­ru (sl. 2: 3) in najbolj oddaljenega groba ob Kamniški cesti (sl. 2: 6) na severovzhodu.37 Grobnemu inventarju verjetno pripada tudi lonček iz Kersnikove ulice (sl. 2: 15).38 Najmlajši grobovi, ki segajo že v poznoantično obdobje, so bili najdeni na vzhodnem obrobju naselja, na Zavrteh (sl. 2: 12,13) in na Zadružniški ulici (sl. 2: 14).39 Ugotovljene so bile različne oblike pokopov. Leta 1834 (točna lokacija ni znana) je bil odkopan obokan grob iz opečnih plošč, v njem pa keramična žara in steklen balzamarij.40 Na grobišču med osnovno šolo in športno dvorano (sl. 2: 11) je bil en žarni grob položen v kvadratno jamo, obloženo s tegulami. Ostali žgani grobovi so bili brez grobne konstrukcije, pepel pokojnika pa je bil bodisi v keramični žari bodisi prosto položen v jamo. Skeletni grob je bil brez grobne konstrukcije. Žarni grob ob Kamniški cesti (sl. 2: 6) je bil obzidan s kamnitim zidcem in pokrit s tegulo. Z območja Pristave (sl. 2: 17) 33 CIL 3895; AIJ 216. 34 Šašel 1954, 15. 35 Sagadin 1997, 172; Sagadin 1999, sl. 66–71. 36 Sagadin 1989, 228. 37 Sagadin 1999, 41. 38 Zupančič 1979, 20, t. 1: 6. 39 Železnikar 2000. 40 Šašel 1954, 14. CEMETERIES Only one tombstone was discovered in Mengeš.33 The names carved on it suggest an either Italian (Rusticus) or autochthonous (Lascontia) origin of the inhabitants.34 The earliest Roman cemeteries are arranged along the edge of the identified inhabited area, while the later cemeteries lie also within the inhabited area. Between the school and the sports hall, six cremation graves and one inhumation grave were discovered (Fig. 2: 11). They are dated from the second half of the 2nd century to the 4th century AD (Fig. 7). The inventory of the cremation graves shows similarities with the cemeteries of Emona.35 There is no doubt that the cemetery extended all the way to the sports hall (Fig. 2: 10), during the construction of which isolated human bones were discovered.36 Other graves were unearthed in the area between Muljava (Fig. 2: 18), Pristava (Fig. 2: 17), and Slomškova ulica (Fig. 2: 7) in the south, and the Stare tree nursery in the north (Fig. 2: 3). The most distant grave lies near Kamniška cesta (Fig. 2: 6) in the northeast.37 A small pot from Ker-snikova ulica probably comes from a grave as well (Fig. 2: 15).38 The latest graves are already from the Late Antique period and were discovered on the eastern edge of the settlement, in Zavrti (Fig. 2: 12,13) and in Zadružniška ulica (Fig. 2: 14).39 Various forms of burial have been identified. In 1834, a vaulted grave was excavated (exact location unknown). Made of brick slabs, it contained a ceramic urn and a glass balsamarium.40 In the cemetery between the school and the sports hall (Fig. 2: 11), an urn was discovered in a square pit lined with tegulae. Other cremation graves had no grave constructions and the ashes of the deceased were either in a ceramic urn or deposited directly into 33 CIL 3895; AIJ 216. 34 Šašel 1954, 15. 35 Sagadin 1997, 172; Sagadin 1999, figs. 66–71. 36 Sagadin 1989, 228. 37 Sagadin 1999, 41. 38 Zupančič 1979, 20, pl. 1: 6. 39 Železnikar 2000. 40 Šašel 1954, 14. Sl. 9: Rimska poselitev okolice Mengša. Fig. 9: Roman settlement in the surroundings of Mengeš. je ohranjena zgoraj odrezana amfora, ki dokazuje tudi pokop v amfori. V Slomškovi ulici (sl. 2: 7) je bil odkrit grob v kamniti pepelnici, pokriti s kamnito ploščo.41 OKOLICA MENGŠA Naselbinske najdbe so v starejših virih (1847) omenjene na območju Drnovega (sl. 2: 2), severno od sedanjega jedra Mengša. Danes je to obsežna poljedelska površina brez sledov arheoloških ostalin antične dobe. Iz neposredne okolice Mengša (sl. 9) so znani antič­ni naselbinski kompleksi pri cerkvi sv. Primoža in Felici­jana v Loki pri Mengšu, iz Suhadol, Žej pri Komendi in Most pri Komendi. Širše območje Mengša – od Loke na 41 Sagadin 1999, 42; sl. 37; 39; 40; št. 62. the pit. The inhumation grave had no special grave con­struction. The urn grave at Kamniška cesta (Fig. 2: 6) was surrounded with a small stone wall and covered with a tegula. Burial in amphoras was practiced as well, as sug­gested by an amphora with its upper part cut off, which survived in the Pristava area (Fig. 2: 17). In Slomškova ulica (Fig. 2: 7), a burial in a stone ash chest covered with a stone slab was discovered.41 THE SURROUNDINGS OF MENGEŠ According to earlier sources (1847), settlement remains are mentioned in the area of Drnovo (Fig. 2: 2), north of the present-day centre of Mengeš. Today, this 41 Sagadin 1999, 42, fig. 37, 39, 40, no. 62. jugu do Komende na severu, Križa na vzhodu in Vodic na severozahodu – pa izkazuje izredno veliko gostoto naselbin (sem sodijo še Lahovče, Sp. Brnik – letališče, Gora pri Komendi, Brezovica pri Komendi), ki dosegajo že medsebojno oddaljenost okoli 1,4 km, kar govori o optimalni izkoriščenosti razpoložljive plodne zemlje.42 RAZVOJ NASELJA Glede na široko razprostranjenost naselbinskih os­tankov na območju Mengša ne moremo računati na vilo rustiko, ampak na obstoj večjega poselitvenega območja. Jedro naselbine se je ob vznožju Gobavice širilo v smeri sever–jug vsaj 1,2 km daleč, v smeri vzhod–zahod pa ne več kot 300 m. Verjetno je naselbina ležala ob vicinalni cesti, ki se je nekje pri Grobljem odcepila od glavne ceste Emona–Celeia. Niz naselbinskih točk, kot so Loka pri Mengšu, Mengeš, Suhadole, Žeje, Moste in Lahovče, nakazuje potek te komunikacije. Ostanki rimske ceste se sicer omenjajo v gozdu med Žejami in Lahovčami, vendar trasa še ni bila raziskana. Obsežna prazgodovinska utrjena naselbina na Go-bavici, ki je segala tudi na vzhodno vznožje, je doživela še prihod Rimljanov. Ti so v stari naselbini sprva vsaj občasno nastanili manjši vojaški oddelek, hkrati se je razvilo novo naselje v ravnini. Zgodnejše naselbinske najdbe pokrivajo območje od osnovne šole do športne dvorane (sl. 2: 8,10). Glede na najdbe z Jelovškove ceste (sl. 2: 16) moramo s posameznimi objekti računati še na širšem območju. Razporeditev sočasnih grobišč določa mejo naselbine na severu v Staretovi drevesnici (sl. 2: 3), na jugu na Muljavi (sl. 2: 18) in Pristavi (sl. 2: 17), na jugovzhodu na Slomškovi ulici (sl. 2: 7), grob na Kamniški cesti (sl. 2: 6) pa verjetno označuje potek lokalne komunikacije proti Kamniku in Tuhinjski dolini. Konec 2. in v 3. st. se pojavijo grobovi znotraj tako začrtanega območja (sl. 2: 11). Zdi se, da se v tem času naselbina pomakne proti severu, na območje vrtca Go-bica (sl. 2: 9) in Semesadike (sl. 2: 5; 4; 8). Tu se pojavijo značilne poznoantične polzemljanke (sl. 10). Pri naj­bolje ohranjenih smo lahko ugotovili približno velikost (3,5 × 4 m) in približno pravokotno obliko. Objekti so bili delno vkopani v prodnato podlago (do 40 cm), pri nekaterih (objekta 2 in 6) je bilo opaziti, da je strešna konstrukcija slonela na vertikalnih kolih. Stene so bile ometane z ilovico. Najmlajši grobovi z Zadružniške ulice (sl. 2: 14) in Zavrti (sl. 2: 12,13) lahko pripadajo tej naselbini. Posa­mezni grobovi s tega območja se lahko že povezujejo s podatki o najdbi gotske ločne fibule z Zadružniške ulice (sl. 2: 14) in o najdbi dveh pasnih spon polihromnega stila pri župni cerkvi sv. Mihaela (sl. 2).43 Tudi na Pri­stavi naj bi bila najdena ogrlica z jagodami millefiori 42 Sagadin 2008, 174. 43 Knific 1999, 47. is an extensive agricultural area with no archaeological traces from the Roman period. Roman period settlement complexes (Fig. 9) in the immediate surroundings of Mengeš are situated at the church of Sts Primus and Felician in Loka near Mengeš,in Suhadole, in Žeje pri Komendi, and in Most pri Ko­mendi. The wider surroundings of Mengeš – from Loka in the south to Komenda in the north, and from Križ in the east to Vodice in the northwest – show a very dense settlement pattern (including the settlements in Lahovče, Spodnji Brnik – airport, Gora pri Komendi, Brezovica pri Komendi), with the distance between the sites about 1.4 km, which implies a great efficiency in the use of the available arable land.42 DEVELOPMENT OF THE SETTLEMENT Judging by the wide distribution of settlement re­mains in the area of Mengeš, we cannot speak about a villa rustica, but about a larger inhabited area. The core of the settlement at the foot of Gobavica extended at least 1.2 km in the north–south direction, but no more than 300 m in the east–west direction. The settlement was probably situated along the vicinal road, which branched off from the main Emona–Celeia road in the vicinity of Groblje. A series of settlement points such as Loka pri Mengšu,Mengeš, Suhadole, Žeje, Moste, and Lahovče, suggests the course of the communication route. While there are several references to a Roman road in the forest between Žeje and Lahovče, no surveys have been undertaken. The large prehistoric fortified settlement on top of Gobavica, which stretched all the way to the eastern foot of the hill, survived into the time of the Roman arrival. While a small Roman military unit was at least occasion­ally stationed in the old settlement, a new settlement developed in the plain below. Early settlement finds cover the area between the school and the sports hall (Fig. 2: 8,10). Judging by the finds from Jelovškova cesta (Fig. 2: 16), isolated buildings can be expected in the wider area. The arrangement of contemporary cemeteries defines the northern border of the settlement in the Stare tree nursery (Fig. 2: 3), the southern border in Muljava (Fig. 2: 18) and Pristava (Fig. 2: 17), the southeastern border in Slomškova ulica (Fig. 2: 7), while the grave in Kamniška cesta (Fig. 2: 6) probably marks the course of the communication route towards Kamnik and the Tuhinj Valley. At the end of the 2nd century and in the 3rd century, graves also appear in the middle of this area (Fig. 2: 11). It seems that in that time the settlement moved towards the north, to the present-day Gobica kindergarten (Fig. 2: 9) and Semesadike company (Fig. 2: 5), where typical Late Antique pit houses were discovered. The best-preserved ones enable an estimation of their size (3.5 x 4 m) and 42 Sagadin 2008, 174. (sl. 2: 17). Nadaljevanje poselitve v zgodnji srednji vek dokazujejo staroslovanski grobovi, odkriti v okolici sv. Mihaela (sl. 2).44 Naselje se – posredno z omembo listinske pri-če Dietricusa de Meingosburg – prvič omenja leta 1154/1156.45 44 Knific 1999, 50. 45 Železnikar 1999b. AIJ = V. Hoffiller, B. Saria, Antike Inschriften aus Jugoslavien 1. Noricum und Pannonia Superior, Zagreb 1938. CIL = Corpus inscriptionum Latinarum. GABROVEC, S. 1965, Kamniško ozemlje v prazgodovini. – Kamniški zbornik 10, 1965, 89–134. GABROVEC, S. 1975, Mengeš. – V / In: Arheološka najdišča Slovenije. – Ljubljana. HORVAT, J. 2015, Early Roman military finds from prehistoric settlements in the Gorenjska region / Zgodnjerimske vo­jaške najdbe s prazgodovinskih naselbin na Gorenjskem. – V / In: J. Istenič, B. Laharnar, J. Horvat (ur. / eds.), Evidence of the Roman army in Slovenia / Sledovi rimske vojske na Slovenskem, Katalogi in monografije 41, 171–208.¸ KNIFIC, T. 1999, Zgodnji srednji vek. – V / In: J. Železnikar (ur. / ed.), Poselitvena podoba Mengša in okolice od prazgodovine do srednjega veka, 47–50, Mengeš. MÜLLNER, A. 1879, Emona. Archäologische Studien aus Krain. – Laibach. MÜLLNER, A. 1898, Prähistorischer Stahl von Mannsburg. – Argo 6, 56. NOVŠAK, M., R. ERJAVEC, T. VERBIČ 2009, Arheološko vrednotenje dveh vrtin na lokaciji OŠ Mengeš. – Poročilo / Report; Arhej, d.o.o. (neobjavljeno / unpublished). shape (approximately rectangular). The buildings were partly (up to 40 cm) sunk into the gravel ground (Fig. 10). The roof of some of them (Buildings 2 and 6) was supported by vertical posts. The walls had a clay coating. The latest graves from Zadružniška ulica (Fig. 2: 14) and Zavrti (Fig. 2: 12,13) might belong to this settlement. Individual graves from this area could already be associ­ated with the mention of a Gothic bow brooch, discovered in Zadružniška ulica (Fig. 2: 14), and two polychrome style belt buckles found near the parish church of St Michael (Fig. 2).43 A necklace with millefiori beads was supposedly discovered in Pristava (Fig. 2: 17). The area continued to be inhabited into the Early Middle Ages, as proven by the early Slavic graves discovered in the vicinity of the church of St Michael (Fig. 2).44 The settlement is first mentioned – indirectly, via Dietricus de Meingosburg, who was a witness to a deed – in 1154/1156.45 Translation: Meta Osredkar 43 Knific 1999, 47. 44 Knific 1999, 50. 45 Železnikar 1999b. NOVŠAK, M., A. ŠINKOVEC, T. VERBIČ 2008, Arheološko vrednotenje vrtin na lokaciji Mengeš – Osnovna šola, Po­ročilo. – Poročilo / Report; Arhej, d.o.o. (neobjavljeno / unpublished). NOVŠAK, M., T. VERBIČ, R. ERJAVEC 2010, Arheološko vrednotenje treh vrtin na lokaciji OŠ Mengeš; dodatek k poročiloma. Mengeš – Osnovna šola. Predhodne arheološke raziskave – vrednotenje arheološkega potenciala. – Poročilo / Report; Arhej, d.o.o. (neobjavljeno / unpublished) [http:// www.arhej.com/datoteke/Pdf/0052010_porocilo_menges. pdf]. SAGADIN, M. 1989, Mengeš. – Varstvo spomenikov 31, 1989, 227–228. SAGADIN, M. 1995, Mengeš v antiki (Mengeš in the Roman Period). – Arheološki vestnik 46, 1995, 217–245. SAGADIN, M. 1997, Mengeš. – Varstvo spomenikov 36, 1997, 172–173. SAGADIN, M. 1999, Rimsko obdobje. – V / In: J. Železnikar (ur. / ed.), Poselitvena podoba Mengša in okolice od praz­godovine do srednjega veka, 39–46, Mengeš. SAGADIN, M. 2006, Mengeš – poznoantično grobišče na Zavrteh. – Varstvo spomenikov 39/41, 2000–2004 (2006). Poročila, 101. SAGADIN, M. 2008, Od Karnija do Kranja. Arheološki podatki o razvoju poselitve v antičnem in zgodnjesrednjeveškem obdobju. – Doktorska disertacija / PhD thesis, Oddelek za arheologijo Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). SAGADIN, M., Š. PEČE 2007, Mengeš. Arheološko območje Gobavica. Jelovškova ulica 2. – Varstvo spomenikov 44, 2007 (2008). Poročila, 153–154. STRAŽAR, S. 1993, Mengeš in Trzin skozi čas. – Mengeš, Trzin. ŠAŠEL, J. 1954, Rimska doba. – Mengeški zbornik 1, 13–17. ŠTIBERNIK, G. 2001, Mengeš, Zadružniška 2001. Poročilo o izkopavanju. – Poročilo / Report; ZVKDS OE Kranj (ne­objavljeno / unpublished). ŠTIBERNIK, G. 2006a, Mengeš. Mengeš – arheološko območje Gobavica. – Varstvo spomenikov 39/41, 2000–2004 (2006). Poročila, 98–99. ŠTIBERNIK, G. 2006b, Mengeš. Mengeš – grobišče na Zadružniški ulici. – Varstvo spomenikov 39/41, 2000–2004 (2006). Poročila, 99–100. TURK, P. 1999, Starejša železna doba – halštatsko obdobje. - V / In: J. Železnikar (ur. / ed.), Poselitvena podoba Mengša in okolice od prazgodovine do srednjega veka, 31–39, Mengeš. URANKAR, R., H. BEŠTER 2016, Arheološko izkopavanje v Mengšu zaradi prizidka OŠ Mengeš, poročilo. – Poročilo / Report; ZVKDS OE Kranj (neobjavljeno / unpublished). ZUPANČIČ, M. 1979, Arheološki najdišči Trojane in Mengeš. – V / In: Zbornik občine Domžale, 15–22, Domžale. ŽELEZNIKAR, J. 1999a, Mlajša železna doba – latenskoobdobje. – V / In: J. Železnikar (ur. / ed.), Poselitvena po­doba Mengša in okolice od prazgodovine do srednjega veka, 37–39, Mengeš. ŽELEZNIKAR, J. 1999b, Srednji vek. – V / In: J. Železnikar (ur. / ed.), Poselitvena podoba Mengša in okolice od prazgodovine do srednjega veka, 50–51, Mengeš. ŽELEZNIKAR, J. 2000, Preliminarno poročilo o zaščitnem arheološkem izkopavanju na Zavrteh. – Poročilo / Report; ZVKDS OE Kranj (neobjavljeno / unpublished). Milan SagadinŽupančičeva ulica 39 SI-4000 Kranj sagadin.milan@gmail.com Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 201–212 CARNIUM – KRANJ Milan SAGADIN Izvleček Kranj leži na naravno zavarovanem skalnem pomolu nad sotočjem Save in Kokre. Neolitska naselbina je obsegala skrajni južni del pomola, staroželeznodobna poselitev je presegla meje pomola in segla vse do Save, medtem ko je bila latenska poselitev manj intenzivna. V srednje- in poznoavgustejskem obdobju je na pomolu ležala naselbina priseljencev iz Italije, obdana s kamnitim obzidjem in stolpi, v notranjosti pa so stale predvsem lesene stavbe. Med drobnimi najdbami prevladuje fina keramika, uvožena iz severne Italije, malo je staroselske keramike, nekaj predmetov pa kaže tudi na prisotnost rimske vojske. Po koncu avgustejskega obdobja je naselbina zamrla. V pozni rimski dobi je zaradi naravno zavarovanega položaja Kranj znova pridobival na pomenu. V 4. st. sodijo ka­kovostno zidani objekti. Naselje Carnium je prvič omenjeno v pisnem viru, ki se nanaša na 5. in 6. st. V 6. st. je Carnium obsegal ves pomol, obdan je bil s kamnitim obzidjem, imel je cerkev ter dve obsežni grobišči. V notranjosti je bila odkrita cela vrsta bivalnih polzemljank, raziskani pa so bili tudi ostanki obrtniških delavnic. Ključne besede: Italija (10. regija), Kranj, Carnium, rimska doba, avgustejska doba, pozna antika, naselbina, obzidja, polzemljanke, keramika, steklo, orožje Abstract The town of Kranj is situated on a naturally protected rocky promontory above the confluence of the Sava and Kokra rivers. Neolithic settlement covered the southernmost part of the promontory. In the Early Iron Age, the inhabited area ex­tended beyond the limits of the promontory, all the way to the Sava, while in the La Tene period, settlement was less intense. In the Middle and Late Augustan periods, the promontory hosted a settlement of immigrants from Italy. The settle­ment was surrounded by a stone wall and towers, while the buildings in the interior were mostly made of wood. Fine pottery imports from northern Italy predominate among the small finds and local autochthonous pottery is uncommon. Some artefacts indicate the presence of the Roman army. The settlement died out after the end of the Augustan period. The significance of Kranj increased again in the Late Roman period, due to its naturally defended position. Some well-constructed buildings date to the 4th century. The settlement of Carnium is first mentioned in a written source referring to the 5th and 6th centuries. In the 6th century, Carnium extended over the entire promontory. Surrounded by a stone wall, it had a church and two large cemeteries. A series of pit houses and remains of craft workshops were discovered in the interior. Keywords: Italy (Regio X), Kranj, Carnium, Roman period, Augustan period, Late Antiquity, settlement, defensive wall, pit house, pottery, glass, weapons https://doi.org/10.3986/9789610502586_11 LEGA Kranj leži v severozahodnem delu Ljubljanske kot-line, na konglomeratnem pomolu nad sotočjem Kokre in Save, ki tukaj tvorita preko 30 m globoka kanjona. Konglomeratna osnova pripada konglomeratnemu za­sipu Save in Kokre. Ta je na tem mestu višji kot mlajši prodni zasip, ki tvori okoliško Kranjsko in Sorško polje. Za prehod iz Ljubljanske kotline v severno Italijo ali na Koroško ima Kranj izrazito strateško pomembno lego, saj varuje dohode v Gornjesavsko dolino ali do gorskih prelazov v Karavankah. Najstarejše ohranjeno ime naselja na sotočju Save in Kokre je Carnium. Prvič je zabeleženo v delu Cosmo-graphia Anonimnega geografa iz Ravene. Tekst sodi na konec 7. ali v začetek 8. st., vendar po lastnih navedbah avtorja uporablja vire iz 5. in 6. st. Politična situacija, ki jo opisuje v zvezi z opisom pokrajine Carneole, ustreza položaju sredine 6. st. Med petindvajsetimi naselji, ki jih našteva, je na prvem mestu omenjen Carnium. Navaja tudi ime reke Corcac (Kokra).1 Očitna imenska kontinuiteta in veliko število arheoloških najdb iz 6. st. v Kranju opravičujeta identifikacijo antičnega Karnija s sedanjim Kranjem. ZGODOVINA RAZISKAV Prvi zgodovinopisci so svoje zanimanje za arheo­loška obdobja gradili predvsem na najdbah iz obdobja antike. Z območja Kranja so pozornost pritegnile novčne najdbe in nagrobnik iz kapele sv. Petra iz Stražišča pri Kranju (na desnem bregu Save).2 Na podlagi teh najdb in navedb pri Klavdiju Ptolemeju ter v Antoninskem itinerariju so Kranj identificirali z antičnim Santikom.3 Vendar sta že Müllner in Globočnik opustila to identi­fikacijo.4 V devetdesetih letih 19. stoletja so pozornost začele buditi prazgodovinske (starejšeželeznodobne) grobne najdbe z območja Mayerjeve pristave (današnja Gregorčičeva ulica in okolica, sl. 1: 3), tedanjega mestne­ga pokopališča (sedaj Prešernov gaj, sl. 1: 1–2), mestne sirotišnice in Mladinskega doma (današnja Stritarjeva ulica, sl. 1: 4).5 Konec 19. in v začetku 20. stoletja se je pozornost arheologije preusmerila na izkopavanja veli­kega grobišča iz obdobja preseljevanja ljudstev v Lajhu (sl. 4: 8).6 Tik pred 1. svetovno vojno so se nadaljevale naključne najdbe prazgodovinskih grobov zahodno od 1 Šašel 1970–1971, 33–44. 2 CIL III 3892; AIJ 215. Linhart 1788, 322; Radics 1862; Hitzinger 1856, 26; Maurig 1893, 463. 4 Müllner 1879, 262; Globočnik 1889. 5 Pečnik 1893, 80; Pečnik 1894, 184; Pečnik 1904, 127– 128 (Ljubljana, 2. Kranj); Schmid 1908, 214–215; Schmid 1909, 154–155. 6 Obsežen historiat prinaša Knific 1995, 23–24. POSITION Located in the northwestern part of the Ljubljana Basin, Kranj lies on a conglomerate promontory above the confluence of the Kokra and Sava rivers, which here form two more than 30 m deep canyons. The conglomerate base belongs to the conglomerate accumulation deposited by the Sava and the Kokra, which here rises above the later accumulation of pebbles forming the surrounding plain of Kranj and Sora. Protecting the access points to the Upper Sava Valley and the mountain passes of the Karavanke Mountains, Kranj has great strategic significance for the transport from the Ljubljana Basin to northern Italy and Carinthia. The earliest known name of the settlement at the confluence of the Sava and the Kokra is Carnium. It is first mentioned in the work Cosmographia by the anonymous geographer of Ravenna. While the text can be dated to the end of the 7th or the beginning of the 8th century, the author claims to have used sources from the 5th and 6th centuries. The described political situation in the land of Carneola corresponds to the situation in the mid-6th century. Carnium is the first of the 25 listed settlements. The name of the river Corcac (Kokra) is also mentioned.1 The evident continuity of the name and a large number of 6th century archaeological finds in Kranj justify the identification of Carnium with present-day Kranj. RESEARCH HISTORY The interest of the first historians focused primar­ily on the finds from the Roman period. Their attention was caught by coins from Kranj and a tombstone from the chapel of St Peter in Stražišče near Kranj (right bank of the Sava).2 On the basis of these finds and references from Claudius Ptolemy and the Antonine Itinerary, Kranj was identified with the Roman Santicum.3 But Müllner and Globočnik already abandoned this identification.4 In the 1890s, the focus shifted to the prehistoric (Early Iron Age) grave finds from Mayerjeva pristava (present-day Gregorčičeva ulica and its surroundings, Fig. 1: 3), the then town graveyard (present-day park of Prešernov gaj, Fig. 1: 1–2), the town orphanage and youth centre (present-day Stritarjeva ulica, Fig. 1: 4).5 At the end of the 19th and the beginning of the 20th century, archaeology focused on the excavations of a large Migration period cemetery in Lajh (Fig. 4: 8).6 Just before World War I, 1 Šašel 1970-1971, 33–44. 2 CIL III 3892; AIJ 215. 3 Linhart 1788, 322; Radics 1862; Hitzinger 1856, 26; Maurig 1893, 463. 4 Müllner 1879, 262; Globočnik 1889. 5 Pečnik 1893, 80; Pečnik 1894, 184; Pečnik 1904, 127– 128; Schmid 1908, 214–215; Schmid 1909, 154–155. 6 Overview by Knific 1995, 23–24. Sl. 1: Kranj. Prazgodovinska najdišča. Grobni ostanki (1–8) in depo (9). M. = 1:10.000. Fig. 1: Kranj. Prehistoric sites. Graves (1–8) and hoard (9). Scale = 1:10,000. (Vir / Source: GURS©2004 TTN5, pomanjšano / diminished) 1, 2 – Prešernov gaj (staro mestno pokopališče); 3 – Gregorčičeva ulica (nekdanja Mayerjeva pristava z okolico); 4 – Stritarjeva ulica (nekdanja mestna sirotišnica in mladinski dom); 5 – Mladinska ulica; 6 – Bleiweisova cesta; 7 – Stošičeva ulica (nekdanja Prahova vila); 8 – Koroška cesta; 9 – Ljubljanska cesta Mayerjeve pristave (Mladinska ulica, sl. 1: 5), ki so se nadaljevale in zaključile pred 2. svetovno vojno z najd­bami ob Prahovi vili (Stošičeva ulica in Koroška cesta, sl. 1: 7–8).7 Po vojni je bil najden še mlajšehalštatski bojevniški grob v parku ob Bleiweisovi cesti (sl. 1: 6).8 Sredi tridesetih let dvajsetega stoletja je bilo najdenih more prehistoric graves were accidentally discovered west of Mayerjeva pristava (Mladinska ulica, Fig. 1: 5). Several similar discoveries followed until the beginning of World War II, the last being the finds at Vila Prah (Stošičeva ulica and Koroška cesta, Fig. 1: 7–8).7 After the war, a Late Hallstatt warrior grave was discovered in the 7 Pregled: Škvor Jernejčič 2017, 117–196. 8 Stare 1954, 112–127. 7 Škvor Jernejčič 2017, 117–196. nekaj antičnih najdb v starem mestnem jedru, ki so bile interpretirane kot grobne.9 Med 2. svetovno vojno je bil ob gradnji stavbe Stritarjeva 8 (na območju prazgodo­vinskega grobišča) najden rimski grob (sl. 2: 11),10 več antičnih najdb iz bližnje Mayerjeve pristave pa opozarja na morebiten obstoj antičnega grobišča na tem območju (sl. 2: 12).11 Po vojni, od 1953 do 1972, so bile arheo­loške raziskave osredotočene na izkopavanje velikega srednjeveškega in novoveškega grobišča ob župni cerkvi, leta 1984 v notranjosti cerkve.12 Od leta 1990 pa so se domala vsakoletno opravljala zaščitna arheološka izko­pavanja v srednjeveškem mestnem jedru. Po letu 2004 so se znova razmahnila izkopavanja grobišča v Lajhu, ki so v velikem obsegu potekala vzporedno z raziskavami naselbine do leta 2013.13 PRAZGODOVINA Najstarejša poselitev je iz obdobja mlajšega neoliti­ka, pripadala je t. i. savski skupni. Naselbina je bila v tem času omejena na skrajni južni konec konglomeratnega pomola med Savo in Kokro (današnji Trubarjev trg in južni del Cankarjeve ulice, sl. 1).14 Iz obdobja bronaste dobe domala ni najdb, velik razcvet pa je naselbina doživela v začetku stare železne dobe, na prehodu med HaB3 in HaC1. Takrat je prekri­vala celoten areal kasnejšega srednjeveškega mesta, na severu in zahodu pa ga je celo presegala, saj se je širila do levega brega Save na zahodu in do začetka današnje Koroške ceste na severu (sl. 1). Najdeni so bili značilni sledovi metalurške dejavnosti v naselju, ostanek tega je tudi izkopani depo 22 kg razbitih bronastih predmetov in surovcev iz konglomeratnega spodmola pod zahod­nim robom naselja, ob Ljubljanski cesti (sl. 1: 9).15 Pripadajoče prazgodovinsko grobišče se je raztezalo po robu nekdanje rečne terase severno od naselja, od današnjega Prešernovega gaja na vzhodu (Partizanska in Gregorčičeva ulica, sl. 1: 1–3), preko Stritarjeve, Mla­dinske in Stošičeve ulice (sl. 1: 4–7) do Koroške ceste na zahodu (sl. 1: 8).16 V obdobju mlajše železne dobe je naselbina močno opešala, vendar je vztrajala do prihoda Rimljanov, ko je prišlo do izrazitega mešanja poznolatenskih najdb 9 Žontar 1939, 11. 10 Valič 1960, 33. 11 Pečnik 1894, 184; Pečnik 1904, 127–128 (Ljubljana, 2. Kranj); Knific 1971, 76. 12 Kastelic 1960, 41; Valič 1975, 159–167; Pleterski, Štu­lar, Belak 2016; 2017; Pleterski, Štular, Belak , Bešter 2019. Najobsežnejši posegi, predstavljeni v poročilih: Lux 2011; Tomažinčič 2011; Urankar 2011; Urankar, Bešter 2014; Urek et al. 2016. 14 Sagadin 2005, 33; Urankar, Bešter 2014, 24–29. 15 Horvat 1983, 140–218; Modrijan 1988; Rozman 2004, 55–109. 16 Gabrovec 1960, 11–30; Škvor Jernejčič 2017, 117–196. park near Bleiweisova cesta (Fig. 1: 6).8 In the mid-1930s, Roman period finds – interpreted as grave finds – were discovered in the old core of the town.9 In the time of World War II, a Roman grave was discovered during the construction of the building at Stritarjeva 8 (Fig. 2: 11).10 Roman period finds from the nearby Mayerjeva pristava suggest a possible Roman cemetery at that loca­tion (Fig. 2: 12)11. Between 1953 and 1972, archaeological research focused on the excavation of a large Medieval and early modern cemetery at the parish church. In 1984, excavations were conducted in the interior of the parish church.12 and since 1990, there have been almost annual rescue excavations in the medieval town core. After 2004, excavations of the Lajh cemetery were again in full swing and lasted, simultaneously with the investigations of the settlement, until 2013.13 PREHISTORY The earliest settlement is from the Late Neolithic period and it belongs to the so-called Sava Group. It was limited to the southernmost part of the conglomerate promontory above the Sava and the Kokra (present-day Trubarjev trg and the south part of Cankarjeva ulica, Fig. 1).14 While there are almost no finds from the Bronze Age, the settlement flourished at the beginning of the Early Iron Age, i.e. at the transition between the HaB3 and HaC1 phases. At that time, it covered the entire area of the later medieval town, and extended even further to the north and west: to the left bank of the Sava on the west and to the beginning of present-day Koroška cesta on the north (Fig. 1). Typical traces of metallurgical activity within the settlement have been discovered, including a 22 kg hoard of broken bronze objects and ingots from a conglomerate rock shelter at Ljubljanska cesta (Fig. 1: 9).15 The corresponding prehistoric cemetery extended along the edge of a former river terrace north of the settlement, from present-day Prešernov gaj on the east (Partizanska in Gregorčičeva cesta, Fig. 1: 1–3), over Stritarjeva, Mla­dinska and Stošičeva ulica (Fig. 1: 4–7) to Koroška cesta on the west (Fig. 1: 8).16 While the settlement experienced a strong decline in the Late Iron Age period, it survived until the arrival of the Romans, when late La Tene finds 8 Stare 1954, 112–127. 9 Žontar 1939, 11. 10 Valič 1960, 33. 11 Pečnik 1894, 184; Pečnik 1904, 127–128 (Ljubljana, 2. Kranj); Knific 1971, 76. 12 Kastelic 1960, 41; Valič 1975, 159–167; Pleterski, Štular, Belak 2016; 2017; Pleterski, Štular, Belak , Bešter 2019. 13 Lux 2011; Tomažinčič 2011; Urankar 2011; Urankar, Bešter 2014; Urek et al. 2016. 14 Sagadin 2005, 33; Urankar, Bešter 2014, 24–29. 15 Horvat 1983, 140–218; Modrijan 1988; Rozman 2004, 55–109. 16 Gabrovec 1960, 11–30; Škvor Jernejčič 2017, 117–196. Sl. 2: Kranj. Rimskodobna najdišča. Ostanki obzidja, stavb in grobov. M. = 1:5.000. Fig. 2: Kranj. Roman period. Remains of town walls, buildings, and graves. Scale = 1:5,000. (Vir / Source: GURS©2004, TTN5) 1, 4 – Tomšičeva 42 in 44 (severno dvorišče gradu Kieselstein); 2 – Tomšičeva 44 (južno dvorišče gradu Kieselstein); 3, 6 – Tom-šičeva ulica 38 (notranjost in dvorišče); 5 – Tomšičeva ulica 26; 7 – Reginčeva ulica; 8 – Ljubljanska cesta; 9 – Glavni trg 21; 10 – Prešernova ulica; 11 – Stritarjeva ulica; 12 – Gregorčičeva ulica (nekdanja Mayerjeva pristava z okolico) (pripadajočih mokronoški skupini) in zgodnjerimskega (srednjeavgustejskega) importa v istih kontekstih.17 RIMSKA NASELBINA Arheološki ostanki rimske dobe (drobne najdbe in objekti) se pojavljajo po celotnem konglomeratnem pomolu (od Trubarjevega trga oz. Pungarta do začetka Ljubljanske ceste), poleg tega pa tudi po pobočjih do levega brega Save (nekdanja ledina Lajh, zdaj Sejmišče in Savska cesta). Posamezne grobne najdbe so znane sever-no od tod – ob gradnji občinske stavbe Stritarjeva 8, pri Mayerjevi pristavi (Prešernov gaj in Gregorčičea ulica, sl. 2: 11,12), drobne naključne najdbe pa tudi v širši okolici naselja (v smeri proti grobišču – Ulica XXXI. divizije; Stražišče, na desnem bregu Save, itd.). Objekti, v katerih so bile srednje- in poznoavgustej­ske najdbe, so bili izdelani s še povsem prazgodovinsko tehniko – lesene sohe, stene iz vejnatega prepleta, prema­zane z ilovico (najdišča: severno dvorišče Kieselsteina, Ljubljanska cesta, Glavni trg 21, Prešernova ulica, itd. – sl. 2: 4,8–10). Večina najdb iz tega obdobja pa je bila najdena v različnih vkopih v matično osnovo, ker je bila kulturna plast zlasti po sredini kranjskega pomola zaradi kasnejših izravnav večidel odstranjena. Edini zidani objekt, ki zanesljivo spada v zgodnje­rimsko obdobje, je obzidje. Ugotovljeno je bilo v več odsekih na severnem in južnem dvorišču gradu Kiesel-stein (Tomšičeva 42 in 44, sl. 2: 1,2; 3) ter na dvorišču stavbe Tomšičeva 38 (sl. 2: 3; 3). Na tem delu je obzidje imelo tudi vsaj en (po vsej verjetnosti pa dva) pravokotni stolp (sl. 3). Debelo je bilo od 80 do 90 cm, zidano iz lomljencev, večjih rečnih oblic in mestoma tudi iz kosov konglomerata. Ohranjene višine so skromne in dosegajo 0,5 m.18 Glede na ostanke freskiranega/slikanega ometa na dvorišču stavbe Tomšičeva 38 lahko predvidevamo, da so tudi v zgodnji antiki tu stali še drugi zidani objek-ti. Tudi nekateri šibko ohranjeni temelji na severnem dvorišču gradu Kieselstein so datirani v zgodnje rimsko obdobje. Drobne najdbe pripadajo predvsem sever-noitalski teri sigilati (oblike Consp. 12, 13, 14, R 2),19 skodelicam Sarius, čašam Aco, keramiki tankih sten (primerjave v oblikah 1–6, 9, 12–13, 27, 68 s Štalenške gore),20 cilindričnim in reliefnim oljenkam, amforam (Dressel 7, 6b). Novčne najdbe obsegajo nekaj primerkov vzhodnokeltskega kovanja in republikanske novce, med kovinskimi najdbami pa najdemo fibule poznolatenske sheme, fibule tipov Nova vas, Jezerine, Alezija, Aucissa in močno profilirane fibule. Nekaj kovinskih najdb pripada tudi odlomkom srednje- in poznoavgustejske 17 Gabrovec 1966, 243–270; Sagadin 2008, 84–87; Guštin 2011. 18 Sagadin 2010, 16–25, sl. 2–5. 19 Conspectus 1990. 20 Schindler-Kaudelka 1975. (from the Mokronog Group) and early Roman (Middle Augustan) imports are present in the same contexts.17 ROMAN SETTLEMENT Archaeological remains (small finds and buildings) from the Roman period occur all over the conglomerate promontory (from Trubarjev trg to the beginning of Ljub­ljanska cesta) and also on the slopes all the way down to the left bank of the Sava (former Lajh, now Sejmišče and Savska cesta). Individual burial finds are known north of this area: they were discovered during the construction of a municipal building (Stritarjeva 8), near Mayerjeva pristava (Prešernov gaj and Gregorčičeva ulica, Fig. 2: 11, 12). Stray small finds were also discovered in the wider surroundings (north of the settlement: Ulica XXXI. di­vizije; right bank of the Sava: Stražišče, etc.). The buildings containing Middle and Late Augus­tan finds were still constructed in an entirely prehistoric fashion – wooden vertical beams, wattle-and-daub walls (north courtyard of Kieselstein Manor, Ljubljanska cesta, Glavni trg 21, Prešernova ulica, etc. – Fig. 2: 4,8–10). Most of the finds from that period were discovered in various pits dug into the parent material, since the cultural layer, especially along the middle of the Kranj promontory, had been largely removed due to the levelling of the ground. Town walls is the only masonry construction that can be dated with certainty to the Early Roman period. Several segments been identified in the north and south courtyards of Kieselstein Manor (Tomšičeva 42 and 44, Figs. 2: 1,2; 3) and in the courtyard of the building at Tomšičeva 38 (Figs. 2: 3; 3). Here, the walls had at least one (but probably two) rectangular towers (Fig. 3). The walls were between 80 and 90 cm thick, made of roughly cut stones, large river cobbles, and some occasional conglomerate pieces. Their preserved height is modest and reaches up to 0.5 m.18 Remains of frescoed/painted plaster in the courtyard of the building at Tomšičeva 38 imply that other masonry buildings from the Early Roman period may be expected at that location. Some poorly preserved stones in the north courtyard of Kieselstein Manor also date to the Early Roman period. Small finds are mostly represented by North Italian terra sigillata (the Consp. 12, 13, 14, R 2 types),19 Sarius cups, Aco beakers, thin-walled pottery (corresponding to Forms 1–6, 9, 12–13, 27, 68 from Magdalensberg),20 cylindrical and relief oil lamps, and amphorae (Dressel 7, 6b). Among the coins, there are specimens of East Celtic minting as well as Republican coins. Metal finds include fibulae of the late La Tene period, fibulae of the Nova vas, Jezerine, 17 Gabrovec 1966, 243–270; Sagadin 2008, 84–87; Guštin 2011. 18 Sagadin 2010, 16–25, fig. 2–5. 19 Conspectus 1990. 20 Schindler-Kaudelka 1975. vojaške opreme.21 Zgodnjerimski import se dosledno pojavlja v istih kontekstih kot poznolatenske najdbe mokronoške skupine. Po koncu avgustejskega obdobja je bila poselitev pomola na sotočju Save in Kokre prekinjena. Od 4. st. dalje pa je naselbina v Kranju ponovno oživela. Predvsem po zahodnem robu kranjskega po­mola (severno dvorišče gradu Kieselstein, Tomšičeva ulica 26, 28, 38, Reginčeva ulica, sl. 2: 4–7) so bili odkriti ostanki kakovostno zidanih objektov s pohodno povr­šino iz maltnega estriha, na severnem dvorišču gradu Kieselstein (sl. 2: 4) je bila ugotovljena tudi uporaba hipokavsta. Zidovi teh objektov so bili grajeni pretežno iz večjih oblic, manj je bilo uporabljenih lomljencev. Debelina zidov je znašala od 70 do 80 cm, žal pa ohranje­nost objektov ni bila tolikšna, da bi lahko rekonstruirali kolikor toliko zaključene stavbne celote.22 V nekaterih kontekstih (Tomšičeva ulica 26, sl. 2: 5) so bile najdene tudi posamezne mozaične kocke, vendar se kulturna plast tu ni navezovala na ostanke zidov.23 Datacija te faze poselitve je izrazito slabo podprta z drobnimi najdbami. Ostanki zidanih stavb so stratigrafsko mlajši kot zgodnje 21 Sagadin 2003, 71–81; Sagadin 2015, 185–189. 22 Urankar 2011. 23 Sagadin 2008, 44. Sl. 3: Kranj. Zgodnjeantično obzidje med Tomšičevo 38 in 44 – odseki na sl. 2: 1–3. M. = 1:1000 (po Sagadin 2015, sl. 2). Fig. 3: Kranj. Early Roman town walls between Tomšičeva 38 and 44 – segments in Fig. 2: 1–3. Scale = 1:1000 (after Sagadin 2015, sl. 2). Alesia, Aucissa types, and strongly profiled fibulae (kräftig profilierte Fibeln). There are also fragments of Middle and Late Augustan military equipment.21 Early Roman imports consistently appear in contexts together with Late La Tene finds of the Mokronog Group. After the Augustan period, there was a settlement hiatus on the promontory at the confluence of the Sava and Kokra rivers. From the 4th century on, the Kranj settlement recovered. Remains of well-built masonry houses with mortar walking surfaces were discovered especially along the western edge of the Kranj promontory (the north courtyard of Kieselstein Manor, Tomšičeva ulica 26, 28, 38, Reginčeva ulica, Fig. 2: 4,5,6,7), and a hypocaust was identified in the north courtyard of Kieselstein Manor (Fig. 2: 4). Mostly made of large cobbles and – to a lesser degree – quarry stones, the walls of these buildings were between 70 and 80 cm thick. Unfortunately, they were not sufficiently preserved to allow reconstructions of more or less complete buildings.22 Individual mosaic tiles were discovered in certain contexts (Tomšičeva ulica 26, Fig. 2: 5), but this cultural layer was not in contact with the remains of the walls.23 The dating of this settlement phase is poorly supported by small finds. The remains of masonry buildings are stratigraphically later than the Early Roman layer, and hypocausts do not appear in the Gorenjska region before the second half of the 2nd century AD.24 Finds from the 4th century AD (a Roman crossbow fibula, a finger ring with a key, a bronze stylus, bone spindle whorls)25 also appear elsewhere in Kranj. The only clues to the location of the Roman cemetery are modest remains in the area of Mayerjeva pristava and a Roman grave, excavated in the time of World War II during the construction of the building at Stritarjeva 8 (Fig. 2: 11,12). Individual Roman artefacts were discov­ered also north of that area.26 Since grave goods are not preserved, it is not possible to determine the period the cemetery belonged to. 21 Sagadin 2003, 71–81; Sagadin 2015, 185–189. 22 Urankar 2011. 23 Sagadin 2008, 44. 24 Judging by the dating of Roman settlement agglomera­tions (Sagadin 1995, 13–22); for the wider area oral confir­mation by Plesničar Gec (Mestni muzej Ljubljana). 25 Sagadin 2008, 61–78. 26 Valič 1960, 31–37. rimska naselbinska plast, hipokavst pa se na Gorenjskem ne pojavlja pred drugo polovico 2. st. po Kr.24 Tudi dru-god na območju Kranja se pojavljajo najdbe, ki sodijo v 4. st. po Kr. (fibula s čebuličastimi zaključki, prstan s ključem, profiliran bronast stilus, koščena vretenca).25 Pri lociranju antičnega grobišča se lahko opiramo le na skromne ostanke na območju Mayerjeve pristave in na med vojno izkopani rimski grob ob gradnji stavbe Stritarjeva 8 (sl. 2: 11,12). Posamezni rimski predmeti so bili najdeni tudi severno od tod.26 Grobni pridatki se niso ohranili, zato ne moremo ugotoviti, kateremu obdobju rimske poselitve bi to domnevno grobišče pripadalo. POZNOANTIČNA NASELBINA Poselitev se je intenzivirala v 5. in 6. st. Konec 5. ali v začetku 6. st. je bilo zgrajeno novo obzidje, katerega deli so bili odkriti na zahodnem robu Trubarjevega trga (sl. 4: 1,2), na severnem in južnem dvorišču gradu Kieselstein (Tomšičeva 44, sl. 4: 3,4) in na dvorišču stavbe Tomšičeva 38 (sl. 4: 5). Obzidje je bilo zgrajeno iz lomljencev in po­sameznih večjih rečnih oblic, široko zamazane fuge med kamenjem so splošna značilnost poznoantične zidave.27 V sredino 6. st. sodijo ostanki steklarske delavnice na se­vernem dvorišču gradu Kieselstein in večja količina stekla, predvsem steklenih čaš (sl. 4: 6),28 v drugi polovici 6. st. pa je nastal velik cerkveni kompleks z zgodnjekrščansko cerkvijo in krstilnico na lokaciji sedanje župne cerkve (sl. 4: 7).29 Posebnost naselja v Kranju so predvsem številne polzemljanke (sl. 4), ki so bile najdene po celotnem ob-močju starega mestnega jedra, od sredine Prešernove ulice na severu do Trubarjevega trga na jugu.30 Drobne najdbe iz tega obdobja predstavljajo predvsem odlomki grobe kuhinjske keramike, pa tudi uvoženega poznoantičnega posodja, zlasti amfor (Hayes 87c, 90b, 32/58, 99; amfore LRA 1, 2, 4; Keay 26F, 26G, 26H; poznorimske cilindrič­ 24 Sodeč po datacijah rimskih naselbinskih aglomeracij (Sagadin 1995, 13–22); za širše območje ustna potrditev Ples­ničar Gec (Mestni muzej Ljubljana). 25 Sagadin 2008, 61–78. 26 Valič 1960, 31–37. 27 Primerjave zlasti na poznoantičnih objektih na Ajdni nad Potoki in na Tonovcovem gradu pri Kobaridu. 28 Ostanki struktur so interpretirani kot steklarska de­lavnica na podlagi številnih odlomkov steklenih kozarcev, deponije apna, žganega iz školjk, ostankov pokvarjenih stek­lenih izdelkov in razlike med kuriščema, katerih eno pripada apnenici, za drugo pa še ni ugotovljenih primerjav (Sagadin 2004, 107–114). Drugačna interpretacija obe kurišči identi­ficira kot apnenici, spremljajoče stekleno gradivo pa naj bi dokazovalo deponijo, odpad ali reciklažo (Lazar, 2003, 216; Lazar 2018). 29 Sagadin 2017, 35–57. 30 Sagadin 2008, 145–146; Urankar, Bešter 2014; Urek, Rozman 2012. LATE ANTIQUE SETTLEMENT Settlement intensified in the 5th and 6th centuries. At the end of the 5th or the beginning of the 6th century, new town walls were built, the segments of which were discovered along the western edge of Trubarjev trg (Fig. 4: 1,2), in the north and south courtyards of Kieselstein Manor (Tomšičeva 44, Fig. 4: 3,4), and in the courtyard of the building at Tomšičeva 38 (Fig. 4: 5). The walls were made of quarried stones and individual large river cobbles. Wide mortar-filled gaps between the stones are a general characteristic of Late Roman construction.27 Remains of a glass workshop from the north courtyard of Kieselstein Manor and a number of glass artefacts, especially beakers (Fig. 4: 6), are dated to the middle of the 6th century.28 A large ecclesiastical complex with an Early Christian church and a baptistery at the location of the present-day parish church (Fig. 4: 7) was constructed in the second half of the 6th century.29 A specific feature of the Kranj settlement are numerous pit houses (Fig. 4), discovered all over the old town core, from the middle of Prešernova ulica in the north to Trubarjev trg in the south.30 Small finds from this period are predominantly fragments of coarse kitchen pottery, but also imported Late Antique vessels, especially amphorae (Hayes 87c, 90b, 32/58, 99; amphorae LRA 1, 2, 4; Keay 26F, 26G, 26H, late Roman cylindrical amphorae; a Hayes IIB oil lamp, etc.). Germanic stamped pottery and pottery with burnished decoration appear in the same contexts.31 When it comes to Late Antique metal finds, two lamellar armours and an angon from the courtyard of the building at Tomšičeva 38 stand out (Fig. 4: near the site 5).32 Of similar significance are the finds from the large cemetery at Lajh. The latter extended in the 6th and 7th centuries on the slopes and flat ground at the confluence of the Sava and the Kokra, and contained numerous finds from the Migration period (Fig. 4: 8).33 At the same time, another ecclesiastical centre with a cemetery was founded on the 27 The best analogies come from the buildings in the Late Antique fortified settlements of Ajdna nad Potoki and Tonovcov grad near Kobarid. 28 The interpretation of the remains as a glass workshop is based on numerous fragments of glass goblets, a depot of lime burned from shells, remains of broken glassware, and two different fireplaces: one is a lime kiln, while the analo­gies for the other one have not been found (Sagadin 2004, 107–114). A different interpretation identifies both fireplaces as lime kilns, and the corresponding glassware is supposed to be evidence of a depot, a junkyard, or a recycling centre (Lazar 2003, 216; Lazar 2018). 29 Sagadin 2017, 35–57. 30 Sagadin 2008, 145–146; Urankar, Bešter 2014; Urek, Rozman 2012. 31 Sagadin 2008. 32 Pflaum, Sagadin 2016. 33 Stare 1980; Knific 1995, 23–40; Knific, Lux 2005, 331– 343; Knific, Lux 2010, 26–36; Knific, Lux, 2015, 29–41. 5 6 grobiš~e / cemetery polzemljanka / pit-house obzidje / defence wall steklarska delavnica / glass production cerkev / church 0 100 m Sl. 4: Kranj. Poznoantična najdišča. M. = 1:5.000. Fig. 4: Kranj. Late Antique sites. Scale = 1:5,000. (Vir / Source: GURS©2004, TTN5) 1, 2 – Trubarjev trg (Pungart). 3, 4 – Tomšičeva 44 (severno in južno dvorišče gradu Kieselstein); 5 – Tomšičeva ulica 38; 6 – Tomšičeva 44 (severno dvorišče gradu Kieselstein); 7 – Župnijska cerkev sv. Kancijana; 8 – Lajh (Savska cesta in Sejmišče); 9 – Sv. Martin (lokacija nekdanje cerkve). ne amfore; oljenka Hayes IIB, itd.). V istih kontekstih se pojavlja tudi germanska glajena in žigosana keramika.31 Med poznoantičnimi kovinskimi najdbami po svojem pomenu izrazito izstopajo dva lamelna oklepa in ango z dvorišča stavbe Tomšičeva 38 (sl. 4: pri najdišču 5).32 Pomembnost te najdbe ustreza pomenu najdb z velikega grobišča v Lajhu, ki je v 6. in 7. st. nastalo na pobočjih in na ravnici ob sotočju Save in Kokre in ki mu dajejo pečat številne najdbe preseljevanja ljudstev (sl. 4: 8).33 Istočasno je na nasprotnem bregu Save nastalo še eno cerkveno središče z grobiščem (sl. 4: 9).34 V tem obdobju je Kranj predstavljal najpomembnejšo naselbino na območju današnje Slovenije. RAZVOJ NASELJA Za avgustejskodobno naselbino v Kranju je značil­na velika količina uvožene kakovostne keramike (sl. 5) iz severne Italije, pri čemer zlasti izstopa večje število čaš vrste Aco. Zelo malo pa je lokalne latenske keramike. Sestava drobnega gradiva sicer kaže priselitev iz Italije, vendar so bivalni objekti zelo skromni. Verjetno gre za načrtno koncentracijo nove oblasti v natanko določe­nem naselju. Zdi se, da je v zgodnji antiki Kranj vsaj nekaj časa predstavljal nekakšno trgovsko postojanko na poti iz Emone proti severu. Časovna pripadnost kranj­skih najdb omogoča tudi domnevo, da se je to zgodilo po aneksiji Noriškega kraljestva leta 15 pr. Kr., ko so bile zelo verjetno vzpostavljene nove trgovske povezave med Emono in na novo priključenimi ozemlji na severu. Ob-zidje s stolpi in pojav posameznih kosov vojaške opreme uvršča Kranj med tista naselja, ki obvladujejo prostor ter nadzorujejo prometne poti v zaledju vojaških spo­padov v osrednjem Podonavju.35 Očitno so bile manjše rimske vojaške posadke postavljene na vse strateško oz. prometno pomembne točke. Po končani pacifikaciji, ki je sledila generacijo kasneje, so ta naselja pogosto izgubila veljavo. Pomembnejše postanejo lokacije z izrazitejšim ekonomskim potencialom, kar v pogojih primarne proizvodnje pomeni bližino plodne zemlje in izrazit transportni potencial – bodisi za kopenski bodisi vodni promet. S tem lahko razlagamo popolno prenehanje življenja naselbine v Kranju od sredine 1. st. dalje. Kranj ima namreč izrazito naravno zavarovano, torej ugodno obrambno lego, vendar otežen dostop do vode in plodne zemlje. Zato se je težišče poselitve preneslo v številne vile rustike in manjša naselja ob robovih Kranjskega in Sorškega polja.36 31 Sagadin 2008, 95–107. 32 Pflaum, Sagadin 2016. 33 Stare 1980; Knific 1995, 23–40; Knific, Lux 2005, 331– 343; Knific, Lux 2010, 26–36; Knific, Lux, 2015, 29–41. 34 Sagadin 1987. 35 Horvat 2015, 171–208. 36 Sagadin 2008, 171–176. opposite bank of the Sava (Fig. 4: 9).34 During this period, Kranj was the most important settlement in the area of present-day Slovenia. SETTLEMENT DEVELOPMENT The Augustan-period settlement in Kranj is char­acterized by large amounts of imported pottery from northern Italy (Fig. 5), especially a large number of Aco beakers. There is very little local La Tene pottery. While the composition of small finds indicates Italian immigra­tion, dwellings are very modest. It seems that at least for a while during the Early Roman period, Kranj was some sort of a trade post on the route from Emona towards the north. The dating of the finds from Kranj allows the supposition that this happened after the annexation of the kingdom of Noricum in 15 BC, when new trade connections were most likely established between Emona and the newly annexed territories in the north. The town walls with towers and a small amount of military equip­ment indicate that Kranj was one of the settlements that controlled the territory and traffic routes in the hinterland of the armed conflicts in the Central Danube area.35 It appears that small Roman garrisons were stationed in all the locations that were significant from either a strategic or transport perspective. One generation later, when the pacification had been completed, the significance of these settlements slowly dwindled and settlements with higher economic potential gained importance. In the conditions of primary production, economic potential was the proximity of arable land as well as transport possibilities – for land or water traffic. This can explain the complete cessation of life in the Kranj settlement after the mid-1st century. While the position of Kranj is naturally well-protected, the access to water and arable land is inconvenient. Thus, the focus of settlement shifted 34 Sagadin 1987. 35 Horvat 2015, 171–208 Ponovna oživitev naselbine na konglomeratnem pomolu med Savo in Kokro od 4. st. dalje je verjetno posledica naraščajoče potrebe po varnosti, ki je postala pomembnejša v nemirnih časih druge polovice 3. st. Dobra naravna zavarovanost in strateška lega ob poti v severno Italijo preko Gornjesavske doline je zagotavljala izjemen pomen naselbini tudi v 5. in 6. st., ko postane Italija glavni cilj preseljujočim se ljudstvom.37 37 Sagadin 2008, 176–178. AIJ = V. Hoffiller, B. Saria, Antike Inschriften aus Jugoslavien 1: Noricum und Pannonia Superior, Zagreb 1938. CIL = Corpus inscriptionum Latinarum. Conspectus 1990 = Cospectus formarum terrae sigillatae Italico modo confectae. – Materialien zur Römisch-germanischen Keramik 10, Bonn 1990. GABROVEC, S. 1960, Mesto Kranja v prazgodovini slovenske­ga ozemlja. – V / In: 900 let Kranja, 11–30, Kranj. GABROVEC, S. 1966; Latensko obdobje na Gorenjskem (Die Latenezeit in Oberkrain). – Arheološki vestnik 17, 243–270. GLOBOČNIK, A. 1889, Archäeologische Karte von Krain. – Mittheilungen des Musealvereines für Krain 2. GUŠTIN, M. 2011, Carnium (Kranj, Slovenia): insediamento dei Carni. – V / In: S. Casini (ur. / ed.), “Il filo del tempo”. Studi di preistoria e protostoria in onore di Raffaele Carlo de Marinis, Notizie archeologiche Bergomensi 19, 447–458. HITZINGER, P. 1856, Kje je stal stari Santicum. – Kmetijske in rokodelske novice 14, 26. HORVAT, J. 1983; Prazgodovinske naselbinske najdbe pri farni cerkvi v Kranju (Vorgeschichtliche Siedlungsfunde bei der Pfarrkirche in Kranj). – Arheološki vestnik 34, 140–218. HORVAT, J. 2015, Early Roman military finds from prehistoric settlements in the Gorenjska region / Zgodnjerimske voja­ške najdbe s prazgodovinskih naselbin na Gorenjskem. – V / In: J. Istenič, B. Laharnar, J. Horvat (ur. / eds.), Evidence of the Roman army in Slovenia / Sledovi rimske vojske na Slovenskem, Katalogi in monografije 41, 171–208. KASTELIC, J. 1960, Staroslovanski Kranj. – V / In: 900 let Kranja, 38–50, Kranj. KNIFIC, T. 1971, Arheološka najdišča v Kranju. – Kronika 19, 70–76. KNIFIC, T. 1995, Vojščaki iz mesta Karnija. – Kranjski zbornik 1995, 23–40, Kranj. KNIFIC, T., J. LUX 2005, Gospe iz mesta Karnija. – Kranjski zbornik 2005, 331–343, Kranj. KNIFIC, T., J. LUX 2010, Otroci iz mesta Karnija. – Kranjski zbornik 2010, 26–36, Kranj. to numerous villae rusticae and smaller settlements along the edges of the plain of Kranj and Sora.36 The revival of the settlement on the conglomerate promontory between the Sava and the Kokra, which took place from the 4th century onwards, was most likely a consequence of an increasing need for security in the turbulent times of the second half of the 3rd century. The naturally well defended position and the strategic location on the route towards north Italy along the Upper Sava Valley ensured the significance of the settlement in the 5th and 6th centuries, when Italy became the main destination of the migrating peoples.37 Translation: Meta Osredkar 36 Sagadin 2008, 171–176. 37 Sagadin 2008, 176–178. KNIFIC, T., J. LUX 2015, Kranj z okolico v pozni antiki – zapis geografa iz Ravene in arheološki podatki. – Kranjski zbornik 2015, 29–41, Kranj. LAZAR, I. 2003, Rimsko steklo Slovenije / The Roman Glass of Slovenia. – Opera Instituti Archaeologici Sloveniae 7. LAZAR, I. 2018, Glass workshops in the Alpine and Eastern Adriatic area. – V / In: F. Wiesenberg, B. Birkenhagen (ur. / eds.), Roman glass furnaces – contexts, finds and reconstruc­tion synthesis, 31–52, Borg-Koeln. LINHART, A. T. 1788, Verzuch einer Geschichte von Krain und der übrigen südlichen Slaven Österreichs 1. – Laibach. LUX, J. 2011, Poročilo o izvedenih predhodnih arheoloških raziskavah na območju predvidene gradnje stanovanjskega objekta Zupanec v Kranju. – Poročilo / Report; ZVKDS OE Kranj (neobjavljeno / unpublished). MAURIG, J. 1893, Kranj. – Dom in svet 6, 463. MODRIJAN, Z. 1988; Prazgodovinska keramika iz farne cerkve v Kranju. – Seminarska naloga / Thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). MÜLLNER, A. 1879, Emona. Archäeologische Studien aus Krain. – Laibach. PEČNIK, J. 1893, Starinske najdbe pri mestu Kranju. – Izvestja Muzejskega društva za Kranjsko 3, 80. PEČNIK, J. 1894, 132. Die Grabungen in Krain während des Jahres 1893. – Mittheilungen der k. k. Central-Commision 20, 183–184. PEČNIK, J. 1904, Prazgodovinska gradišča na Kranjskem. – Izvestja Muzejskega društva za Kranjsko 14, 125–143. PFLAUM, V., M. SAGADIN 2016, V blesku kovinske oprave. Poznoantična lamelna oklepa iz Kranja (Gleaming in ar­mour. Late Antique lamellar armours from Kranj). – Kranj. PLETERSKI, A., B. ŠTULAR, M. BELAK (ur. / eds.) 2016, Izkopavanja srednjeveškega in zgodnjenovoveškega grobišča pri župni cerkvi v Kranju v letih 1964-1970 (Excavations of Medieval and Post-Medieval cemeteryat the Župna cerkev in Kranj from 1964 to 1970). – Opera Instituti Archaeologici Sloveniae 35, Grobišče Župna cerkev v Kranju 1, Ljubljana.PLETERSKI, A., B. ŠTULAR, M. BELAK (ur. / eds.) 2017, Arheološka raziskovanja grobov najdišča Župna cerkev v Kranju med letoma 1972 in 2010 (Archaeological investiga­tion of graves from Župna cerkev site in Kranj between 1972 and 2010). – Opera Instituti Archaeologici Sloveniae 37, Grobišče Župna cerkev v Kranju 2, Ljubljana. PLETERSKI, A., B. ŠTULAR, M. BELAK, H. BEŠTER (ur. / eds.) 2019, Začetek in konec raziskovanj grobišča pri Župni cerkvi v Kranju (1953 in 2011-2013) [The beginnning and end of exploration at the Župna cerkev cemetery in Kranj (1953 and 2011-2013)]. – Opera Instituti Archaeologici Sloveniae 38, Grobišče Župna cerkev v Kranju 3, Ljubljana. RADICS, P. von 1862, Archäologische Karte von Krain, 1862 (priloga k / appendix to: Geschichte Krains, Handbuch, 1. snopič / part 1, 1864). ROZMAN, B. 2004; Keramika iz prazgodovinske naselbine v Kranju (Pavšlarjeva hiša) (Pottery from the prehistoric settlement in Kranj (the Pavšlar house)). – Arheološki vestnik 55, 55–109. SAGADIN, M. 1987; Kranj – križišče Iskra. Nekropola iz časa preseljevanja ljudstev in staroslovanskega obdobja / Kranj – Iskra Crossroads. A Cemetery from the Migration Period and the Early Slavic Period. – Katalogi in monografije 24. SAGADIN, M. 1991, Najstarejša cerkvene stavba v Kranju. – V / In: Pod zvonom sv. Kancijana, 31–44, Kranj. SAGADIN, M. 1995, Poselitev rimskega podeželja na Gorenj­skem. – Kranjski zbornik 1995, 13–22, Kranj. SAGADIN, M. 2003, Zgodnjeantični Kranj (Ancient Kranj). – V / In: B. Jenčič et al. (ur. / eds.) Avguštinov zbornik. 50 let Gorenjskega muzeja, 71–81, Kranj. SAGADIN, M. 2004, Poznoantična steklarska delavnica (?) v Kranju [A Late Roman Glass workshop (?) in Kranj]. – V / In: I. Lazar (ur. / ed.), Drobci antičnega stekla (Fragments of ancient Glass), Annales Mediterranea, 107–114, Koper. SAGADIN, M. 2005, Nekatere neolitske najdbe z vzhodne Gorenjske (Some neolithic finds from the eastern Gorenj-ska). – V / In: M. Guštin (ur. / ed.), Prvi poljedelci. Savska skupina lengyelske kulture (First Farmers. The Sava Group of the Lengyel culture), Annales Mediterranea, 29–36, Koper. SAGADIN, M. 2008, Od Karnija do Kranja. Arheološki podatki o razvoju poselitve v antičnem in zgodnjesrednjeveškem obdobju. – Doktorska disertacija / PhD thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (unpublished / neobjavljeno). SAGADIN, M. 2010, Zgodnjeantično obzidje Karnija. – In / V: Jubilejni Kranjski zbornik 2010, 16–25, Kranj. SAGADIN, M. 2015, Evidence of the Roman army in Early Roman Kranj / Sledovi rimske vojske v zgodnjeantičnem Kranju. – V / In: J. Istenič, B. Laharnar, J. Horvat (ur. / eds.), Evidence of the Roman army in Slovenia / Sledovi rimske vojske na Slovenskem, Katalogi in monografije 41, 209–220. SAGADIN, M. 2017, Prva cerkev v luči arheoloških raziskav. – V / In: R. Peskar, M. Sagadin, A. Šebalj, Župnijska cerkev sv. Kancijana v Kranju. Njeno obličje in pomen, 35–55, Ljubljana. SCHINDLER-KAUDELKA, E. 1975, Die dünnwandige Gebrauchskeramik vom Magdalensberg. – Kärntner Mu-seumsschriften 58, Archäologische Forschungen zu den Grabungen auf dem Magdalensberg 3, Klagenfurt. SCHMID, W. 1908, Kleine Mitteilungen. – Carniola 1, 214–215. SCHMID, W. 1909, Kleine Mitteilungen. – Carniola 2, 154–155. STARE, F. 1954, Dva ilirska grobova iz Kranja. – Arheološki vestnik 5, 112–127. STARE, V. 1980; Kranj – nekropola iz časa preseljevanja ljudstev. - Katalogi in monografije 18. ŠAŠEL, J. 1970–1971, Alpes Iuliana. – Arheološki vestnik 21–22, 33–44. ŠKVOR JERNEJČIČ, B. 2017, Starejšeželeznodobne gomile na Gorenjskem. Žgani grobovi pri Vili Prah in na Koroški cesti v Kranju / Early Iron Age tumuli in the Gorenjska region. Cremation burials at Vila Prah and Koroška cesta in Kranj. – Arheološki vestnik 68, 117–196. TOMAŽINČIČ, Š. 2011, Poročilo o arheoloških raziskavah Jelenovega klanca na Ljubljanski cesti v Kranju. – Poročilo / Report; ZVKDS, OE Kranj (neobjavljeno / unpublished). URANKAR, R. 2011, Poročilo o arheoloških izkopavanjih na lokaciji Kranj – Kieselstein (severno dvorišče). – Poročilo / Report; ZVKDS, OE Kranj (neobjavljeno / unpublished). URANKAR, R., H. BEŠTER 2014, Arheološka izkopavanja v Kranju Mestno jedro. – Poročilo / Report; ZVKDS, OE Kranj (neobjavljeno / unpublished). UREK, M., L. ROZMAN 2012; Poročilo o zaščitnih arheolo­ških izkopavanjih na lokaciji Tomšičeve – Tavčarjeve (južni del). – Poročilo / Report; ZVKDS, OE Kranj (neobjavljeno / unpublished). UREK et al. 2016 = UREK, M., T. PODOBNIK, Š. TOMAŽIN­ČIČ, S. DJOKIĆ 2016, Končno strokovno poročilo o arheolo­ških raziskavah grobišča v Lajhu in prazgodovinske poselitve na območju Savske ceste in Sejmišča v Kranju. – Poročilo / Report; ZVKDS, OE Kranj (neobjavljeno / unpublished). VALIČ, A. 1960, Kranj z okolico v rimski dobi. – V / In: 900 let Kranja, 31–37, Kranj. VALIČ, A. 1975, Oris 20-letnih raziskovanj grobišča v Kra­nju. – V / In: Kranjski zbornik 1975, 159–167, Kranj. ŽONTAR, J. 1939, Zgodovina mesta Kranja. – Kranj. Milan SagadinŽupančičeva ulica 39 SI-4000 Kranj sagadin.milan@gmail.com Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 213–229 ŠMARTNO PRI CERKLJAH Špela TOMAŽINČIČ, Draško JOSIPOVIČ Izvleček Na območju prazgodovinske naselbine Straža je bila od sredine 1. st. pr. Kr. do zgodnjetiberijskega časa manjša ali občasna vojaška postojanka. Pozneje je bila v prostor današnje vasi Šmartno umeščena rimskodobna naselbina, na robu katere so bili odkriti tudi ostanki zgodnjekrščanske cerkve. Iz severozahoda je proti naselbini vodila tlakovana cesta, zgrajena v tiberijskem obdobju. Od nje se je odcepila pot do grobišča z 49 grobovi, datiranimi od sredine 1. do 4. st. po Kr. Prevladuje žgan pokop v preprostih grobnih jamah, izstopata dve zidani grobnici, odkrita je bila ustrina. Najmlajši grobovi so skeletni. Ob grobišču je stala stavba, ki jo vkopano kurišče, talilna peč v bližini ter večja količina železove žlindre opredeljujejo kot kovaško delavnico. Ključne besede: Italija (10. regija), Šmartno pri Cerkljah, rimska doba, naselbina, vojaška postojanka, zgodnjekrščanska cerkev, cesta, železarska delavnica, grobišče, grobnice, ustrina Abstract From the middle of the 1st century BC to the early Tiberian period, there was a small Roman military outpost in the area of the prehistoric settlement of Straža. Later, a Roman settlement was located in the area of today’s village of Šmartno, on the edge of which some remains of an early Christian church were discovered. A paved road that led towards the settle­ment from the northwest was constructed in the Tiberian period; from it, a trail split leading to the cemetery with 49 graves dating from the mid- 1st to the 4th century. They are predominantly simple cremation graves, but there are also two vaults and an ustrina. The latest graves are inhumations. A building with a hearth stood next to the cemetery. A melting furnace was located near-by; judging by large quantities of slag, this was a blacksmith’s workshop. Keywords: Italy (Regio X), Šmartno pri Cerkljah, Roman period, settlement, military outpost, early Christian church, road, iron workshop, cemetery, vault, ustrina https://doi.org/10.3986/9789610502586_12 Sl. 1: Šmartno pri Cerkljah. Arheološka najdišča. M. = 1:20.000. Fig. 1: Šmartno pri Cerkljah. Archaeological sites. Scale = 1:20,000.(Po / After: Horvat 2015, 173, sl. 2; Škerjanec 2016) 1 - Straža; 2 – Milharjev hrib; 3 – Pristava / “Vas brez ovir – Taber”; 4 – Pšata; 5 – Šmartno (območje vasi / village area); 6 – Sv. Martin; 7 – Cerkovnik; 8 – Brinovec; 9 – Straža – gomile / tumuli LEGA Med številnimi lokacijami stare poselitve na severo­vzhodnem robu Cerkljanskega polja, ob vznožju gričev­natih Tunjiških dobrav, ima posebno mesto ravninska vas Šmartno pri Cerkljah (373 m n. m.). Umeščena je pod hrib Stražo (492 m n. m.) in nižji Milharjev hrib in je za Poženikom tretja vas, ki leži ob potoku Pšata (sl. 1). Ta izvira v vasi Pšata in se po 28 km v vasi z enakim imenom izliva v Kamniško Bistrico. Lokacija iz krajine izstopa že zaradi spominske vrednosti svojega toponima, ki izvira od svetniškega patrocinija vaške podružnične cerkve sv. Martina.Toponim Šmartno in za starost indikativen patrocinij cerkve namreč kažeta zgodnji nastanek vasi in njene cerkve,1 zaradi česar predstavljata enega redkih prostor­skih elementov, ki se ni spremenil. Cerkev sv. Martina v Šmartnem je po izročilu prva in najstarejša cerkev v cerkljanski “fari”.2 Glede na pripoved, ki jo navaja Ivan Lavrenčič v svoji Zgodovini cerkljanske fare iz leta 1890, je postavitev cerkve sv. Martina, datira jo v 8. oziroma 1 Za zgodovino svetniških patrocinijev glej Höfler 2013, 373–376. 2 Lavrenčič 1890, 24–27; Kos 1960, 62–63; Höfler 2013, 237–238. POSITION Among many old settlements on the north-eastern edge of the Cerkljansko polje plain, at the foot of the hilly Tunjiške dobrave, an especially striking place isŠmartno pri Cerkljah (373 m above sea level). Located below the Straža hill (492 m asl) and the lower Milharjev hrib hill, it is the third village after Poženik to lie on the Pšata stream (Fig. 1), which springs in the village of Pšata and flows after 28 km into the Kamniška Bistrica River in the village of the same name. If not for anything else, the location is outstanding for its toponym, which has its origin in the patronage of the St Martin parish church. The toponym Šmartno and the patronage of the parish church indicate their early origin,1 which is why they are to be understood as one of the rare spatial elements that have not changed. Ac­cording to tradition, the St Martin church in Šmartno is the first and earliest church in the Cerklje parish.2 According to the narrative quoted by Ivan Lavrenčič in his Zgodovina cerkljanske fare from 1890, the construc­tion of the St Martin church, dating back to the 8th or 9th 1 For the history of saint patrocinia see Höfler 2013, 373–376. 2 Lavrenčič 1890, 24–27; Kos 1960, 62–63; Höfler 2013, 237–238. 9. stoletje,3 analogna vzpostavitvi krščanstva na tem prostoru. Po tej pripovedi je dal cerkev sv. Martinav Šmartnem za svoje podložnike sezidati tamkajšnji graščak Brnikar v neposredni bližini svojega gradu, ki je stal na Milharjevem hribu pri vasi Poženik, pod njim pa so bile na severozahodu lepe in prostorne pristave, po katerih se travniki na tem prostoru še danes imenujejo Pristava. Pripoved Brnikarjevo postavitev cerkve “na čast svetemu Martinu” neposredno povezuje s priho­dom oglejskih misijonarjev, ki so s svojim obiskom pri graščaku vplivali na njegovo sprejetje krščanstva in na širjenje nove vere med njegove poganske podložnike. V na videz preprosti pripovedi se skriva vzorec cerkvene organizacije v zgodnjem srednjem veku, ki je slonela na lastniški cerkvi, ki jo je na svoji zemlji za svojo in dušno oskrbo svojih podanikov postavil zemljiški gospod in jo tudi oskrbel s premoženjem in z duhovnikom.4 Tudi kasnejši zgodovinski in pripovedni viri pri­trjujejo, da je bil v času pred Cerkljami (prvi župnik v Cerkljah, Richerus plebanus de Sancta Maria, kjer je bil takrat že sedež župnije, se omenja po letu 1147 in pred1154) sedež “fare” v vasi Šmartno in da je bil sv. Martin prvotna farna cerkev. Cerklje so se prvotno imenovale Trnovlje, sosedje so kraj imenovali Pri Trnovljanih, ljudi pa Trnovljane, ker so prebivali v okolišu poraslem s tr-njem. Šele po postavitvi cerkve sv. Marije so ljudi v naselju ob cerkvi začeli imenovati Cerkevljane ali Cerkljane, od koder izvira današnje ime Cerklje.5 Tudi odprava “fare” vŠmartnem in njen prenos v Cerklje, ki so prevzele vlogo novega cerkvenega središča, je, podobno kot njeni začetki, zabeležen v pripovedi o zakleti graščakinji iz “poženškega gradu”; ta dogodek je namreč plemenito gospo tako zelo razžalostil in užalil, da je najprej zahtevala, da ji pot do Trnovelj posujejo s škrlatom, kasneje pa je kljub temu užaljena pribila, da rajši “zleze v kačo”, kakor da bi šla v Trnovlje. In tako se je tudi zgodilo.6 ZGODOVINA RAZISKAV Starost cerkve v Šmartnem potrjuje arheologija. Lavrenčičeva Zgodovina navaja, da je “pred dobrimi desetimi leti izkopal tamošnji posestnik Blaž Hubad na svojem sadnem vrtu uprav tikoma cerkve mnogo 7 mrtvaških kostij, med njimi lobanjo”. Arheološke raziskave so se že na začetku osredo­točile na pojmovanje tega prostora, kot se kaže skozi omenjene pripovedi in izročilo. Podpovršinski pregled, ki ga je leta 1968 izvedel Andrej Valič, Gorenjski muzej Kranj, je bil tako najprej usmerjen na dve lokaciji, po­vezani s to spominsko tradicijo, na cerkev sv. Martina 3 Lavrenčič 1890, 26. 4 Höfler 2016, 64–90 (posebej za Šmartno: 69 in 71). 5 Kos 1960, 62–63; Močnik 2004, 31–33. 6 Lavrenčič 1890, 3–4. 7 Ib., 57. century,3 is relatable to the establishment of Christianity in this area. According to this narrative, the St Martin church in Šmartno was commissioned to be built by Lord Brnikar for his subjects in the immediate vicin­ity of his castle on the Milharjev hrib near the village of Poženik, northwest of which there was a beautiful and spacious estate (Slov. pristava); the meadows in this area are still called Pristava. The story of Brnikar’s construction of the church “in honour of St Martin” is related to the arrival of missionaries from Aquileia who influenced the lord’s accepting of Christianity and the spreading of a new faith among his pagan subjects. In the seemingly simple narrative, there is a pattern of church organisation in the Early Middle Ages, which was based on a church built and provided for by a landlord on his land, for his own spiritual care and that of his subjects.4 Later historical and literary sources confirm that before Cerklje (the first priest in Cerklje, Richerus plebanus de Sancta Maria, where the parish was then established, is mentioned after 1147 and before 1154) theparish seat was in the village of Šmartno, St Martin being the original parish church. Cerklje was originally called Trnovlje, because the surroundings were full of thorny bushes. Only after the establishment of the Church of St Mary did people in the settlement near the church begin to be referred to as Cerkevljani or Cerkljani, from which the present name “Cerklje” originates.5 The abolition of the parish in Šmartno and its transfer to Cerklje is also recorded in the narrative about an enchanted lady of the castle at Poženik; grieved by the disbanding of the parish, the noble lady first insisted that the road to Trnovlje be strewn with purple for her, but later swore that she would rather be transformed into a snake than go to Trnovlje. And so it all happened.6 RESEARCH HISTORY The age of the church in Šmartno is confirmed by archaeology. Lavrenčič’s Zgodovina cerkljanske fare states that “over ten years ago, the owner of the estate, Blaž Hubad, excavated in his fruit garden right near the 7 church many human bones with a skull among them”. Archaeological research has focused from the very beginning on the concept of this space, as it is presented itself through the above narratives and traditions. The sub-surface survey, carried out by Andrej Valič (Gore-njski muzej, Kranj) in 1968, was thus primarily focused on two locations connected with this memorial tradi­ 3 Lavrenčič 1890, 26. 4 Höfler 2016, 64–90 (esp. for Šmartno: 69 and 71). 5 Kos 1960, 62–63; Močnik 2004, 31–33. 6 Lavrenčič 1890, 3–4. 7 Ib., 57. (sl. 1: 6)8 in na Milharjev hrib (tudi Taber ali Brnikarjev grad; sl. 1: 2).9 Arheološka sondiranja leta 1968 okoli cerkve (sl. 1: 6) so potrdila obstoj grobišča in določila njegovo starost. Na vzhodni strani proti jugu orientirane cerkve sv. Martina je bila odkrita grobnica s petnajstimi skeletnimi grobovi, ki jih pridatki datirajo v 10. in 11. st. Prav tako je bil s sondiranji leta 1968 na nasprotni, za­hodni strani cerkve dokumentiran delni tloris starejše, verjetno že poznoantične sakralne arhitekture, segajoče pod današnjo cerkev.10 Ta najstarejša cerkev je očitno kasneje predstavljala materialni pomnik, na katerega sta se navezala zgodnjesrednjeveška cerkev in grobišče. Leta 1985 so bili v sadovnjaku za hišo Šmartno 13 ob terenskem ogledu (Milan Sagadin, ZVKDS, OE Kranj) pri izkopu jarka za telekomunikacijske vode odkriti ostanki antičnega zidu in rimskega gradbenega materiala (sl. 1: 5).11 Druga terenska raziskava, vodil jo je Draško Josi­povič, je bila zaradi načrtovanega obsežnega gradbenega posega (“Vas brez ovir – Taber”) izvedena leta 2006 na Pristavi (nekdanje Premuzarjevo posestvo) ob jugoza­hodnem vznožju Straže in tudi na nasprotnem, desnem bregu Pšate (sl. 1: 3,4).12 Po izkopavanjih 2006 sta bili na tem prostoru opravljeni še dve arheološki dokumentira­nji ob gradnji, prvo leta 2010 ob gradnji kanalizacije na trasi Pšata–Zalog, odsek Poženik–Šmartno,13 in drugo ob gradnji mostu, dovozne ceste in izkopu temeljev za objekte Doma Taber v letih 2011 do 2013 (sl. 1: 3,4).14 Skupaj so med letoma 2006 in 2013 odkrili ostanke prazgodovinske poselitve (na obeh bregovih Pšate), na levem bregu rimsko cesto in pot, ob poti grobišče in na njegovem severnem robu še poznorimsko stavbo, na desnem bregu pa zgodnjesrednjeveško poselitev. PRAZGODOVINA Najstarejši sledovi poselitve so bili odkriti na desnem bregu Pšate (sl. 1: 4). Pod naselbinskimi depoziti, ki so nastali v zgodnjesrednjeveškem obdobju,15 so bili različni vkopi, stojke, nekatere obložene s kamenjem, in sledovi kurišč, ki predstavljajo ostanke preprostih stavb. Najdbe 8 Okolico cerkve sv. Martina je Valič sondiral na več mestih. Glej njegov načrt v Valič 1970–1971, 276, sl. 1. Obja­ve raziskave: Valič 1968–1969b; 1969; 1970, 189; 1970–1971, 275–287; 1982, XXI. 9 Valič 1968–1969a; 1970, 185; 1982, XXI. 10 Valič 1968–1969b; 1969; 1970–1971, 275–287; 1982, XXI. 11 Sagadin 1986, 274–276. 12 Leta 2005 je bil najprej opravljen podpovršinski pregled s testnimi sondami (Josipovič 2005), naslednje leto so sledilazaščitna izkopavanja (Žižek, Tomažinčič 2007). 13 Josipovič, Rupnik 2010. 14 Josipovič, Rupnik 2013. 15 Odkriti so bili ostanki vsaj ene zgodnjesrednjeveške stavbe. Glej Žižek, Tomažinčič 2007, 2–4. tion, i.e., the church of St Martin (Fig. 1: 6)8 and the Milharjev hrib hill (also Taber, Brnikarjev grad; Fig. 1: 2).9 Archaeological trenching in 1968 around the church confirmed the existence of a cemetery and determined its chronology. On the eastern side of the St Martin church (Fig. 1: 6), which was orientated to the south, a tomb was discovered with fifteen inhumations dating back to the 10th and 11th centuries. On the west side of the church, the foundations of an earlier, probably Late Antique church were located by trenching. It extended below today’s church.10 This earliest church apparently later represented a material memorial to which the early medieval church and the cemetery were connected. In 1985, the remains of a Roman wall and construc­tion material were discovered during an archaeological survey when excavating a telecommunication ditch in the orchard behind the house Šmartno 13 (Fig. 1: 5).11 Another field research (“Vas brez ovir - Taber”), led by Draško Josipovič, was carried out in 2006 on the former Premuzar estate at the southwestern foot of the Straža hill and on the right bank of the Pšata (Pristava; Fig. 1: 3,4).12 After the excavations in 2006, another two archaeological documenting projects were carried out, the first in 2010 during the construction of the sewage system on the Pšata–Zalog road, the Poženik–Šmartno section,13 and the second during the construction of the bridge, access road, and foundation digging for the Dom Taber in the years 2011 to 2013 (Fig. 1: 3,4).14 Thus, between the years 2006 and 2013, the prehistoric settle­ment on both banks of the Pšata was discovered as well as the Roman settlement on the left bank: a road and a path, a cemetery, and a Late Roman building, and the Early Medieval settlement on the right bank. PREHISTORY The earliest settlement traces were discovered in the right bank of the Pšata (Fig. 1: 4). Under the settle­ment deposits from the early medieval period,15 there were different pits, post-holes (some lined with stones) and traces of fireplaces, all the remains of simple build­ 8 The surroundings of the church were examined by Valič in several places. Cf. his plan in Valič 1970–1971, 276, fig. 1. Research publishing: Valič 1968–1969b; 1969; 1970, 189; 1970–1971, 275–287; 1982, XXI. 9 Valič 1968–1969a; 1970, 185; 1982, XXI. 10 Valič 1968–1969b; 1969; 1970–1971, 275–287; 1982, XXI. 11 Sagadin 1986, 274–276. 12 In 2005, the first sub-surface surveys were conducted with the help of test trenches (Josipovič 2005), followed by excavations in the ensuing year (Žižek, Tomažinčič 2007). 13 Josipovič, Rupnik 2010. 14 Josipovič, Rupnik 2013. 15 Remains of at least one early medieval building were discovered. Cf. Žižek, Tomažinčič 2007, 2–4. iz teh struktur so izključno keramične in so datirane v srednjo bronasto dobo. Na levem bregu Pšate, med poto­kom in ob njem umeščeno rimsko cesto, so bili prav tako odkriti posamezni odlomki keramike iz srednje bronaste dobe, sicer pa so najznačilnejše prazgodovinske strukture na prostoru nekdanje pristave in Milharjevega hriba žgani grobovi (sl. 1: 2,3).16 Pridatki, med njimi je za datacijo pomembna predvsem igla s stožčasto glavico, grobove datirajo v iztek pozne bronaste dobe oziroma na začetek železne dobe.17 Pripadajoča prazgodovinska naselbina je bila na hribu Straža (sl. 1: 1), kjer so bile dokumentirane obrambne in naselbinske strukture ter najdbe iz starejše in mlajše železne dobe.18 Naselbini na Straži je pripadalo tudi manjše gomilno grobišče, ki leži ob severovzhodnem vznožju (sl. 1: 9) severnega, nekoliko nižjega vrha Straže.19 RIMSKA DOBA Predmeti, odkriti na Straži, dokazujejo, da je bilo območje prazgodovinske naselbine od sredine 1. st. pr. Kr. do zgodnjega tiberijskega obdobja verjetno uporab­ljeno za manjšo vojaško postojanko (sl. 1: 1).20 Sledovi poselitve, ki so bili odkriti v vasi Šmart­no in njeni neposredni okolici, omogočajo približno rekonstrukcijo prostorske in organizacijske zasnove te podeželske aglomeracije v rimski dobi. Bistveno je namreč, da so bili do zdaj odkriti vsi njeni osnovni strukturni elementi: naselbinski, ki so bili evidentirani na prostoru današnje vasi (sl. 1: 5); daljši odsek vicinalne ceste in grobišče, na levem bregu Pšate, na območju nekdanje pristave Brnikarjevega gradu (oziroma na jugozahodnem vznožju Straže; sl. 1: 3); ter na severnem robu grobišča nekoliko mlajši suhozidni temelji stavb iz poznorimskega obdobja, ki jih povezujemo s predelo­valno ali proizvodno dejavnostjo (sl. 1: 3). CESTI Trasa vicinalne ceste, ki je bila raziskana v dolžini skoraj 50 m, v grobem sledi potoku Pšata in poteka v smeri severo–severozahod od Poženika proti jugo–jugovzhoduv smeri Šmartna (sl. 1: 3; 2; 3). Za gradnjo ceste, ta je bila v povprečju široka med 2,2 do 3,6 m, je bil uporabljen iz­ 16 Žgan grob, v katerem sta bili dve fragmentarno ohran­jeni fibuli očalarki in odlomek keramične žare, je na vrhu Milharjevega hriba pri sondiranju odkril že Valič (Valič 1968–1969a; 1970, 185; 1982, XXI). 17 Glej Žižek, Tomažinčič 2007. 18 Prazgodovinsko naselbino na Straži pod imenom “gra­dišče pri Poženku” omenja že Josip Žontar 1939, 19 in 21, sl. III. Glej tudi Horvat 2015, 173–184 in Škerjanec 2016, 24–27.19 Škerjanec 2016, 26–27. 20 Za interpretacijo vloge Straže v rimskem in v pozno­ rimskem obdobju, od 3. st. do začetka 5. st., glej Horvat 2015, 183–184. ings. Only ceramics were found there, and it is dated to the Middle Bronze Age. On the left bank of the Pšata, between the creek and the adjacent Roman road, Mid­dle Bronze Age ceramics were also discovered. The cremation graves were located on Milharjev hrib (Fig. 1: 2,3).16 The grave goods, the most important being a pin with a cone head, date the graves to the Late Bronze or Early Iron Age.17 The prehistoric settlement was on the Straža hill (Fig. 1: 1), where defensive and settlement structures, as well as small finds from the Early and Late Iron Age, were discovered.18 A small tumulus cemetery north of the Straža hill belonged to the same prehistoric settlement (Fig. 1: 9).19 ROMAN PERIOD Roman military finds are the evidence that the area of prehistoric settlement on Straža hill was used for a small military post from the middle of the 1st century BC until the early Tiberian period (Fig. 1: 1).20 Archaeological traces in the village of Šmartno and its immediate surroundings enable an approximate insight into the settlement organisation in the Roman period. It is crucial that many of its basic structural ele­ments have been discovered thus far; settlement remains in the area of today’s village (Fig. 1: 5), a long section of the vicinal road, a small cemetery and foundations of late Roman buildings associated with production activity, all at the southwestern foothill of Straža hill (Fig. 1: 3). ROADS The route of the vicinal road, which was researched in the length of almost 50 m, follows the Pšata riverbed and runs north-northwest from Poženik towards the south-southeast in the direction of Šmartno (Figs. 1: 3; 2; 3). It was on average between 2.2 and 3.6 m wide and constructed only of local materials: limestone and dolomite pebbles from the Pšata and pieces of sandstone and carbonate conglomerates from Straža. Spacings between up to 10 cm large pavement stones were filled with small, sharp white gravel, also from the Pšata. The 16 A cremation grave, in which two fragmented spectacle fibulae and a fragment of a ceramic jar were located, was dis­covered during probing at the top of the Milharjev hrib by Valič. Valič 1968–1969a; 1970, 185; 1982, XXI. 17 Cf. Žižek, Tomažinčič 2007. 18 The prehistoric settlement on Straža hill named “the Poženk hillfort” is mention also by Josip Žontar 1939, 19 and 21, fig. III. Cf. also Horvat 2015, 174–184 and Škerjanec 2016, 24–27. 19 Škerjanec 2016, 26–27. 20 For the interpretation of the role of Straža hill in the Roman and late Roman period from the 3rd to the 5th century, cf. Horvat 2015, 183–184. 2 5 25 23 zid / wall pe~ / furnace grob / grave grobnica / vault ustrina cesta / road izkopno polje / excavation yeld 0 10 20 m Sl. 2: Šmartno pri Cerkljah, Pristava. Del izkopnega polja iz leta 2006 (“Vas brez ovir – Taber”; sl. 1: 3). M. = 1:500. Fig. 2: Šmartno pri Cerkljah, Pristava. Part of the excavation field 2006 (“Vas brez ovir – Taber”; Fig. 1: 3). Scale = 1:500. Sl. 3: Šmartno pri Cerkljah, Pristava. Odsek rimske ceste obPšati med Poženikom in Šmartnim. Izkopavanja 2006 (sl. 1: 3). Pogled proti jugovzhodu. Fig. 3: Šmartno pri Cerkljah, Pristava. Roman road along Pšatabetween Poženik and Šmartno. Excavations 2006 (Fig. 1: 3). View towards southeast. (Foto / Photo: J. Rupnik) ključno lokalni material, apnenčasti in dolomitni prodniki iz Pšate ter kosi peščenjaka in konglomerata karbonatne sestave, ki so bili prineseni s Straže. Prostor med tlakova­nimi kamni, velikimi do 10 cm, je bil zapolnjen z manjšim belim ostrorobim prodom, ki prav tako izvira iz Pšate. Cesta ni imela posebne podlage, temveč je bila postavljena na droben, dobro nosilen prod, ki ga je verjetno nasul vodotok izpod Straže, preden je dosegel Pšato (sl. 3).21 Najdbe, ki so bile dokumentirane na cesti ali neposredno ob njej, kažejo, da so cesto v takšni zasnovi uporabljali vsaj od tiberijskega časa dalje in je bila verjetno takrat tudi zgrajena. Na cesti je bil najden as cesarja Tiberija (Divus Augustus, 22–30), še en Tiberijev as (34–37) je bil odkrit v sedimentu za utrditev cestišča.22 Kasnejšo uporabo iste komunikacije med drugim dokazujeta bronasta spona, okrašena z dvema konjičkoma, značilna za konec 6. in 21 Glej Verbič 2006, 6,8. 22 Vse numizmatične najdbe, s ceste in iz grobov, je do-ločil Andrej Šemrov (Numizmatični kabinet Narodnega mu-zeja Slovenije). Sl. 4: Šmartno pri Cerkljah, Pristava. Odsek rimske pokopališke poti. Izkopavanja 2006 (sl. 1: 3). Pogled proti vzhodu. Fig. 4: Šmartno pri Cerkljah, Pristava. Cemetery road. Exca­vations 2006 (Fig. 1: 3). View towards east. (Foto / Photo: J. Rupnik) road had no particular foundation; it was constructed on top of small, resistant gravel, which had been probably deposited by water flow from underneath Straža before it reached the Pšata (Fig. 3).21 According to the small finds the road was used at least from the Tiberian time when it was probably also constructed. An Emperor Tiberius as was found on the road (Divus Augustus, 22–30); an­other Tiberius’ as (34–37) was discovered in the road reinforcement sediment.22 Later use of this same road is indicated by the bronze buckle, decorated with two horses, typical of the late 6th and early 7th centuries, and a cast bronze enamelled crescent-shaped earring dating to the 10th or early 11th centuries. A narrower and worse constructed road split off the main road and climbed steadily towards the south–west slopes of Straža. It was researched in the length of 36 m. A small cemetery extended along it (Fig. 2; 4). 21 Cf. Verbič 2006, 6,8. 22 Andrej Šemrov, Narodni muzej Slovenije, determined all the coins. začetek 7. st., ter bronast ulit in emajliran polmesečasti uhan, datiran v 10. ali na začetek 11. st. Od te ceste se je odcepila nekoliko ožja in slabše grajena cesta, ki se je v smeri zahod–vzhod, v raziskani dolžini 36 m, zložno vzpenjala proti pobočjem Straže. Nastanek te stranske ceste povezujemo z grobiščem, ki se je v osrednjem delu navezalo nanjo (sl. 2; 4). GROBIŠČE Večina od skupno 49 grobov (sl. 2),23 je ležala neposredno ob južnem robu stranske ceste, tako da so posamezni grobovi v razporeditvi dosledno sledili poti. Na grobišču so prevladovali žgani grobovi, skeletnih grobov je bilo le šest (grobovi 4, 23, 42–45) in ti so bili tudi najbolj oddaljeni od poti. Severno od poti so ležali le pravokotna grobnica, dva žgana grobova in ustrina, vsi precej oddaljeni drug od drugega in od poti. Prevladujoča oblika pokopa so bile preproste, okrogle (grobovi 1, 3, 6, 7, 11–16, 18, 19, 24, 27, 28, 30, 33–35, 39–41 in 46) ali ovalne grobne jame (grobovi 10, 20, 21, 29, 47), manjše število grobov je imelo pravo­kotno (grobovi 22, 26, 32, 38) ali kvadratno oblikovano grobno jamo (grobovi 17, 25, 31, 36, 37, 48 in 49). V grobne jame so bili poleg sežganih ostankov pokojnikov položeni predvsem keramični pridatki. V tej skupini po pridatkih izstopa grob 34 (sl. 5). Na dnu okrogle grobne jame sta bili na žganino pokojnika položeni pečatni oljenki tipa Loeschcke Xa s pečatom CRESCE in Loeschcke Xb s pečatom FORTIS. Poleg oljenk so bili v grobu še enoročajni vrč, prostoročno izdelan in na vretenu dodelan lonček s poševno metličeno površino na trebuhu in zapestnica iz železne žice (sl. 5: 3) s pomičnim zapiralom (tip 3.26.3 po Rihi), kakršne so bile v rabi skozi celotno rimsko obdobje.24 Po pečatnih oljen­kah Loeschcke Xa in Xb (sl. 5: 1,2) grob datiramo v 2. st. Manjše število grobov je imelo nekoliko kom­pleksnejšo zasnovo. Tako lahko v primeru petih grobov (5, 31, 32, 36 in 48) glede na ostanke kovinskih okovov ali železnih klinov, njihovo lego v grobu in razporeditev grobnih pridatkov sklepamo na pokope v lesenih skrin­jah. V eni od grobnih jam (grob 5) so na primer ležali štirje 20 cm dolgi železni klini, ki so s svojo lego določali obliko in velikost (0,6 × 0,5 m) lesene skrinje, v katero so bili položeni prostoročno izdelan lonec, na vretenu izdelan lonček, odlomki steklene čaše, jantarni obesek v obliki Heraklovega kija in as cesarja Marka Avrelija ali Komoda (161–192). Zanimivi so trije grobovi (25, 26, 37), ki po svoji ar­hitekturi in uporabljenem gradbenem materialu imitirajo domačo hišo. Grob 25, v katerem je bil poleg odlomkov vrča, sigilate in stekla ter železnega klina najden tudi as 23 Na sl. 2 je prikazanih 40 grobov. Na načrtu ni more-bitnih pokopov, ki so sicer všteti v skupno število. 24 Riha 1990, 62. CEMETERY Most of the total of 49 graves (Fig. 2)23 lay directly along the southern edge of the side road, consistently fol­lowing its course. They were predominantly cremation graves, only six of them were inhumations (Graves 4, 23, 42–45), which were also furthest from this cemetery road. Only a rectangular vault, two graves and an ustrina were situated north of the road and quite distant of it. Most of the burials had simple, round (graves 1, 3, 6, 7, 11–16, 18, 19, 24, 27, 28, 30, 33–35, 39–41 and 46) or oval grave pits (graves 10, 20, 21, 29, 47), a smaller number of graves had rectangular (graves 22, 26, 32, 38) or a square-shaped grave pits (graves 17, 25, 31, 36, 37, 48 and 49). Mostly ceramic artefacts were deposited in the graves in addition to the burnt remains. Grave 34 is outstanding in this group (Fig. 5). On the bottom of the round grave pit, two oil lamps were found, one Loeschcke Xa type with the stamp CRESCE and one Loeschcke Xb type with the stamp FORTIS. Fur­thermore, there were a single-handed jug, a hand-made pot that was finished on a potters’ wheel and bearing combed decoration, and an iron-wire bracelet with an adjustable clasp (type 3.26.3 according to Riha), quite common throughout the entire Roman period (Fig. 5: 3).24 Judging by the Loeschcke Xa and Xb oil lamps, the tomb is dated to the 2nd century (Fig. 5: 1,2). A small number of graves were of a somewhat more complex structure. In the case of five graves (5, 31, 32, 36 and 48), the remains of metal fittings or iron spikes specifically positioned in the grave and the arrangement of grave goods led us to believe they were burials in wooden chests. The position of four 20-cm iron spikes in the Grave 5 determined the shape and size (0.6 x 0.5 m) of the wooden chest, in which a hand-made pot and a pot made on potters’ wheel were laid, fragments of a glass cup, an amber pendant in the shape of Hercules’ club and an as of Emperor Marcus Aurelius or Com-modus (161–192). Particularly interesting are three graves (25, 26, 37), which architecture and materials imitate a home. A rectangular Grave 25 that contained a jug, some si­gillata, glass, an iron spike and an as of Antoninus Pius (Diva Faustina I, from 141) had four large stones along the walls that supported the covering of two tegulae. In the middle of the grave pit, under the ridge of this sym­bolic house, there was a small rock which symbolised the home hearth (Fig. 6). In this manner, the image of the grave as an eternal home of the deceased (domus aeterna) or his soul was manifested in the architectural design of the grave. Two masonry tombs are particularly striking in their special architectural design. The first tomb (Fig. 2) 23 Fig. 2 shows 40 graves. Structures, which could be po­tential graves and were also numbered, are not plotted. 24 Riha 1990, 62. Sl. 5: Šmartno pri Cerkljah, Pristava. Grob 34. 1,2,4,5 keramika; 3 železo. M. 1–3 = 1:2; 4,5 = 1:3. Fig. 5: Šmartno pri Cerkljah, Pristava. Grave 34. 1,2,4,5 ceramics; 3 iron. Scale 1–3 = 1:2; 4,5 = 1:3. Antonina Pija (Diva Faustina I., od 141), je imel ob strani­cah kvadratne grobne jame postavljene štiri večje kamne, ki so predstavljali nosilno konstrukcijo za dve teguli. Na sredini grobne jame, domnevno pod slemenom imagi­narne hiške, je bil postavljen manjši kamen, ki bi lahko simbolično nadomeščal domače ognjišče (sl. 6). Tako bi lahko bila predstava o grobu kot posmrtnem bivališču, večnem domu pokojnika (domus aeterna) oziroma domu njegove duše izražena tudi v arhitekturni zasnovi groba. Med grobovi po svoji reprezentativnosti nedvomno izstopata zidani grobnici različne arhitekturne zasnove. Prva grobnica (sl. 2) je bila kvadratne zasnove, velikosti 4,1 × 4,1 m, kar kažejo ohranjeni suhozidni temelji. Približno na sredini grobnice sta bila vkopana dva žgana was square, measuring 4.1 x 4.1 m, as suggested by the preserved drywall foundations. In the middle of the tomb, there were two cremation grave pits (8 and 9) of the same design and containing identical grave goods (Figs. 2; 7). A hand-made pithos was placed in both graves; the one in Grave 8 was decorated on its entire surface with diagonal combed decoration (Fig. 8: 4), while the one in Grave 9 was without decoration (Fig. 9:5). Two glass jars of Isings 67a type were used as urns; they were placed into the pithoi; the jar in Grave 8 was placed on a standing surface, while the one in Grave 9 was laid on the side (Figs. 8: 2; 9: 3). Next to the north edge of each grave pit, one single-handed jug with a spherical-biconical body without a funnel was Sl. 6: Šmartno pri Cerkljah, Pristava. Grob 25. a – dno groba; b – pokriti grob. Fig. 6: Šmartno pri Cerkljah, Pristava. Grave 25. a – bottom of the grave; b – covered grave. (Foto / Photo: J. Rupnik). Sl. 7: Šmartno pri Cerkljah, Pristava. Grobova 8 in 9 v zidani grobnici. Pogled proti jugovzhodu. Fig. 7: Šmartno pri Cerkljah, Pristava. Graves 8 and 9 in the vault. View towards southeast. (Foto / Photo: J. Rupnik). grobova (8 in 9), ki sta imela enako zasnovo in sta vse­bovala tudi identične grobne pridatke (sl. 2; 7). V vsako grobno jamo je bil najprej položen prostoročno izdelan pitos, v grob 8 okrašen po celotni površini s poševnim metličenjem (sl. 8: 4), v grob 9 brez okrasa (sl. 9: 5). Za žari sta bila uporabljena steklena lonca kroglaste oblike tipa Isings 67a, ki sta bila položena v pitos, žara v grobu 8 na stojno ploskev, žara v grobu 9 na bok (sl. 8: 2; 9: 3). Ob severni rob vsake grobne jame sta bila na enak način postavljena po en enoročajni vrč s kroglasto-bikoničnim trupom brez izliva (sl. 8: 3; 9: 4). V stekleni žari iz groba 8 je bila bronasta fibula tipa Almgren 68. Lok fibule je okra-šen s profiliranim gumbom, trapezasto oblikovana noga placed in the same manner (Figs. 8: 3; 9: 4). In the glass urn from Grave 8, a bronze fibula of type Almgren 68 was placed. The arch of the fibula was decorated with a profiled knob; the catch-plate is perforated and ends with a flat-faced knob (Fig. 8: 1). A silver fibula was found in the adjacent Grave 9. Even though the arch of the fibula was deformed, the remainder of the applique is still visible on its lower part. Shaped as a seated bird or duck, the applique is a characteristic addition to the fibulae with animal heads instead of knobs on the arch, which perforated triangular leg does not end with a knob but with an anthropomorphic head (Fig. 9: 1).25 25 Vinko Šribar (1968, 450) thought this version of the Sl. 8: Šmartno pri Cerkljah, Pristava. Grob 8 iz kvadratne grobnice. 1 bron; 2 steklo; 3,4 keramika. M. 1,2 = 1:2; 3,4 = 1:3. Fig. 8: Šmartno pri Cerkljah, Pristava. Grave 8 from the rectangular vault. 1 bronze; 2 glass; 3,4 ceramics. Scale 1,2 = 1:2; 3,4 = 1:3. Sl. 9: Šmartno pri Cerkljah, Pristava. Grob 9 iz kvadratne grobnice. 1,2 srebro; 3 steklo; 4,5 keramika. M. 1,2 = 1:2; 3–5 =1:3. Fig. 9: Šmartno pri Cerkljah, Pristava. Grave 9 from the rectangular vault. 1,2 silver; 3 glass; 4,5 ceramics. Scale 1,2 = 1:2; 3–5 =1:3. ni predrta in se zaključuje z ravno usmerjenim gumbom (sl. 8: 1). V sosednjem grobu 9 je bila najdena srebrna fibu­la (sl. 9: 1). Čeprav je lok fibule deformiran, je na njegovem spodnjem delu še vedno viden ostanek aplike. Ta aplika, v obliki sedeče ptice oziroma račke, je značilen dodatek na fibulah z živalsko glavo na loku namesto gumba, katerih polna trikotna noga predrta z luknjicami se prav tako ne zaključuje z gumbom, temveč z antropomorfno glavico.25 V žari je bil poleg srebrne fibule tudi srebrn obesek v obliki Heraklovega kija, izdelan iz pločevine, na katero je bil apliciran okras v obliki valovnice in vertikalnih kapljic (sl. 9: 2). Pridatki in žari datirajo dvojni pokop v grobnici s kvadratno zasnovo v čas vlade cesarja Klavdija (41–54) oziroma njegovega naslednika cesarja Nerona (54–68). V drugem primeru (grob 2; sl. 2), ki leži južno od poti, lahko na podlagi ohranjenih arhitekturnih elementov, zlasti kamnitega venca z zunanjim preme­ 25 Vinko Šribar (1968, 450) je to različico fibule Almgren 68 označil za lokalno posebnost in jo opredelil za “emonski tip fibule z živalsko glavo na loku”. Pavlin, Dular 2007, 71–73. Furthermore, a silver pendant shaped like Hercules’ club was found there; it was made of sheet metal with applied decoration in form of waves and vertical droplets (Fig. 9: 2). According to the urns and the grave goods, the double-grave tomb is dated to the period of Emperor Claudius (41–54) or his successor Nero (54–68). A special tomb (Grave 2) is located south of the road (Fig. 2). On the basis of the surviving architectural elements, particularly the stone foundation with the diameter of 1.85 m, it can be concluded that there was a simple, domed tomb built with mortar and featuring a white-coated floor. In the eastern part of the tomb, there were two round-shaped grave pits, one without grave goods while the other contained some shards of ceramic vessels and an oil lamp as well as one bronze coin of Constans or Constantius II (348–351). Almgren 68 fibula to be a local specialty, characterising it as an “Emona type fibula with an animal head on the arch”. Pav­lin, Dular 2007, 71–73. rom 1,85 m, sklepamo na preprosto, z malto zidano kupolasto grobnico, ki je imela dno prevlečeno z belim estrihom. V vzhodnem delu grobnice sta bili vkopani okrogli grobni jami. Ena je bila brez pridatkov, v drugi je bilo nekaj odlomkov keramičnih posod in oljenke ter bronast novec Konstansa ali Konstancija II. (348–351). H grobišču sodi tudi ustrina, ki je bila odkrita ob njegovem severozahodnem robu (sl. 2). Za prostor sežiga je bila izbrana manjša, približno 4,5 × 3,1 m velika na­ravna kotanja, ki je bila do 40 cm na debelo zapolnjena z žganinskim sedimentom. Ta je poleg večje količine oglja, žganinskih ostankov ter posameznih koščkov ožganih živalskih in človeških kosti vsebovala tudi več odlomkov rimske keramike. Med skeletnimi grobovi po številu in sestavi pri­datkov izstopa grob 23 (sl. 10). V grobu so bili na enem mestu skupaj položeni vsi pridatki: oljenka brez pečata različice Loeschcke Xb/c, enoročajna čaša z vodoravni-mi žlebovi na vratu in trupu, lonček z rahlo izvihanim ustjem, prav tako na vratu in trupu okrašen s plitvimi vodoravnimi žlebovi, železen nož s širokim nastavkom za držaj in srebrn prstan z ovalnim nastavkom za kamen ali steklen vložek (tip 2.1.2 po Riha), ki pa se ni ohranil (sl. 10: 1). Takšni prstani so se nosili od 1. st. naprej in vsaj do druge polovice 3. st.26 Enoročajna čaša (sl. 10: 4) sodi po obliki, okrasu ter svetli in prečiščeni keramiki v posebno skupino, ki je zelo številna na Hrušici in je datirana v poznorimsko obdobje.27 Način pokopa in pridatki grob datirajo v konec 3. ali v 4. st. Grobni pridatki – med njimi šest novcev, ki so bili z izjemo novca iz kupolaste grobnice vsi kovani po letu 141 – datirajo grobove najprej v klavdijsko obdobje, večina pokopov sodi v 2. st., pokop v kupolasti grobnici (grob 2) pa kaže, da so na tem grobišču očitno pokopa­vali še sredi 4. st. NASELBINA Rimska naselbina, ki ji je pripadalo opisano gro­bišče, je glede na sicer skromne sledove verjetno stala na prostoru današnje vasi Šmartno (sl. 1: 5). To lokacijo za zdaj potrjujejo ostanki antičnega zidu in rimskega gradbenega materiala v središču vasi (Šmartno 13).28 Na pobočje Straže, natančneje na severni rob gro­bišča, sta bila v poznorimskem obdobju sicer umeščena vsaj dva sočasna lesena objekta s kamnitimi temelji, ki pa nista bila bivalnega značaja (sl. 1: 3; 2). Od večjega objek­ta so bili odkriti suhozidni temelji pravokotnega tlorisa velikosti 13,2 × 10,7 m, ki je bil po sredini predeljen. Suhozidni temelji stavbe so bili deloma postavljeni na temelje starejšega objekta, grajenega v podobni tehniki. Severovzhodno in jugozahodno od pravokotne stavbe 26 Riha 1990, 30. 27 Giesler 1981, 90–91, t. 40. 28 Sagadin 1986. Part of the cemetery was also the ustrina discovered in its north-western edge (Fig. 2). A small, round-like natural depression measuring approx. 4.5 x 3.1 m was selected for the incineration site. It was filled about 40 cm deep with burnt residues, which contained large amounts of charcoal, burnt pieces of animal and human bones, and several fragments of Roman ceramics. Inhumation Grave 23 (Fig. 10) stands out with respect to grave goods. All the artefacts were placed together: a Loeschcke Xb/c oil lamp, a single-handed cup with horizontal grooves on its neck and body, a pot with a slightly everted rim and decorated with shallow horizontal grooves, an iron knife with a wide handle base and a silver ring with an oval base for a stone or glass in­sert, which has not survived (Fig. 10: 1). Such rings, type 2.1.2 according to Riha, were worn from the 1st at least until the second half of the 3rd century.26 With regard to its shape, decoration and bright, refined ceramics, the single-handled cup belongs to a select group, which is mainly present in Hrušica and dates to the late Roman period (Fig. 10: 4).27 The burial custom and the grave goods date Grave 23 to the late 3rd or early 4th centuries. Judging by the grave goods, among them six coins, which were all forged after the year 141 (with the excep­tion of the coin from the masonry tomb), the earliest burials date to the period of Emperor Claudius, most of the burials belong to the 2nd century, while the burial in the dome vault (Grave 2) indicates that this site was apparently still in use for burials in the mid-4th century. SETTLEMENT Judging from the scarce evidence, the Roman settlement, to which the above-described cemetery belonged, was most likely in the area of present-dayŠmartno village (Fig. 1: 5). The location is supported by the remains of Roman walls and construction material from the village centre (Šmartno 13).28 Two non-residential wooden buildings were con­structed on the slopes of Straža hill, along the northern edge of the cemetery in the late Roman period (Figs. 1: 3; 2). The bigger of the two buildings had drywall founda­tions measuring 13.2 x 10.7 m and was divided into two rooms. It was placed on top of foundations of an earlier, similarly built building. Northeast and southwest of the building, two corners of similarly constructed drywall foundations were discovered, which probably belonged to a similar wooden building. Judging by an oval-shaped hearth or a furnace in the eastern part of the southern room, as well as the large quantity of slag along the western exterior, this was most likely a smithy. The semi­ 26 Riha 1990, 30. 27 Giesler 1981, 90–91, pl. 40. 28 Sagadin 1986. sta bila ohranjena še dva, na enak način grajena vogala suhozidnih temeljev, ki verjetno prav tako predstavljata ostanke lesenih objektov. Namembnost stavbe, najver­jetneje je bila kovaška delavnica, opredeljuje kurišče ozi­roma peč ovalne oblike, ki je bila vkopana v zahodnem delu južnega prostora, in večja količina železove žlindre, ki je bila najdena ob zahodnem zunanjem robu objekta. Z železarsko dejavnostjo je povezana tudi polkrožno oblikovana talilna peč, ki je bila odkrita približno 20 m južno od objekta. Če sledimo Valičevi interpretaciji arhitekturnih ostankov, ki so bili odkriti ob cerkvi sv. Martina (sl. 1: 6),29 je bila verjetno ob naselbini že konec 4. stoletja zgra­jena manjša cerkev, opremljena z mozaičnim tlakom. FUNKCIJA IN STATUS NASELJA TER RAZVOJ Za rimsko Šmartno se ni ohranil noben epigrafski spomenik in zelo verjetno se nanj ne nanaša noben zgodovinski vir,30 ki bi izpričeval status te naselbine ali kakršenkoli drug podatek o njegovih prebivalcih ali naselbini sami. Pred popolno anonimnostjo rimsko preteklost Šmartna rešuje zgolj arheologija. Podatki iz preteklih arheoloških raziskav namreč omogočajo, da približno rekonstruiramo prostorsko in organizacijsko zasnovo te podeželske aglomeracije ter sklepamo o njeni vpetosti v širši regionalni okvir. Manjše, lepo zasnovano grobišče, ki je bilo umeščeno na jugozahodno pobočje Straže, do njega pa je vodila pokopališka cesta, ki se je odcepila od glavne vicinalne ceste, gotovo pripada manj­ši naselbini, ki je verjetno stala na prostoru današnje vasiŠmartno. Posmrtna bivališča, ki so jih ustvarili prebivalci te naselbine, z izborom pridatkov za posamezne grobove ter pokopavanjem Romano ritu, v skladu z antičnimi predstavami o prehajanju v onstranstvo in življenjem po smrti nedvoumno kažejo, da je bila tu naselbina rimske skupnosti oziroma skupnosti romaniziranega prebivalstva.31 Za naselbino je bilo pomembno, da je bila umeščena ob domnevno vicinalno cesto, ki je iz Emone preko Save vodila v Gameljne, Šmartno pod Šmarno goro, Skaručno, Polje, Utik, Vodice in naprej protiLahovčam, v Šmartno pri Cerkljah, Šenčur in Karnij (Carnium),32 s čimer je bila vključena v mrežo rimskih trgovskih poti. Če sodimo po pridatkih iz grobov, je ta 29 Valič 1970–1971, 275–277 in sl. 1: zid B. 30 V mislih imava sicer poznoantični vir, Kozmografijo (Cosmografia 4,21) anonimnega geografa iz Ravene (okrog 700), ki na ozemlju, imenovanem patria Carneola, našteva 25 naselbin, ki razen ene, imenovane Carnium, niso identifi­cirane. O viru Wolff 2000, 97–101; Bratož 2014, 387–391 in nazadnje Knific, Lux 2015, 29–41. 31 O možnem statusu naselbine, njegovi ekonomski in prostorski vlogi glej Sagadin 2008, 174–175.32 Za traso te ceste glej Truhlar 1975, 101. round melting furnace discovered some 20 m south of the building is another evidence for the ironworks. According to Valič’s interpretation of architectural remains, discovered next to the church of St Martin (Fig. 1: 6), a small church with a mosaic floor was built next to the settlement as early as the end of the 4th century. 29 FUNCTION AND STATUS OF THE SETTLEMENT AND ITS DEVELOPMENT No epigraphic monument has been preservedfor the Roman settlement of Šmartno, and it is very likely that no historical source30 refers to its status, its inhabitants or the settlement itself; only archaeology is saving it from complete anonymity. Archaeological data allow us to roughly reconstruct its structure and its role in the broader regional framework. The small cemetery located on the southwest slopes of Straža and the pertaining cemetery road, which split from the main route, must have belonged to a small settlement, which probably stood in the area of present-day village of Šmartno. With their grave goods and the burials Romano ritu, the post-mortem dwellings created by the inhabitants are congruent to the ancient concepts of life after life, clearly indicating that this was a settlement of a Roman community or a community of Romanised people.31 It was important for the settlement that it was located along the presumed vicinal road that led from Emona across Sava to Gameljne, Šmartno podŠmarno goro, Skaručna, Polje, Utik, Vodice and further towards Lahovče, Šmartno pri Cerkljah, Šenčur and Carnium,32 thus being incorporated into the network of Roman trade routes. Judging by the grave goods, this Roman community probably came to be in the second half of the 1st century and certainly in the 2nd century, persisting until at least the mid-4th century. The com­munication leading past the settlement remained in use in late Antiquity and in the Early Middle Ages. During this time, a new cemetery, built on a new location near the late Roman church, was connected to the former Roman settlement, and a new settlement, also built in a completely new location, on the right bank of the Pšata. 29 Valič 1970–71, 275–277; fig. 1: wall B. 30 Particularly the late ancient Cosmografia (4,21) by an anonymous geographer from Ravenna (around 700), who listed 25 settlements in the area dubbed patria Carneola; all but Carnium are unidentified. On this source cf. Wolff 2000, 97–101; Bratož 2014, 387–391 and Knific, Lux 2015, 29–41. 31 About the possible status of the settlement, its eco­nomic and spatial role cf. Sagadin 2008, 174–175.32 For the route cf. Truhlar 1975, 101. rimska skupnost verjetno zaživela v drugi polovici 1. st., zagotovo v 2. st., in je vztrajala vsaj do sredine 4. st. Ko­munikacija, ki je vodila mimo naselbine, je ostala živa še v pozni antiki in zgodnjem srednjem veku. V tem času sta se na nekdanjo rimsko naselbino navezala grobišče, ki je nastalo na novi lokaciji ob poznoantični cerkvi, in naselbina, ki je bila prav tako umeščena na povsem novo lokacijo, na desni breg Pšate. Zahvale Vsem, ki so nama pomagali pri nastanku članka, se iskreno zahvaljujeva. Vereni Vidrih Perko in Zdenki Kra-mar iz Gorenjskega muzeja, ker sta prijazno pomagali pri vsem potrebnem za izris grobnih celot, in Marku Zoroviću, ki je prispeval risbe predmetov. Hvala Janezu Rupniku za pomoč pri izdelavi načrta rimskega Šmartnega. Zahvalju­jeva se tudi Andreju Šemrovu iz Numizmatičnega kabineta Narodnega muzeja Slovenije, ki je opredelil vse novce. In ne nazadnje, iskrena hvala Jožetu Štuklu iz Loškega muzeja za njegov trud pri tipološki opredelitvi fibul. BRATOŽ, R. 2014, Med Italijo in Ilirikom. Slovenski prostor in njegovo sosedstvo v pozni antiki (Between Italy and Illyricum. Slovene territory and its neighbourhood in late Antiquity). – Zbirka Zgodovinskega časopisa 46. HÖFLER, J. 2013, O prvih cerkvah in župnijah na Slovenskem. K razvoju cerkvene teritorialne organizacije slovenskih dežel v srednjem veku. – Ljubljana. HÖFLER, J. 2016, Lastniške cerkve zgodnjega srednjega veka na Gorenjskem in njihovo zgodovinsko ozadje. – Varstvo spomenikov 49, 64–90. GIESLER, U. 1981, Die Kleinfunde. – V / In: T. Ulbert (ur. / ed.), Ad Pirum (Hrušica), Münchner Beiträge zur Vor- und Frühgeschichte 31, 53–127. HORVAT, J. 2015, Zgodnjerimske vojaške najdbe s prazgodo­vinskih naselbin na Gorenjskem / Early Roman military finds from prehistoric settlements in the Gorenjska re­gion. – V / In: J. Istenič, B. Laharnar, J. Horvat (ur. / eds.), Evidence of the Roman army in Slovenia / Sledovi rimske vojske v Sloveniji, Katalogi in monografije 41, 171–208. JOSIPOVIČ, D. 2005. Poročilo o predhodnih arheološkihsondiranjih v Šmartnem pri Cerkljah (Naselje brez ovir – Taber). – Poročilo / Report; Arhiv ZVKDS, OE Kranj (neobjavljeno / unpublished). JOSIPOVIČ, D., J. RUPNIK 2010, Poročilo o izvedbi arheo­loškega dokumentiranja ob gradnji kanalizacije na trasi Pšata – Zalog, odsek Poženik – Šmartno. – Poročilo / Report; Arhiv ZVKDS, OE Kranj (neobjavljeno / unpublished). JOSIPOVIČ, D., J. RUPNIK 2013, Poročilo o izvedbi arheološke­ga nadzora v Šmartnem pri Cerkljah (Vas brez ovir – Taber) v letih 2011 – 2013. – Poročilo / Report; Arhiv ZVKDS, OE Kranj (neobjavljeno / unpublished). Acknowledgements We would like to extend our thanks to everybody who contributed to this article: to Verena Vidrih Perko and Zdenka Kramar from the Gorenjski muzej, because they kindly helped in everything necessary for the drawings; to Marko Zorović who contributed the drawings of the artefacts; to Janez Rupnik for his help in the drafting of the Roman Šmartno plan. We would like to thank Andrej Šemrov from the Numismatic Cabinet of the National Mu­seum of Slovenia who classified all coins. Last but not least, we would like to thank Jože Štukl from the Loški muzej for his efforts in typologically recognising all the fibulae. Translation: Gregor Pobežin KNIFIC, T., LUX, J. 2015, Kranj z okolico v pozni antiki – zapis geografa iz Ravene in arheološki podatki. – Kranjski zbornik 2015, 29–41. KOS, M. 1960, Starejša naselitev na Kranjski ravnini. – V / In: J. Žontar (ur. / ed.), 900 let Kranja, 51–73, Kranj. LAVRENČIČ, I. 1890, Zgodovina cerkljanske fare. – Ljubljana. MOČNIK, J. 2004, Podobe nekdanjih časov. Ob 850. obletnici prve pisne omembe cerkljanske fare. – Cerklje. PAVLIN, P., J. DULAR 2007, Prazgodovinska višinska naselja v Posavskem hribovju (Prehistoric hilltop settlements in the Posavje Hills). – Arheološki vestnik 58, 65–120. RIHA, E. 1990, Die römische Schmuck aus Augst und Kaiser-augst. – Forschungen in Augst 10.SAGADIN, M. 1986, Šmartno pri Cerkljah. – Varstvo spome­nikov 28, 274–276. SAGADIN, M. 2008, Od Karnija do Kranja. Arheološki podatki o razvoju poselitve v antičnem in zgodnjesrednjeveškem obdobju. – Doktorska disertacija / PhD thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (unpublished / neobjavljeno). ŠKERJANEC, L. 2016, Uporaba državnih lidarskih podatkov varheologiji – primer Šmartno pri Cerkljah. – Profil. Revija študentskega arheološkega društva 10, 19–28. ŠRIBAR, V. 1968, K absolutni kronologiji najdb iz zgodnje Emone. – Arheološki vestnik 19, 445–452. TRUHLAR, F. 1975, Stara pota ter poskus rekonstrukcije nekdanje prometne mreže. – V / In: Arheološka najdišča Slovenije, 99–104, Ljubljana. VALIČ, A. 1968–1969a, Šmartno pri Cerkljah. – Varstvo spo­menikov 13–14, 148–149. VALIČ, A. 1968–1969b, Šmartno pri Cerkljah. – Varstvo spomenikov 13–14, 176. VALIČ, A. 1969, Šmartno pri Cerkljah na Gorenjskem – sta­roslovanska grobnica in kasnoantična arhitektura. – Arhe­ološki pregled 11, 213. VALIČ, A. 1970, Arheološke najdbe v Kranju in okolici od leta 1960 do 1970. – V / In: Kranjski zbornik 1970, 185–191, Kranj. VALIČ, A. 1970–1971, Šmartno pri Cerkljah na Gorenjskem. Poznoantične in zgodnjesrednjeveške najdbe (Šmartno bei Cerklje in Gorenjsko (Oberkrain). Spätantike und frühmittelalterliche Funde). – Arheološki vestnik 21–22, 275–287. VALIČ, A. 1982, Arheološka raziskovanja na kranjskem ob-močju v preteklih štirih desetletjih. – V / In: J. Žontar (ur. / ed.), Zgodovina mesta Kranja, V–XXIV, Kranj. VERBIČ, T. 2006, Poročilo o geološkem ogledu arheoloških izko­pavanj v Šmartnem pri Cerkljah na Gorenjskem. – Poročilo / Report; Arhiv Gorenjskega Muzeja, Kranj (neobjavljeno / unpublished). WOLFF, H. 2000, Die Frage der Besiedlung des heutigen Slowenien im Lichte des Anonymus von Ravenna. – V / In: Bratož, R. (ur. / ed.), Slovenija in sosednje dežele med antiko in karolinško dobo. Začetki slovenske etnogeneze / Slowenien und die Nachbarla¨nder zwischen Antike und karolingischer Epoche. Anfa¨nge der slowenischen Ethnogenese 1, Situla 39 (Razprave 1. razr. SAZU 18), 97–101. ŽIŽEK, T., Š. TOMAŽINČIČ 2007, Poročilo o zaščitnih arheoloških izkopavanjih. Šmartno pri Cerkljah, Vas brez ovir – Taber. – Poročilo / Report; Arhiv ZVKDS, OE Kranj (neobjavljeno / unpublished). ŽONTAR, J. 1939, Zgodovina mesta Kranja. – Kranj. Špela Tomažinčič Univerza v Ljubljani Filozofska fakulteta Oddelek za klasično filologijo Aškerčeva 2 SI-1000 Ljubljana spela.tomazincic@ff.uni-lj.si Draško Josipovič Oldhamska cesta 8 SI-4000 Kranj Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 231–247 BLAGOVICA Ana PLESTENJAK Izvleček Članek obravnava ostanke antične naselbine z grobiščem, postavljene ob vznožje severnega pobočja doline Črni graben, na vzhodnem delu današnje Blagovice. Naselbino z velikopotezno zasnovo so najkasneje v 1. st. n. št. postavili Rimljani. Vzdrževali so jo vsaj do sredine 4. st. n. št. Zaradi plazovitosti terena so bile stavbe večkrat obnovljene in celo popolnoma na novo postavljene, a vselej upoštevajoč prvotno urbanistično zasnovo. V 2. st. je na vzhodnem delu nastalo tudi pripadajoče grobišče. Avtorica predpostavlja, da gre za ostanke rimske poštne postaje. Na osnovi več indicev jo povezuje s postajo Ad publicanos, ki jo omenja Tabula Peutingeriana. Ključne besede: Italija (10. regija), Blagovica, Ad Publicanos, rimska doba, naselbina, poštna postaja, grobišče Abstract This paper addresses the remains of a Roman settlement with a burial ground located at the foot of the northern slope of the Črni graben valley in the eastern part of today’s Blagovica. The settlement with various types of large edifices was built no later than the 1st century AD; it remained in use at least until the mid-4th century AD. Since the terrain was prone to landslides, the buildings had to be repeatedly renovated or even wholly rebuilt, each time according to the original urban design. The cemetery was established in the 2nd century at the eastern side of the settlement. It is assumed that the settlement is the remnants of the Roman postal station. Based on several indications, it might be the Ad publicanos station, mentioned by the Tabula Peutingeriana. Keywords: Italy (Regio X), Blagovica, Ad Publicanos, Roman period, settlement, postal station, cemetery https://doi.org/10.3986/9789610502586_13 Sl. 1: Blagovica. Lokacija rimske naselbine na topografski karti. M. = 1:50.000. Fig. 1: Blagovica. The location of the Roman settlement on the topographic map. Scale 1:50,000. (Vir / Source: https://www.openstreetmap.org/#map=14/46.1853/14.8624&layers=CN) LEGA Antična naselbina je ležala vzhodno od središča današnje vasi Blagovica in severno od današnje ceste proti Trojanam (sl. 1). Tu ob vznožju severnega pobočja doline Črni graben leži aluvialen vršaj (sl. 2; 3). Prepo­znamo ga kot ozek pas pretežno izravnanega zemljišča, ki je blago dvignjeno nad dnom doline in rahlo pada proti jugu, jugozahodu in jugovzhodu, torej proti reki Radomlji. Ta poplavna reka je v osnovi oblikovala dolino in na njenem dnu še danes odlaga svoje nanose. Podlaga, na kateri leži najdišče, je torej aluvialnega izvora. Ker pa je pobočje nad njim zelo vodonosno in že ob manjšem deževju nastajajo številni manjši vodotoki, ki tečejo po hribu ter spirajo zemljino, je na samo najdišče v pretek­losti plazilo tudi več sedimentov s pobočja. ZGODOVINA RAZISKAV Že od leta 1879 se omenjajo najdbe rimskih gro­bov, kamnitih blokov in posameznih najdb na ledini, poimenovani Ajdovski ali Poganski britof (Stražar 1985, 75–83 s starejšo literaturo). Najdišče je bilo prvič razi­skovano leta 1977, ko so ob gradnji plinovoda naleteli na arheološke ostanke in jih pripisali rimskodobnemu grobišču (Sagadin 1977; Zupančič 1979). Najdena sta bila dva večja postamenta, ki so ju pripisali grobnicama, več zidov in njihovih ruševin ter šest poznoantičnih skeletnih grobov. Sagadin (1977) navaja, da je grobišče tudi zgodnjesrednjeveško (slovansko). POSITION The ancient settlement lay east of the centre of today's village of Blagovica and north of today's road towards Trojane (Fig. 1). Here, at the foot of the northern slope of the Črni graben valley, lies an alluvial fan (Figs. 2; 3), recognizable as a narrow belt of predominantly levelled land, which is slightly raised above the bottom of the valley and slightly slanting to the south, south-west and south-east, i.e., towards the Radomlja River, which, with its flooding regime, basically shaped the valley, still laying its deposits on its bottom. The foundation on which the site lies is thus of alluvial origin; however, since the slope above it is very aquiferous, meaning that even small precipitation sets off several smaller watercourses, which wash off the soil running down the slopes. Several sediments from the slopes were deposited over the site in the past. RESEARCH HISTORY The earliest references to the discovery of Roman graves, stone blocks, and individual finds in the fallow called Ajdovski britof or Poganski britof date to as early as 1879 (Stražar 1985, 75-83 with earlier references). The site was first investigated in 1977, when the construction of a gas pipeline revealed archaeological remains attributed to the Roman cemetery (Sagadin 1977; Zupančič 1979). Two large pedestals were found and attributed to the tombs, several walls and ruins thereof, and six late Roman Sl. 2: Blagovica. Lokacija rimske naselbine na lidarskem posnetku. M. = 1:10.000. Fig. 2: Blagovica. The location of the Roman settlement on lidar. Scale 1:10,000. (Vir / Source: http://gis.arso.gov.si/evode/profile.aspx?id=atlas_voda_Lidar@Arso) Sl. 3: Blagovica. Pogled na izkopavanja leta 2013 proti severo­vzhodu. (Foto: Jašar Skorupan) Fig. 3: Blagovica. The view of excavations in 2013 towards northeast. (Photo: Jašar Skorupan) Resnični pomen najdišča je bil ugotovljen šele l. 2013 ob gradnji druge cevi plinovoda, ki je tekel severno od prve cevi. Izsledki teh raziskav so pokazali, da je na ravnici pod severnim pobočjem doline Črni graben ležala rim-skodobna naselbina z grobiščem (Plestenjak et al. 2014). NASELBINA Obseg naselbine pri Blagovici ni natančno določen. Raziskan je bil le 15 m širok pas v dolžini 300 m, kar ponuja omejen vpogled v najdišče (sl. 2). Znan in zame­inhumation graves. According to Sagadin, the cemetery is also of early medieval (Slavic) origin (Sagadin 1977). The real significance of the site was only realised in 2013 during the construction of the second gas pipeline running north of the first one. The results of this investiga­tion showed that on the plain below the northern slopes of the Črni graben valley there was a Roman settlement with a cemetery (Plestenjak et al. 2014). SETTLEMENT The whole settlement area cannot be precisely speci­fied; only the 15 m wide zone was explored in the length of 300 m, which provides limited insight into the site (Fig. 2). Only the eastern edge of the settlement is well-known. In the northern part, the settlement probably bordered on the slope of the valley, while it possibly bordered the Radomlja River to the south, which used to flow further to the northern side of the valley than today. It is not known how far the settlement extended towards the west. It was presumably confined to the alluvial fan, other areas being less suitable for settlement. It can be surmised from the findings that the settlement was founded by the Romans no later than the 1st century AD. As early as the first phase, the basic layout of the settlement was put into place and remained unaltered until its abandonment in the 4th century. Its main feature was a series of large buildings, their longer sides facing north and south. This orientation changed only slightly over time, indicating their probable position along the road running on the east-west direc­ Sl. 4: Blagovica. Tloris prve faze poselitve. Fig. 4: Blagovica. Ground plan of settlement Phase 1. jen je le vzhodni rob naselja. Na severnem delu je bila verjetno meja pobočje doline, na južnem delu pa reka Radomlja, ki je nekdaj tekla bolj po severni strani doline kot danes (med izkopavanji je bila domnevna nekdanja struga odkrita le nekaj metrov južno od stavb; prim. sl. 4; 6; 8; 12). Kako daleč je poselitev segala proti zahodu, ni znano. Domnevno je bila naselbina vezana na aluvialni vršaj, saj so bili ostali predeli za poselitev manj primerni. Izsledki kažejo, da so naselbino najkasneje v 1. st. n. št. zasnovali Rimljani. Že v prvi fazi poselitve je bila začrtana osnovna stavbna zasnova, ki jo je naselbina ohranjala vse do opustitve v 4. st. Glavna značilnost so velike stavbe, ki so se nizale druga ob drugi. Z daljšima stranicama so bile obrnjene proti severu oz. jugu. Ta usmeritev se je skozi čas le minimalno spreminjala, kar nakazuje lego ob cestni povezavi, usmerjeni vzhod–za-hod. Cesta ni bila odkrita, vendar domnevamo, da je ležala severno od raziskanega območja. Kljub ohranjanju osnovne zasnove se je naselbina skozi čas nenehno spreminjala. Glavni vzrok za to so naravni procesi. Zaradi velike vodonosnosti pobočja nad najdiščem ob večjem deževju nastajajo hudourniki, ki so si (do regulacije v sedemdesetih letih 20. stoletja) pot v dolino utirali po površju in s seboj prenašali ma­terial. Prvi poskusi regulacije so bili izvedeni že takoj ob rimski naselitvi območja, ko so naravna korita skušali zamejiti z zidci (sl. 4; 8; 12). Vendar pa pri tem niso bili najuspešnejši. Struge so se nenehno premikale. Prebivalcem so preglavice povzročala tudi pogo-sta plazenja zemljine s pobočja. Nekateri zdrsi so tako močno poškodovali objekte, da so se ti podrli. Zato je naselbina doživela številne pozidave in prezidave, večkrat pa tudi celovite prenove. Ob tem so graditelji preizkušali tudi različne tehnike gradnje, s katerimi so bolj ali manj učinkovito kljubovali naravnim procesom. Prav te razmere, ki so bile za prebivalce izrazito neugodne, so omogočile stratigrafsko lepo berljiv arhe­ološki zapis. Na podlagi odlaganja plazin smo dogajanje lahko umestili v 4 glavne kronološke faze, ki jih lahko le okvirno časovno opredelimo, saj so se zaradi plazenja starejše najdbe mešale z mlajšimi. FAZA 1 Najstarejši ostanki so vezani na zahodni del razi­skanega območja (sl. 4). Tu so bili najdeni številni jarki in jame, ki jim težko pripisujemo funkcijo. Izpostaviti velja le večjo naravno jamo (SE 3415 = 3432), globoko 1,2 m, ki je bila zapolnjena z ostanki kurjenja (plasti bogate z ogljem in ožgano zemljo). Ti ostanki nakazujejo, da so v bližini delovale peči ali podobni objekti. V spodnjih polnilih ni bilo najdb. Radiokarbonska analiza oglja enega izmed njih je pokazala datacijo 2045 ± 30 BP,1 1 Poz-63337, 95 cal. BC – 2 cal. AD (kalibrirana starost 1-.: (68,2 % ver.)) oziroma 165 cal. BC – 24 cal. AD (kalibri­ tion. The road was not discovered, but it is supposed it lay north of the explored area. Despite its unaltered basic outline, the settlement has undergone constant change, mainly due to natural processes. Because of the aquiferous nature of the slopes above the site, torrents are regular during heavy rainfall; prior to the amelioration projects in the 20th century, they deposited debris in the valley below. The first attempts at amelioration were conducted in the Roman period when walls were installed in order to block the waterways but to no avail (cf. Figs.: 4; 6; 8; 12): the troughs were constantly moving. Frequent landslides were another concern. Some­times they damaged the houses so severely they collapsed; this is why the settlement underwent several restorations, which could be quite extensive. During these restoration projects, new construction techniques were employed in order to hinder the detrimental natural processes. It was precisely these unfavourable conditions that created a neat, archaeologically readable stratigraphy. Based on landslide deposits, the events could be divided into four major chronological phases, which division is only very approximate, since the earlier layers tended to mix with the later ones due to constant landslides. PHASE 1 The earliest remains are confined to the western part of the investigated area (Fig. 4), where several ditches and pits were found. Their function is difficult to ascertain, with the exception of the large natural pit (SE 3415=3432) 1.2 m deep, which was filled with burning residues (lay­ers rich in charcoal and burnt soil). The latter indicates the close vicinity of furnaces or similar objects. There were no finds in the lower fills. Radiocarbon analysis of the charcoal from one of these fills yielded the dating of 2045 ± 30 BP,1 which, calibrated, places the origin of charcoal between the mid- 2nd century BC and the first quarter of the 1st century AD. In the layer above this fill, a fragment of the Dres­sel 6B type amphora was discovered (appearing between the second half of the 1st century BC and the end of the 1st century AD (Bezeczky 1998, 6-10; Carre, Pesavento Mattioli 2003, 461-463)), based on which the earliest traces of settlement in Blagovica can be dated no later than the first half of the 1st century AD, possibly even the end of the 1st century BC. At the same time, two large, parallel adjacent wooden Buildings VIIIa and VIIIb were erected (7.3 m apart) in the eastern part of the settlement. They were over 30 m long (Fig. 5). Building VIIIb was 5 m wide, while the width of Building VIIIa was not determined. 1 Poz-63337, 95 cal. BC–2 cal. AD (calibrated age 1-.: (68.2% ver.)) or 165 cal. BC–24 cal. AD (calibrated age 2-. (95.4% ver.)) Sl. 5: Blagovica. Prva faza, objekt VIIIb, pogled proti severozahodu. Fig. 5: Blagovica. Phase 1, Building VIIIb, view towards northeast. (Foto / Photo: Maja Korošec) kar ob kalibriraciji nastanek oglja uvršča med sredino 2. st. pr. n. št. do prve četrtine 1. st. n. št. V plasti, ki je ležala nad tem polnilom, je bil odkrit odlomek amfore tipa Dressel 6B, kakršne se pojavljajo od druge polovice 1. st. pr. n. št. do konca 1. st. n. št. (Bezeczky 1998, 6–10; Carre, Pesavento Mattioli 2003, 461–463). Na podlagi slednjega začetke poselitve na Blagovici umeščamo najkasneje v prvo polovico 1. st. n. št., morda pa že v konec 1. st. pr. n. št. Sočasno sta bila na vzhodnem delu naselja postav­ljena dva velika lesena objekta VIIIa in VIIIb, ki sta ležala vzporedno, drug ob drugem (razdalja med njima je znašala 7,3 m). Dolga sta bila več kot 30 m (sl. 5). Širina objekta VIIIb je znašala 5 m, širine objekta VIIIa pa ne poznamo. Nosilno konstrukcijo obeh so sestavljali vertikalni tramovi, zabiti v zemljo (razdalja med njimi pribl. 3 m). Zaradi pomanjkanja najdb ni jasno, čemu sta objekta služila. Lahko le domnevamo, da gre za gospodarska objekta – hlev, skladišče ipd. Velikost objektov in močni temelji kažejo, da poselitev že od začetka ni bila prepuščena naključne-mu razvoju, temveč lahko že vse od njenega začetka govorimo o načrtni obsežni izgradnji naselja, ki je po­trebovalo objekte velikih dimenzij. Prav tako že v 1. fazi lahko govorimo o funkcionalni razdelitvi naselbine. Na zahodnem delu so prepoznani znaki obrtnih dejavnosti, na vzhodnem delu pa so verjetno hlevi oz. skladišča. FAZA 2 Enaka funkcionalna delitev je značilna tudi za dru-go fazo gradnje. Na vzhodnem delu je v tem času objekta VIIIa in VIIIb nadomestil objekt IX (sl. 6; 7). Tudi v tem primeru gre za veliko, več kot 30 m dolgo in vsaj 8 m široko leseno stavbo. Zanjo so uporabili drugačno tehniko gradnje. Stavba je bila zgrajena iz kombinacije horizontalnih tramov, položenih v vnaprej izkopane jarke, in vertikalnih tramov, zabitih v tla, med katere so bile vpete stene. Stavbo je obdajal portik. V notranjosti je bila razdeljena na manjše prostore (4,5–3,5 m × 3,3 m). Zaradi tega je najverjetnejša domneva, da je stavba rana starost 2-. (95,4 % ver.)). The supporting construction of both buildings was made up of vertical posts, driven into the ground (set approx. 3 m apart). Due to the lack of finds, it is not clear what the purpose of these two buildings was. We can only as­sume they were commercial-related constructions, such as stables, warehouses, etc. The size of the buildings and the strong foundations indicate that the development of the settlement was care­fully planned from the very beginning and that this was a large-scale building project. It is also safe to say that the settlement had a utilitarian ground plan established in Phase 1, according to which crafts were concentrated in the western part of the settlement, while the stables and warehouses were located to the east. PHASE 2 The same functional division was still in place during the second construction phase. In the eastern part of the settlement, the Buildings VIIIa and VIIIb were replaced by Building IX (Figs. 6; 7), a large, over 30 m long and 8 m wide wooden construction. In this case, a different con­struction technique was employed, using a combination of wooden beams horizontally placed into pre-dug ditches and vertical posts driven into the ground, between which walls were constructed. The building was surrounded by a portico. It was divided into smaller rooms (4.5-3.5 m x 3.3 m). For this reason, the most likely assumption is that the building served as a warehouse. Unfortunately, the in situ finds were scarce. At the same time, a small furnace (SE 3229) was constructed in the western part of the site (near the filled earlier pit 3415). Next to it, a large pit was excavated (SE 3297) alongside several smaller waste pits. It was filled (like the pit SE 3415 from Phase 1) with several sediments containing fire residues. Judging by such a large amount of waste, it is safe to assume that there were several furnaces located nearby. During Phase 2, stone was first used as a building material. Pebbles and quarry stones were used for foun­dations (wall above the furnace, Building XI, Building XIII). In some parts, impressions of wooden posts were Fig. 6: Blagovica. Ground plan of settlement Phase 2. Sl. 7: Blagovica. Druga faza, objekt IX, pogled proti severu. Fig. 7: Blagovica. Phase 2, Building IX, view towards north (Foto / Photo: Maja Korošec) služila kot skladišče. Žal pa so bile tudi v tem objektu najdbe in situ redke. Sočasno je bila na zahodnem delu najdišča (ob v tej fazi že zasuti starejši jami 3415) postavljena manjša peč (SE 3229). Ob njej je bila poleg več manjših odpad­nih jam izkopana tudi večja jama (SE 3297). Ta je bila (prav tako kot jama 3415 iz 1. faze) zapolnjena z več sedimenti, ki so vsebovali ostanke kurjenja. Sodeč po tako veliki količini odpada domnevamo, da je bilo v bližini še več peči. V 2. fazi se je kot gradbeni material prvič pojavil tudi kamen. Oblice in lomljenci so bili uporabljeni za temelje zidov (zidec nad pečjo, objekt XI, objekt XIII). Na nekaterih delih so bili na dnu jarkov za temelje naj­deni odtisi kolov. To kaže, da so bili objekti še vedno leseni, vendar pa so imeli kamnite temelje. Zaradi slabe ohranjenosti teh ostalin je težko reči, ali gre za ostanke stavb. Morda v njih lahko prepoznamo ostanke urejanja s škarpami in palisadami. Najdbe te faze so podobne najdbam prve faze. Večinoma izhajajo iz plazin in jih je težko ozko časov-no zamejiti. Zato predvidevamo, da je druga faza prvi sledila zelo kmalu. Fazo 2 tako okvirno umeščamo v čas sredine oz. druge polovice 1. st. n. št., morda še v čas začetka 2. stoletja. FAZA 3 Ob koncu 1. oz. na začetku 2. st. n. št. je večji plaz zasul osrednji del najdišča in uničil vse lesene objekte. Po tem dogodku je naselbina doživela celovito preno­vo in tudi funkcionalne spremembe. Ostanki ruševin lesenih objektov so bili skrbno odstranjeni, predel pa izravnan s 30 cm debelim nasutjem in ponovno pozidan. Nekoliko se je spremenila tudi funkcionalna delitev prostora (sl. 8). Na vzhodnem delu je bila že leta 1977 najdena pravokotna jama, velika 3,7 × 3 × 1m, zasuta s kamni. Interpretirali so jo kot postament za grobnico (Zu­pančič 1979, 279–281; Stražar 1985, 82). Ta ob gradnji found on the bottom of foundations’ ditches, indicating that buildings were wooden, but with stone foundations. Due to the poor state of these remains, it is difficult to say whether they belong to previous buildings; perhaps they can be seen as traces of scarps and palisades. The finds from this phase are similar to the finds from Phase 1. They were mostly located in landslide de­posits and are therefore difficult to date, which makes it possible to assume that Phase 2 closely followed Phase 1. Phase 2 can thus be dated to the mid-1st or the second half of the 1st century AD, possibly even the early 2nd century AD. PHASE 3 By the end of the 1st century or the beginning of the 2nd century AD, a major landslide struck the central part of the settlement, ruining all wooden structures. After this, the settlement underwent major reconstruction as well as functional change. The remains of wooden structures were thoroughly removed; the central area was levelled with a 30 cm thick layer of gravel and then rebuilt. The functional division also changed (Fig. 8). In the eastern part, a rectangular pit was found in 1977, measuring 3.7 x 3 x 1 m. Filled with stones, it was interpreted as a pedestal of a burial chamber (Zupančič 1979, 279-281; Stražar 1985, 82). It was not removed dur­ing the construction of the gas pipeline, which allowed re-documenting in the time of new surveys. During the clearing of the surrounding area, traces of burnt material were found around the pedestal, in which several plates with red-slip were found that appeared between the 1st and 4th century (Plesničar-Gec 1972, 55, Pl. 33. 2; Plesničar-Gec 1977, 54-56, Pl. 7; Krajšek, Stergar 2008, 252). During Phase 3, the riverbed ran west of the pedestal, next to which Building I was constructed. It was 29 m long and 14 m wide and had stone foundations. In the northern part, a portion of a stone wall was preserved on the foundations. On the SE corner, a rectangular stone extension was uncovered in 1977 (Fig. 8), which was Sl. 8: Blagovica. Tloris tretje faze poselitve. Fig. 8: Blagovica. Ground plan of settlement Phase 3. 239 Sl. 9: Blagovica. Tretja faza. Steklene posode iz groba 1. Fig. 9: Blagovica. Phase 3. Glass vessels from Grave 1. (Foto / Photo: Jašar Skorupan) plinovoda ni bil odstranjen, kar je omogočilo ponovno dokumentiranje ob novih raziskavah. Ob čiščenju okolice so bile ob postamentu najdene sledi žganine, v kateri je ležalo več krožnikov z rdečim premazom, ki se pojavljajo od 1. st. do 4. st. (Plesničar-Gec 1972, 55, t. 33: 2; Plesničar-Gec 1977, 54–56, t. 7; Krajšek, Stergar 2008, 252). Zahodno od postamenta je v 3. fazi poselitve tekla struga. Tik ob njo so postavili t. i. objekt I. To je bil 29 m dolg in 14 m širok prostor s kamnitimi temelji. Na severnem delu se je na temeljih ohranil tudi del kam­nitega zidu. Ob JV vogalu so l. 1977 našli pravokotno kamnito razširitev (sl. 8), ki so jo interpretirali kot še en postament za grobnico (Zupančič 1979, 279–281; Stražar 1985, 82). Razlog za to interpretacijo je bila podobnost s prvim postamentom in to, da je bilo v mlajših ruševin­skih plasteh tega objekta najdenih nekaj poznorimskih oz. zgodnjesrednjeveških skeletnih grobov. Na podlagi tega bi lahko domnevali, da objekt I predstavlja mejo grobne parcele. Vendar pa v notranjosti objekta z novimi izkopavanji do 4. faze ni bilo odkritih nobenih pokopov ali morebitnih drugih posegov. Z izjemo dveh manjših kurišč je bil prostor povsem prazen. Tudi tlak ni bil najden. Zato teze o grobni parceli ne moremo potrditi. Le 7 m zahodno od objekta I je ležal še en večji objekt – objekt III. Na predelu med obema objektoma je bilo tik pod ornico najdenih več pravokotnih jam, ki so bile večinoma zasute s kamni. V nekaterih so bili najdeni odlomki lončenine in stekla, v dveh tudi izjemno redki drobci ožganih kosti. Na dnu ene izmed jam je bil najden maltni tlak, stene pa so bile obzidane. V ruševini je ležalo več fragmentov keramičnih posod, 9 steklenih skodelic, dva steklena krožnika, steklen lonec, ki bi lahko služil kot žara, ter dva steklena balzamarija (sl. 9). Skodelice interpreted as another tomb pedestal (Zupančič 1979, 279-281; Stražar 1985, 82). This explanation was based on the similarity to the first pedestal and the fact that several late Roman and early medieval inhumation graves were found in destruction layers of this construction. There is reason to believe that Building I represented the bound­ary of a grave plot. However, new excavations of Phase 4 yielded no burials or other interventions in the interior of Building I; with the exception of two small fireplaces, the room was empty and was not even paved. That is why the grave plot hypothesis cannot be confirmed. Only 7 m west of Building I lay another large build­ing, i.e. Building III. In the area between the two buildings, several rectangular pits were found immediately beneath the surface layer. Mostly filled with stones, some of these pits contained fragments of pottery and glass. Two pits also contained very few fragments of burnt bones. On the bottom of one of these pits, which happened to be walled, mortar paving was found. There were several fragments of ceramic vessels in the debris, nine glass cups, two glass plates, a glass pot, which could serve as an urn and two glass balsamaria (Fig. 9). The cups (Lazar 2003, 77, Figs. 29, 30), balsamaria (Lazar 2003, 180, 195, Fig. 50; Istenič 2000, 144) and the plates (Lazar 2003, 64, Fig. 28; Istenič 2000, 40) are dated to the second half of the 1st and the beginning of the 2nd centuries, while similar glass jars appear in the 2nd and 3rd centuries (Lazar 2003, 164-168, Fig. 48; Istenič 1999, 74, Fig. 56). Based on the inventory, it is assumed that this was a walled grave (Grave 1); ac­cording to the grave goods, the grave is dated to the early 2nd century. Other nearby pits can also be interpreted as potential graves or tombstone pedestals. Building III was built only 3 metres from Grave 1 (Fig. 10). Its function is less ambiguous than that of (Lazar 2003, 77; sl. 29, 30 ), balzamarija (Lazar 2003, 180, 195, sl. 50; Istenič 2000, 144) in krožnika (Lazar 2003, 64, sl. 28; Istenič 2000, 40) datiramo v drugo polovico 1. in začetek 2. st., podobni stekleni lonci pa se pojavljajo v 2. in 3. st. (Lazar 2003, 164–168, sl. 48; Istenič 1999, 74, sl. 56). Glede na inventar domnevamo, da gre za zidano grobnico (t. i. grob 1). Grob pa, glede na najdbe, dati­ramo v čas začetka 2. st. Tudi ostale jame v neposredni bližini interpretiramo kot potencialne grobne jame oz. postamente za nagrobnike. Nenavadno pa je, da je bil le 3 metre zahodno od groba 1 postavljen objekt III (sl. 10). Funkcija slednjega je bolj nedvoumna od funkcije objekta I. Gre za stavbo s kamnitimi temelji, ki je bila predeljena na tri prostore. Vhod vanjo je bil iz zahodne smeri v prostoru IIIc. V srednjem in vzhodnem prostoru (IIIa in IIIb) je bilo v notranjosti še več manjših kamnitih temeljev, ki pa niso v pomoč pri določanju funkcije prostorov. V tem pogledu je zanimivejši manjši zahodni prostor (IIIc). V tem so bili najdeni trije vzporedni kamniti temelji. Razdalja med njimi je bila 0,9 m. Na podlagi slednjega domnevamo, da gre za ostanke kašče oz. skladišča za žito. Ti temelji bi lahko nosili lesena tla, ki so bila v kaščah zaradi potrebe po kroženju zraka dvignjena (Rickman 1971). Celoten objekt bi torej lahko služil kot skladišče. Zahodno so bili v tem času postavljeni kar trije manjši objekti (V, VI, VII; sl. 8; 11). Vsi so imeli zidane temelje, objekta V in VI pa zagotovo tudi kamnite stene (oz. vsaj del njih). Malte med kamni ni bilo oz. se vsaj ni ohranila. Tudi strešne kritine ne poznamo. Na celot­nem območju je bilo najdeno razmeroma malo tegul in imbreksov. Funkcije teh objektov ne poznamo, treba pa je izpostaviti, da se stavbe na tem delu po obliki in velikosti zelo razlikujejo od tistih na vzhodnem delu najdišča. Predvsem so manjše. Še posebej zanimiv je objekt VI, saj je dobro raz­vidno, s kakšnimi razmerami so se spopadali graditelji. Leži na predelu, ki močno pada proti jugu, kar pomeni, da je bila moč plazenja tu še večja. Verjetno je imel prav zato objekt na vzhodni in zahodni strani dvojne temelje (sl. 11). Ali je bila taka gradnja že od samega začetka ali pa je bil drugi temelj dodan kasneje, ni jasno. Južno od objekta VI je bila naselbina v tej fazi za­mejena z zidom. Ohranila se je le ena lega kamnov zidu (sl. 8). Način gradnje in njegova majhna širina (30 cm) pa ne nakazujeta, da bi šlo morda za “obzidje”, temveč prej za simbolično mejo naselja ali pa za obrambo pred reko. Struga slednje je namreč takrat tekla le 5 m južneje. Enako usmerjenost kot zid ima plast kamnov na načrtu izkopavanj iz l. 1977 (sl. 8). To so sicer pripiso­vali ostankom ceste, a ni bila do konca raziskana. Zato obstaja možnost, da gre za ruševino zidu, ki je na jugu zamejeval naselje. V tem obsegu je naselje delovalo vse do sredine 3. st. Building I. It was constructed with stone foundations and divided into three rooms. The entrance was from the west in Room IIIc. In both eastern rooms (IIIa and IIIb), several short stone foundations were found, which did not help determine the function of the rooms. In this respect, the small western room (IIIc) is more interesting. Here, three parallel stone foundations were found. The distance between them was 0.9 m, based on which it is safe to as­sume it is the remains of a granary; the foundations could have supported a raised wooden floor, which provided much-needed air circulation (Rickman 1971). The entire facility could have served as a warehouse. During this period, three small buildings were constructed to the west (V, VI, VII; Figs. 8; 11), all with stone foundations. Buildings V and VI featured stone walls (at least partially) with no mortar; at least, none was preserved. Roofing is also unknown; in the whole area, relatively few tegulae and imbrices were found. The purpose of these constructions is unclear, but it should be pointed out that the buildings in this part are very different in shape and size than those in the eastern part of the site. They are smaller. Building VI is of particular interest, as it shows the conditions masons faced. It lies in the section sloping towards the south, which means that land sliding was even more of an issue here. This is probably why the building was reinforced with double foundations on the eastern and western sides (Fig. 11). It is unclear whether this was the original plan or an additional improvement. South of Building VI, the settlement was walled dur­ing this phase. Only one line of stones survived (Fig. 8); however, the construction and negligible width (30 cm) of the wall do not indicate that this was a protective rampart, but rather a symbolic boundary of the settlement or de-fence against the river, which flowed only 5 m to the south. The layer of stones on the 1977 excavation plan features the same orientation (Fig. 8). This was attributed to the remains of the road, but it was not fully explored. Therefore, there is a possibility that it is, in fact, the re­ FAZA 4 V naslednji fazi se je obseg naselja nekoliko zmanj­šal (sl. 12), pomembnejši je postal zahodni predel. Na mestu nekdanjega objekta V so postavili nov objekt IV (sl. 11; 13). Razdeljen je bil v dva prostora. Tlak iz drobirja v južnem prostoru in peč v severnem kažeta, da so tu potekale gospodarske dejavnosti. Objekt IV je imel 1,1 m globoke temelje, zgrajene v tehniki ribje kosti, kar naj bi zagotavljalo tudi stabilnejšo gradnjo. Na tem delu je podlaga namreč statično izjemno nestabilna, kar kažejo temelji, ki so bili ob našem izkopu močno nag-njeni v smeri po pobočju navzdol. Zamik med zgornjim in spodnjim robom temelja je mestoma presegal 0,15 m, kar priča o silovitih pritiskih zemeljskih mas. Te so bile usodne tudi za starejši objekt VI, ki mu niso pomagali niti dvojni temelji. Ta objekt je bil v celoti porušen in prekrit s pobočnimi nanosi. Na njegovem mestu so postavili nov, lesen objekt XII (sl. 12). Tudi objekt III (iz faze 3) je bil v tem času vsaj delno podrt. V nekdanjem prostoru IIIa je bil vzpostavljen metalurški obrat z več pečmi. Ob njih so bili najdeni številni kosi železne žlindre. Analiza je pokazala, da je šlo za žlindro iz rude lokalnega izvora z majhno vsebnostjo železa, ki so jo talili prav tu (Verbič 2014, 170–171). Torej ne govorimo le o izdelovanju končnih produktov, temveč se je na tem prostoru odvijal celoten postopek pridobivanja železa. Seveda le v manjših količinah. V zasutjih peči sta bila najdena dva novca cesarja Konstantina II., kovana med letoma 351 in 355, ter novec Konstantina I., datiran v čas med letoma 330 in 333. V mains of the wall, which confined the settlement at the southern part. The settlement remained in this layout until the mid-3rd century. PHASE 4 The extent of the settlement decreased slightly (Fig. 12), its western part gaining in significance. At the site of the former Building V, the new Building IV was built (Figs. 11; 13), which was divided into two rooms. The gravel paving in the southern part and the furnace in the northern part of the room indicate that economic activities took place here. Building IV had a 1.1 m deep foundation, built in fishbone technique, which provided a more stable construction. The terrain is exceptionally unstable in this part: when excavated, the foundations were leaning heavily down the slope, the gap between their upper and lower edges exceeding 0.15 m at places, which indicates the severe pressure. This must have been the undoing of the earlier Building VI, which even the double foundation could not prevent. This building was completely demolished and covered by landslides from the slopes. In its place, the new, wooden Building XII was constructed (Fig. 12). During this time, Building III (from Phase 3) was at least partly ruined. In former Room IIIa, a metallurgical plant with several furnaces was established. Quite a few pieces of iron slag were found nearby. Analyses showed that this was slag from local ore with a low iron content, Sl. 12: Blagovica. Tloris četrte faze poselitve. Fig. 12: Blagovica. Ground plan of settlement Phase 4. 243 ruševini objekta III pa je bil najden tudi novec cesarja Jovijana (Iovianus), kovan v letih 363–364. Na vzhodnem robu naselja iz tega časa ni sledi poselitve. Del zahodnega zidu objekta I (iz faze 3) je bil porušen, zato domnevamo, da ni bil več namenjen svoji prvotni funkciji. V notranjosti nekdanjega objekta I je bilo brez pridatkov pokopano dekle staro med 12 in 15 let (t. i. grob 2). Ena izmed kosti roke je bila radiokar­bonsko datirana.2 Na podlagi tega grob datiramo med drugo polovico 3. in koncem 4. st. V ta čas bi lahko umestili tudi štiri skeletne grobo­ve, ki so bili v ruševini na jugovzhodni strani objekta I najdeni že l. 1977 (Zupančič 1979; sl. 12). Tudi ti grobovi so večinoma brez pridatkov. Le v enem je bil najden kuhinjski lonček, okrašen z valovnico (Stražar 1985, 83). Tudi to fazo z najdbami težko datiramo. Ume-ščamo jih pretežno med drugo polovico 1. in sredino 3.  st., vendar je treba poudariti, da so najdbe večinoma sekundarno odložene (plazine, ruševinske plasti, …) ter pravzaprav datirajo 3. fazo. Novčne najdbe v kontekstih 4. faze pa kažejo na čas sredine 4. st. Torej bi celotno fazo 4 nekako umestili v čas 3. in 4. st. Mlajših sledov na raziskanem predelu ni bilo za­slediti. Od opustitve naselja je bil prostor izkoriščan le v kmetijske namene. Modernih posegov je bilo razmeroma malo. Izjema so le posegi ob obstoječem gospodarskem poslopju na jugozahodnem delu raziskanega predela, kjer je bil zgrajen verjetno novoveški vodnjak. Tudi najdb iz mlajših zgodovinskih obdobij je izjemno malo. 2 Poz-63338. Datacija 1695 ± 30 BP ustreza času 264–273 cal. AD (kalibrirana starost 1-.: 6,1 % ver.) ali 331–394 cal. AD (kalibrirana starost 1-.: 62,1 % ver.), oziroma 255–302 cal. AD (kalibrirana starost 2-.: 19,8 % ver.) ali 316–412 cal. AD (ka­librirana starost 2-.: 75,6 % ver.). which was melted at this location (Verbič 2014, 170-171). Therefore, not only finished products were produced here; instead, the whole process of iron extraction took place in the area, but in limited quantities. In the filling of two furnaces, two coins were found from the period of Emperor Constantine II, minted be­tween 351 and 355, and one coin of Constantine I, dating from the period between 330 and 333. In the ruins of Building III, a coin from the period of Emperor Jovian, minted 363–364, was found. On the eastern edge of the excavation area, there is no trace of settlement during this period. The western wall of Building I was partially demolished, making it possible to assume that it no longer served its original purpose. In the interior of the former Building I, a girl aged between 12 and 15 years was buried without any grave goods (Grave 2). One of the arm bones was radiocarbon dated,2 based on which the grave was dated between the second half of the 3rd and the end of the 4th centuries. This could also be the timeframe for four inhuma­tion graves, located in the ruins of the south-eastern side of Building I and found in 1977 (Zupančič, 1979). These graves, too, are mostly without any grave goods; only one featured a kitchen pot decorated with a wavy line (Stražar 1985, 83). Phase 4 is not easily dated by small finds. It should be emphasized that most of the small finds were discov­ered in secondary deposits (landslides, ruins, etc.) and dated between the second half of the 1st and the mid-3rd centuries. Thus they originate from Phase 3. In contrast, the coins belonging to the contexts of Phase 4 indicate the time of the mid- 4th century. Therefore, Phase 4 should be dated to the 3rd and 4th centuries. No later traces were observed in the investigated area. After the abandoning of the settlement, the area was used only for agricultural purposes. Modern interventions were relatively scarce, with the exception of interventions alongside the existing farm building in the southwestern part of the explored area, where a modern-era fountain was most likely built. Finds from later historical periods are also scarce. THE PURPOSE AND STATUS OF THE SETTLEMENT The Roman settlement at Blagovica had a clearly defined layout from the very start. In the 1st century AD and quite possibly even as early as the 1st century BC, a settlement with several multi-purpose buildings was built. In the eastern part, large commercial buildings, possibly 2 Poz-63338. Dating 1695 ± 30 BP corresponds to 264– 273 cal. AD (calibrated age 1-.: 6.1% ver.) or 331–394 cal. AD (calibrated age 1-.: 62.1% ver.) or 255–302 cal. AD (cali­brated age 2-.: 19.8% ver.) or 316A–412 cal. AD (calibrated age 2-.: 75.6 % ver.). FUNKCIJA IN STATUS NASELJA Antična naselbina pri Blagovici je imela že ob vzpo­stavitvi načrtno in velikopotezno urbanistično zasnovo. V 1. st. n. št., morda pa že v 1. st. pr. n. št. je bilo zgrajeno naselje z več poslopji, ki so imela različne funkcije. Na vzhodnem delu so se nizala velika gospodarska poslopja, morda skladišča ali hlevi (VIIIa in VIIIb). Na zahodnem delu je bilo več neopredeljivih jam z ostanki žganin. Čeprav so prvotne objekte že zelo kmalu opustili (v sredini 1. st.), je zasnova novih objektov sledila prvotni. Na zahodnem delu so stala gospodarska poslopja (XIII, XI in skladišče IX), na zahodnem delu pa znova predel s pečjo. Največji razcvet je naselbina doživela med koncem 1. in 3. stoletjem, ko so lesene objekte zamenjali objekti s kamnitimi temelji. Na vzhodnem delu so še vedno stali veliki objekti, ki so lahko služili kot ograde oz. hlevi (I) in skladišča (III), na zahodu pa so zgradili več manjših objektov (VI, V, VII), katerih funkcije ne poznamo. Poleg tega je na vzhodnem delu naselbine najpozneje od 2. st. dalje ležalo tudi pripadajoče grobišče, ki je bilo stisnjeno kar med dva gospodarska objekta. V sredini 3. st. se je naselje nekoliko skrčilo. Vzhod­ni objekti so bili opuščeni, v nekdanjem objektu I so za-čeli pokopavati. Na mestu objekta III je bil vzpostavljen predel namenjen metalurškim postopkom, ki je deloval vsaj do sredine 4. st. Neka gospodarska dejavnost je potekala tudi v novonastalem objektu V. Treba je poudariti, da se usmerjenost objektov ves čas obstoja naselbine ni spreminjala, kar sicer ni nenavadno, saj je bila stisnjena med takratno strugo Radomlje in brežino. Poleg tega je treba opozoriti tudi na za naselbino netipično majhno količino najdenega kuhinjskega posodja ter razmeroma veliko količino uvožene lončenine (fina namizna lončenina, sigila­ta …). Sklop steklenih posod iz groba 1 pa priča tudi o visokem standardu prebivalcev. Dostop do uvoženih predmetov kaže, da je naselbina ležala ob prometni cestni povezavi, verjetno kar ob glavni prometnici med Italijo in Norikom. Tudi urbanistična zasnova naselja nakazuje, da je tu najverjetneje ležala poštna postaja. Te so bile v rimskem času navadno nekoliko zunaj naselij ali pa so ležale celo povsem na samem. Sestavljal jih je sklop več poslopij. Od gospodarskih do skladišč, shramb za vozove, poslopij za cestno policijo, do bivalnih predelov, ki so po možnosti imeli še terme (Šašel Kos 1997, 24). Slednje sicer niso bile odkrite, vse ostalo pa bi v Blagovici lahko prepoznali. Ob tej predpostavki lahko domnevamo, da bi tu lahko stala na Tabuli Peutingeriani omenjena rimska poštna postaja Ad publicanos.3 Ta naj bi ležala ob cesti Emona (Ljubljana)–Celeia (Celje) pred Atransom (Troja-ne). Od Emone naj bi bila oddaljena 20 milj, od Atransa 3 Tab. Peut. IV, 2. warehouses or stables, were concentrated (VIIIa and VIIIb). In the western part, several hard-to-explain pits with burnt residues were located. Although the original buildings were abandoned rather soon (mid-1st century AD), the original architectural design was followed. In the western part, there were commercial Buildings XIII and XI and Warehouse IX, and in the western part again an area with a furnace. The settlement mainly flourished between the late 1st and the 3rd centuries, when wooden buildings were replaced by buildings with stone foundations. The eastern part still featured large buildings, which either served as stables (I) or warehouses (III), while several smaller facilities (VI, V, VII) with unknown purpose were built to the west. In the eastern part of the settlement, there was a small 2nd-century cemetery, which was positioned between two commercial buildings. In the mid- 3rd century the settlement shrunk a bit. The eastern buildings were abandoned, in the former Building I burials took place. At the site of former Building III, a metallurgical plant was established, which operated at least until the mid-4th century. No well-defined activity took place in the newly constructed Building V. It should be emphasized that the orientation of the buildings did not change during the time, which is not unusual since it was compressed between the Radomlja River and the adjacent slopes. The structure of the ceramic finds is not quite typical for the ordinary rural settlement. A small quantity of kitchenware was found, and a rela­tively large quantity of imported pottery (fine tableware, terra sigillata, etc.). A set of glass vessels from Grave 1 shows a high standard of living. Access to imported items indicates that the settlement lay along a busy road connection, probably just off the main road between Italy and Noricum. The layout of the settlement suggests that this was most likely the location of a Roman postal station. They were usually somewhat outside of settlements, or even completely on their own, comprising a set of several buildings from commercial facilities to warehouses, cart storages, road police buildings, to residential areas andpossibly thermae (Šašel Kos 1997, 24). The latter were not discovered, but everything else could be recognized in Blagovica. This leads to the assumption that this could be the Ad publicanos postal station mentioned by the Tabula Peutingeriana.3 It supposedly lay along the Emona (Ljub­ljana)–Celeia (Celje) road before Atrans (Trojane), about 20 miles from Emona and 6 miles from Atrans. The Roman mile estimated at 1.48 km, the station was lying approximately 8.88 km from today’s Trojane and 29.6 km from Ljubljana. In the past, researchers placed the Ad Publicanos station around Lukovica, Podpeč, or in Krašnja (Müllner1879, 37, 85; Šašel 1954, 16). This thesis was influenced 3 Tab. Peut. IV, 2. pa 6 milj. Rimska milja znaša današnjih 1,48 km, kar pomeni, da je postaja ležala pribl. 8,88 km od Trojan in 29,6 km od Ljubljane. V preteklosti so raziskovalci Ad publicanos po­stavljali na območje Lukovice, v Podpeč ali v Krašnjo (Müllner 1879, 37, 85; Šašel 1954, 16). Na te teze je vplivalo dejstvo, da je v novem veku v Podpeči pri Lukovici delovala pomembna poštna postaja, zadnja pred Trojanami. Poleg tega do l. 2013 rimskodobne naselbinske ostaline v dolini Črnega grabna niso bile poznane. Vendar pa razdalje med kraji ponujenih tez ne potrjujejo najbolje. Razdalja med Podpečjo in Trojanami po današnji cesti namreč znaša 16,25 km. Ustrezneje bi bilo, da bi postajo locirali na območje Blagovice, saj so Trojane od središča vasi po današnji cesti oddaljene 7,3 km (od našega najdišča pa 7 km; sl. 1). To je ob dejstvu, da natančne trase rimske ceste ne poznamo,4 zagotovo bližje 6 miljam, kot Lukovica. Tabula Peutingeriana ne navaja statusa naselja Ad publicanos. Ne vemo torej, ali gre za postajo s preno-čiščem (mansio) ali postajo za menjavo vprege (mutatio). Slednje se sicer omenjajo šele od 4. st. dalje (zlasti na Burdigalskem oz. Jeruzalemskem itinerariju, ki pa posta­je Ad publicanos ne omenjata). Razdalje med postajami so znašale med 6 in 12 milj, pri čemer je bilo razmerje med mansio in mutatio 1:2 ali 1:3. Ker za Atrans vemo, da gre za mansio, je mogoče sklepati, da le 6 milj pred počivališčem ni ležala postaja, ki bi služila enakemunamenu (Šašel Kos 1997, 31). Nekaj več o statusu naselja lahko izvemo iz njego­vega imena. Ad publicanos (Pri zakupnikih) po mnenju Marjete Šašel Kos nakazuje, da je v naselju delovala mitnica, saj so te službe v rimskem času pogosto dajali v zakup (Šašel Kos 1997, 31). Poleg tega je ime naselja povsem rimskega izvora, kar nakazuje, da pred nastan­kom pošte v bližini ni bilo starejše naselbine, saj bi bila v tem primeru postaja poimenovana po njej (Šašel Kos 1997, 31). Tudi to se sklada s tezo o lociranju te postaje na Blagovico. Na najdišču namreč ni bilo najdenih pred­metov, ki bi jih lahko povezovali s staroselsko poselitvijo. Tudi v kuhinjskih posodah ni sledi staroselske tradicije. Naselbino so na samem povsem na novo postavili Rim-ljani. Poleg tega ne smemo pozabiti, da leži na točki, ko se razmeroma ravni del doline Črni graben zaključuje, cesta pa se začne dvigovati proti Trojanam. Torej na točki, kjer bi popotnik zagotovo želel zamenjati vprego. 4 Obstajajo domneve, da se je rimska cesta ognila dolini Črnega grabna in je tekla po zalednih hribih (Sagadin, ustno 19.11.2013). Možno je, da se je ognila zoženemu zgornje-mu delu doline Radomlje, ki je bila v tem delu izrazito hu­dourniška. Če bi se pri Petelinjeku povzpela proti Javorniku in tekla mimo Prvin do Šentožbolta, bi razdalja do Trojan povsem ustrezala 6 miljam. by the fact that in Podpeč near Lukovica an important modern postal station was located, the last one before Trojane. Furthermore, until 2013, Roman settlements in the Črni graben valley were not known. However, the distances between the sites do not quite confirm this the­sis. The distance between Podpeč and Trojane on today’s road is 16.25 km. It would be more appropriate to locate the station in the area of Blagovica, since Trojane is now 7.3 km away from the centre of the village (7 km from our site; Fig. 1). Accounting for the fact that we do not know the exact route of the Roman road,4 this is certainly closer to 6 miles than Lukovica. Tabula Peutingeriana does not mention the status of the Ad publicanos settlement. We do not know, there­fore, whether it was a mansio or a mutatio. The latter are mentioned only from the 4th century onwards e.g. in Itinerarium Burdigalense, which does not mention the Ad publicanos station. The distances between stations ranged between 6 and 12 miles, the ratio between mansio and mutatio being 1:2 or 1:3. Since we know that Atrans was a mansio, it can be concluded that only 6 miles before the resting place there was no station that would serve thesame purpose (Šašel Kos 1997, 31). We can learn more about the status of the settlement from its name. According to Marjeta Šašel Kos, the name Ad publicanos (at “The tax collection agents”) indicates that a toll station was operating in the settlement, as theseservices were frequently leased (Šašel Kos 1997, 31). Fur­thermore, the name of the settlement is of purely Roman origin, which suggests that prior to the creation of the post office there was no earlier settlement, since in this case the station would be named after it (Šašel Kos 1997, 31). This confirms the location of the station at Blagovica. Namely, there were no objects found on the site that could be associated with the indigenous settlement. Even the kitchenwares have no trace of indigenous tradition. The settlement was completely built anew by the Romans. It must not be forgotten that the settlement lies on a point at which the relatively levelled part of the Črni graben valley ends, and the road begins to rise towards Trojane – at a point, then, where a traveller would definitely want to replace their beasts of burden. Translation: Gregor Pobežin 4 There are assumptions that the Roman road took a turn away from the Črni graben valley, running along the surrounding hills (Sagadin, orally 19.11.2013). It is possible it avoided the torrential upper part of the Radomlja River valley. If it took a turn towards Javornik at Petelinjek, running past Prvine to Šentožbolt, the distance to Trojane would perfectly correspond 6 miles. BEZECZKY, T. 1998, The Laecanius amphora stamps and the villas of Brijuni. – Osterreichische Akademie der Wissen­schaften, philosophisch-historische Klasse, Denkschriften 261. CARRE, M.-B., S. PESAVENTO MATTIOLI 2003, Tentativo di classificazione delle anfore olearie adriatiche. – Aquilea Nostra 74, 453–476. GABROVEC S. 1975, Blagovica. – V / In: Arheološka najdišča Slovenije, 175, Ljubljana. ISTENIČ, J. 1999, Poetovio, zahodna grobišča I. Grobne celote iz Deželnega muzeja Joanneuma v Gradcu / Poetovio, the Western Cemeteries I. Grave-Groups in the Landesmuseum Joanneum, Graz. – Katalogi in Monografije 32. ISTENIČ, J. 2000, Poetovio, zahodna grobišča II. Grobne celote iz Deželnega muzeja Joanneuma v Gradcu / Poetovio, the Western Cemeteries II. Grave-Groups in the Landesmuseum Joanneum, Graz. – Katalogi in Monografije 33. KRAJŠEK, J., P. STERGAR 2008, Keramika s svetiščnega območja v Podkraju pri Hrastniku (The pottery material from the Roman sanctuary area at Podkraj near Hrastnik). – Arheološki vestnik 59, 245–278. LAZAR, I. 2003, Rimsko steklo Slovenije / The Roman glass of Slovenia. – Opera Instituti Archaeologici Sloveniae 7. MÜLLNER, A. 1879 (repr. 2013), Emona. – Ljubljana. PLESNIČAR-GEC, L. 1972, Severno emonsko grobišče / The northern necropolis of Emona. – Katalogi in Monografije 8. PLESNIČAR-GEC L. 1977, Keramika emonskih nekropol / The Pottery of Emona Necopolises. – Disertationes et mo-nographiae 20. PLESTENJAK et al. 2014 = A. PLESTENJAK, M. HORVAT, T. VERBIČ, I. BEKLJANOV ZIDANŠEK, R. ERJAVEC, B. TOŠKAN, T. LESKOVAR, T. TOLAR, J. LORBER 2014, Končno poročilo o arheoloških raziskavah na lokaciji Blago­vica – grobišče. – Poročilo / Report, Arheološki konzorcij (UL FF, IZA ZRC SAZU, Arhej d.o.o.) (neobjavljeno / unpublished). RICKMAN, G. 1971, Roman Granaries and Store Buildings. – Cambridge. SAGADIN, M. 1977, Trasa magistralnega plinovoda – ar­heološki nadzor. – Poročilo / Report; ZVKDS OE Kranj (neobjavljeno / unpublished). SAGADIN, M. 2004, Arheološka preteklost občine Lukovica. – V / In: Zbornik občine Lukovica 2004, 41–50, Ljubljana, Lukovica. STRAŽAR, S. 1985, Rimska nekropola v Blagovici. – V / In: P. Stražar, Črni graben. Od Prevoj do Trojan, 81–84, Lukovica. ŠAŠEL, J. 1954, Arheološka podoba Mengša. Rimska doba. – V / In: Mengeški zbornik, 800 let Mengša, 13–17, Mengeš.ŠAŠEL KOS, M. 1997, Pošta v antiki. – V / In: A. Hozjan (ur. / ed.), Pošta na slovenskih tleh, 18–43, Maribor. VERBIČ T. 2014, Analiza žlindre. – V / In: Plestenjak et al. 2014, 170–171. ZUPANČIČ, M. 1979, Blagovica. – Varstvo spomenikov 22, 278–281. Ana Plestenjak Arhej, d.o.o, Drožanjska 23, SI-8290 Sevnica ana@plestenjak.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 249–293 ATRANS – TROJANE Janja ŽELEZNIKAR, Julijana VISOČNIK Izvleček Naselje Atrans je ležalo ob itinerarski cesti na prelazu Trojane, še v agru Celeje v Noriku in v bližini meje z Italijo. Tu je delovala carinska postaja v sklopu ilirske carine, katere uslužbenci so izpričani na epigrafskih spomenikih. V prispevku smo analizirali pisne vire, kamnite spomenike, slikovno dokumentacijo, predvsem pa zapuščino raziskovalca Walterja Schmida in načrte iz zaščitnih izkopavanj Mirine Zupančič. Poskušali smo rekonstruirati topografsko sliko rimske nasel­bine ter velikost in položaj dvanajstih rimskih objektov in njihovo funkcijo. Drobne najdbe, ki so datirane od 1. do 4. st., ne izvirajo iz dobro dokumentiranih kontekstov. Ključne besede: Norik, Trojane, Atrans, rimska doba, naselbina, poštna postaja, ilirska carina, beneficiarska postaja, epigrafski spomeniki, arhivski viri, Walter Schmid Abstract The settlement of Atrans on the Trojane Pass was situated on an itinerary road, still in the territory of Celeia in the province of Noricum, but close to the Italian border. It was a customs station within the framework of the Illyrian cus­toms, and its officials are attested in epigraphic sources. This contribution presents an analysis of written sources, stone monuments, and images, with the emphasis on the heritage of the researcher Walter Schmid and the plans from a rescue excavation led by Mirina Zupančič. The authors attempt to reconstruct the topography of the Roman settlement, includ­ing the size, position, and function of twelve Roman buildings. Dated between the 1st and 4th centuries, small finds do not originate from well-documented contexts. Keywords: Noricum, Trojane, Atrans, Roman period, settlement, postal station, Illyrian customs, statio of the ben­eficiarii, epigraphic monuments, archives, Walter Schmid https://doi.org/10.3986/9789610502586_14 Sl. 1: Trojane. Pogled iz zraka proti zahodu. Fig. 1: Trojane. Aerial view from the east. (Foto / Photo: Primož Hieng) LEGA IN LOCIRANJE Trojane se razprostirajo na nadmorski višini okoli 563 m. Ležijo na sedlu hribovja, ki se spušča od vzhodnih Karavank do savskega kanjona ter ločuje Ljubljansko odCeljske kotline, Kranjsko od Štajerske. Naselbina (tudi antična) je zrasla na ozkem in podolgovatem platoju (sl. 1). Omejena je s tremi potoki, ki se stekajo v tri do-line: Radomlja teče v Črni graben, Orehovica v dolino Izlak, Bolska proti Vranskemu.1 V 1. st. n. št. (ohranjene drobne najdbe), še bolj verjetno pa v 2. st. (epigrafski spomeniki), je ob cesti Emona–Celeja–Petoviona zraslo antično naselje Atrans, prva cestna postojanka na teritoriju Celeje. Zahodno od Atransa je potekala meja med Italijo in provinco Norik, skozenj pa via publica – od Akvileje do Emone in preko Trojan naprej do Celeje (Celeia) in Petovione (Poetovio).2 Geografski viri navajajo Atrans kot poštno postajo: mansio, ki pa je v vsakem izmed njih zapisana nekoliko drugače. Pod imenom Adrante mansio m. p. XXV se pojavlja v Antoninskem itinerariju (129.3), kot Mansio Hadrante s pripisom fines Italiae et Norici v Burdigal­ 1 Orožen Adamič, Perko, Kladnik et al. 1995, 397. Šašel, Bolta 1975; Horvat 1999, 253–254; Šašel Kos POSITION AND LOCALIZATION The settlement of Trojane lies about 563 m above sea level, on a pass in the hills descending from the eastern Karavanke Mountains towards the Sava canyon, separat­ing the Ljubljana and Celje Basins, as well as the regionsof Kranjska and Štajerska. The settlement (including the Roman settlement) was built on a narrow, elongated plateau (Fig. 1). It is bounded by three creeks that flow into three valleys: the Radomlja into the valley of Črni graben, the Orehovica into the valley of Izlake, and the Bolska towards Vransko.1 The Roman settlement of Atrans was built on the Emona – Celeia – Poetovio road in either the 1st (small finds) or – even more likely – 2nd century AD (epigraphic monuments). It was the first road station in the territory of Celeia. The border between Italy and the province of Noricum ran west of Atrans, and the via publica from Aquileia to Emona and across Trojane to Celeia and Poetovio ran through the settlement.2 Atrans is mentioned in geographical sources as a postal station – mansio. The spelling of the name, how­ 1 Orožen Adamič, Perko, Kladnik et al. 1995, 397. 2 Šašel, Bolta 1975; Horvat 1999, 227, 229; Šašel Kos 1999, 238–240. 1999, 238–240. skem itinerariju (560.9) in kot Adrante na Tabuli Peu­tingeriani (IV, 2).3 Ime Atrans naj bi bilo predrimsko.4 V starih listinah je ime sodobnega kraja Trojane zapisano leta 1229 kot latere ville Troye, okoli leta 1400 am Troyn, leta 1466 vber 5 den Troyan in leta 1496 Troyan. Avtor nepodpisanega članka v časniku Slovenec leta 1939 pojasnjuje, da je vzhodni del antičnega Atransa 6 dobil ime “v Zideh”. Arheološke ostaline predrimske dobe s Trojan in bližnje okolice, niso znane. ZGODOVINA RAZISKAV Trojane so bile skozi stoletja večkrat predmet raz­iskav. Do zdaj pa še niso dobile celovitega pregleda in objave. Gradnja nove ceste preko trojanskega klanca in Trojan v šestdesetih letih 20. stoletja je terjala rušenje šte­vilnih lesenih in tudi zidanih objektov vzdolž nekdanje trase. Naselitvena ravnica je spremenila svojo podobo, drugačen je potek ceste, ki se je iz vasi navezovala na staro cesto. Ta je, v nasprotju z današnjo, tekla po levem bregu Bolske. Trojančani so morali zaradi arheoloških ostalin večkrat spreminjati načrte za novogradnje. Najstarejše omembe najdb s Trojan so iz 16. stolet­ja. Alfonz Müllner poroča, da je že Auguštin Tyffernus pisal o najdbah različnih starih kamnov z napisi.7 Leta 1507 je bila najdena ara (kat. št. 3; CIL III 5120), ki jo je postavil Gaj Antonij Julijan. Janez Vajkard Valvasor omenja zidovje, ruševine ter izredno “množico rimskih novcev, ki še danes ležijo v zemlji”.8 Müllner navaja, da sta Richard Pococke in Jeremiah Milles, angleška antikvarja in popotnika, na Trojanah leta 1752 videla tri napisne kamne.9 Ker gre samo za omembo, kamnov ni mogoče 3 Šašel 1975, 75–78. 4 Repanšek 2016, 107–108. 5 Kos 1975; Snoj 2009 in Repanšek 2016, 107–108 (razla­gajo ime Trojane iz predromanskega substrata); Torkar 2012, 698–699 (slovanski izvor imena Trojane). 6 Besedilo z naslovom “Zgodovinske Trojane” je istega leta izšlo v treh revijah: Slovenec 1939; Slovenski list 1939; Glasilo 1939 (“... ker so se tam nahajali še ostanki obzidja; v dolini pod zidom je nastala vas Podzidom; zahodno naselbino pa so imenovali Tvrjani, Trjani, Trjanje (prim. vtrjenje) zato, ker so stanovali v nekdanji rimski trdnjavi; od tod je nastalo sedanje popačeno knjižno ime Trojane.”). V zapuščini J. Šašla (glej poglavje: Pisni viri in dokumentacija, vir 17) je fotoko­pija članka s Šašlovim pripisom, da je avtor znani in prizna­ni slovenski arhitekt, domačin z Zavrha pri Trojanah, Niko Kralj (prim. vir 10). 7 Müllner 1879, 82. 8 Valvasor 1689a, 124; 1689b, 266–268. Valvasor naj bi si prav iz trojanskih novcev osnoval svojo preko 8000 tisoč ko­sov obsegajočo zasebno numizmatično zbirko (Pegan 2016, 56, op. 121 in 122). Müllner 1879, 83. 9 Müllner 1879, 83–84; to notico je povzel tudi Stražar 1995, 84. ever, differs slightly between the sources. It is mentioned as Adrante mansio m. p. XXV in the Antonine Itinerary (129.3), as Mansio Hadrante with the comment fines Italiae et Norici in the Bordeaux Itinerary (560.9), and as Adrante in Tabula Peutingeriana IV 2.3 The name Atrans is probably pre-Roman.4 The name of the modern settlement of Trojane appears in old docu­ments in 1229 as latere ville Troye, around 1400 as am Troyn, 1466 as vber den Troyan, and in 1496 as Troyan. In 1939, an unsigned article in the Slovenec news­paper claimed that the eastern part of the Roman Atrans was called V Zideh (“between the walls”).6 No pre-Roman archaeological remains are known from Trojane and its immediate surroundings. RESEARCH HISTORY While Trojane has been the subject of research sev­eral times during the centuries, there has been no com­prehensive overview and Reference. In the 1960s, when a new road up the hill towards Trojane and through the village itself was constructed, many wooden and masonry buildings along the old course of the road were demol­ished. The flat area of the settlement was transformed and the road that came from the village and joined the old road changed its course. Unlike the present-day road, the old road ran along the left bank of the Bolska. Plans for new buildings were often changed due to archaeological remains. The earliest records of finds from Trojane date to the 16th century. According to Müllner, Augustinus Tyffernus already wrote about the discoveries of various old stones with inscriptions.7 In 1507, an altar (Cat. no. 3; CIL III 5120) by Gaius Antonius Iulianus was found. Janez Vajkard Valvasor mentioned walls, debris, and an extraordinary “number of Roman coins which have re­mained in the soil until this day”.8 According to Müllner, 3 Šašel 1975, 75–78. 4 Repanšek 2016, 107–108. 5 Kos 1975; Snoj 2009, 439–440 and Repanšek 2016, 107– 108 (explain the name Trojane from a pre-Roman substrate); Torkar 2012, 698–699 (Slavic origin of the name Trojane). 6 Slovenec 1939; Slovenski list 1939; Glasilo 1939 (the same text: “... since there were still remains of walls in the val­ley under the wall, the village was called Podzidom [“under the wall”]; while the western settlement was called Tvrjani, Trjai, Trjanje (cf. vtrjanje [“fortification”], because they lived in a former Roman fortress; this is the origin of the present-day disfigured standard name of Trojane.”). Šašel’s archives (see Source 17) contain a photocopy of the article with Šašel’s comment saying that the author is probably the native of Za­vrh near Trojane, well-known Slovenian architect Niko Kralj (cf. Source 10). 7 Müllner 1879, 82. 8 Valvasor 1689a, 124; 1689b, 266–268. The coins from Trojane were supposedly the basis of Valvasor’s private nu­ identificirati. Müllner piše, da je Valentin Vodnik leta 1808 našel več vzidanih napisov na Trojanah.10 Vodnik naj bi videl še dva napisna kamna s Trojan (kat. št. 1, 4), ki sta bila vzidana v Attemsovo palačo v Gorici, o čemer poročata Rechfeld in Hitzinger.11 Največ podatkov o najdbah in različnih izkopa­valnih posegih na Trojanah pa imamo od sredine 19. stoletja naprej (tab. 1; sl. 2). PROSTORSKA UMESTITEV STAVB IN DRUGIH NAJDB V graškem muzeju Joanneum (v nadaljevanju: UMJG) odkrite najdbe, terenski dnevniki in nekaj do-kumentacije Walterja Schmida, prvega strokovnega ra­ziskovalca Trojan, na Inštitutu za arheologijo ZRC SAZUpregledani arhivi Jaroslava Šašla, arheološko gradivo z izkopavanj Mirine Zupančič iz Medobčinskega muzeja Kamnik (MMK) ter novejše raziskave na Trojanah in v okolici nam odstirajo nekatere tančice v poplavi ugank antičnega Atransa.12 Iz Schmidovih terenskih dnevnikov je razvidno, da je na Trojanah kopal kar dvakrat in sicer leta 1941 in jeseni leta 1942. Skiciral, opisal ali evidenti­ral je skupno osem rimskodobnih objektov.13 Od druge polovice 19. stoletja do leta 2018 je bilo tako ali drugače prekopano območje znotraj ožjega dela Trojan, ki meri približno 500 × 80 m: od nekdanjih Konjškovih njiv (danes Trojane 22) na zahodu, do Hotela Trojane (Trojane 27) na vzhodu (sl. 2; 3: 2–13). Skupno lahko do tega trenutka na Trojanah z veliko verjetnostjo umestimo dvanajst antičnih zgradb (sl. 20; 22). 10 Müllner 1879, 83. 11 Rechfeld 1848, 89; Hitzinger 1855. 12 Za podrobna pojasnila o arhivskih virih glej poglavje: Pisni viri in dokumentacija. 13 Viri 1–3 in 8; Schmid 1941, 5; Jutro 1941. English antiquaries and travellers Richard Pococke and Jeremiah Milles saw three inscription stones in Trojane in 1752.9 Based on this record alone, the stones cannot be identified. According to Müllner, in 1808, Vodnik found several inscriptions built-in the houses of Trojane.10 Vodnik apparently saw two other inscription stones from Trojane, built in the Attems mansion in Gorizia (Cat. nos. 1, 4), as reported by Rechfeld and Hitzinger.11 The largest amount of information on the finds and various excavations in Trojane, however, has been available since the 19th century onwards (Tab. 1; Fig. 2). POSITION OF BUILDINGS Some of the mysteries of the Roman Atrans have been unveiled by the finds discovered in the Joanneum Museum in Graz, by Walter Schmid’s field journals anddocumentation, by the inspected archives of Jaroslav Šašel, by the material from the archaeological excavations led by Mirina Zupančič, and by more recent surveys in Trojane and its surroundings.12 Walter Schmid’s field journals reveal that he exca­vated in Trojane twice; in 1941 and again in the autumn of 1942. He recorded and described eight Roman buildings and sketched six of them.13 Between the second half of the 19th century and 2018, an area of about 500 × 80 m in the central part of Trojane was excavated in one way or another (Fig. 2; 3: 1–13). Up to this moment, twelve Roman buildings can be located with high probability (Figs. 20; 22). mismatic collection of more than 8000 pieces (Pegan 2016, 56, notes 121, 122). Müllner 1879, 83. 9 Müllner 1879, 83–84; the note was quoted by Stražar 1995, 84. 10 Müllner 1879, 83. 11 Rechfeld 1848, 89; Hitzinger 1855. 12 For details see below: Written sources and documentation. 13 Sources 1–3 and 8; Schmid 1941, 5; Jutro 1941. KATALOG ARHEOLOŠKIH OBMOČIJ [CATALOGUE OF ARCHAEOLOGICAL AREAS] (Sl. / Figs. 2; 3) V besedilu uporabljene okrajšave: ARS = Arhiv Republike Slovenije IzA ZRC SAZU = Inštitut za arheologijo ZRC SAZU MMK = Medobčinski muzej Kamnik NMS = Narodni muzej Slovenije PMC = Pokrajinski muzej Celje UMJG = Universalmuseum Joanneum, Graz ZVKDS = Zavod za varstvo kulturne dediščine Slovenije Območje (Sl. 2) Area (Fig. 2) Raziskovalec Excavator Leto Year Nekdanje ime Former name Nekdanji lastnik / Owner in the time of excavations Današnja hišna ali parc. št. (k. o. Trojane) / Present-day house or land plot no. Objava, poročila, vir / Publication, reports, sources 6 Dornik (župnik / parish priest, Šentgotard) 1846 Stojčev vrt Štefan Brvar Trojane 10 Müllner 1879, 83; Müllner 1878 5 Konjšek (župan / mayor) 1866 Konjškovi vrtovi Konjšek Trojane 11 Deschmann 1866 6 Dornik? 1874 posestvo / property: Brvar (?) Štefan Brvar Trojane 10 Müllner 1878 6 Dornik? 1876 posestvo / property: Brvar (?) Štefan Brvar Trojane 10 Müllner 1879, 83 5, 6 Pečnik 1886 posestvi / properties: Konjšek in Brvar (pri Stojc) Štefan Brvar in / and Konjšek Trojane 11 in / and 10 Pečnik 1904; Orožen 1902, 20; Rutar 1890, 121–122. Viri / Sources: 19; 20 1 Pečnik 1889 Konjškove njive Konjšek zahodno od glavne ceste / west of the main road Viri / Sources: 22; 23 5, 6 – 1890 – – Rutar 1890, 121; Prokop 1891 5, 6 – 1891 – – Rutar 1890, 121; Rutar 1891, 209; Dimitz 1874, 56, op. / n. 5 6 Štefan Brvar 1936 posestvo / property:Štefan Brvar (p. d. Stojc) Štefan Brvar Trojane 10 Slovenec 1939, 10, 12; Ložar 1937; Zupančič et al. 2003, 12 7 Schmid 1941 med Konjškom in Ržunom Konjšek, Ržun in drugi / and others od hiš Trojane 11 do 29 / from the houses Trojane 11 to 29 Schmid 1941; Jutro 1941; Viri / Sources: 1; 6; 9 3 Schmid 1941 Zadružni dom Trojane 6 Šašel 1954a, 160. Vir / Source: 2 9 Schmid 1941 Ržunov vrt Trojane 9 in 29 Jutro 1941; Šašel 1954a, 160; Zupančič 1970; Zupančič 1971, 221; Zupančič 1979, 17. Viri / Sources: 1; 8; 15 13 Schmid 1941 Hotel Trojane Trojane 27 Ustni podatek / Personal information: Janez Ocvirk. 2 Schmid 1942 – od hiše Trojane 6 do 11 / from the houses Trojane 6 to 11 Viri / Sources: 2; 15 15 Schmid 1942 Učak Šašel 1954a, 160–161; Vir / Source: 2; 4; 5; 13 6 nemški ujetniki German PoWs 1945 posestvo / property:Štefan Brvar (p. d. Stojc) Štefan Brvar Trojane 10 Ustni podatek / Personal information: M. Zupančič. Območje (Sl. 2) Area (Fig. 2) Raziskovalec Excavator Leto Year Nekdanje ime Former name Nekdanji lastnik / Owner in the time of excavations Današnja hišna ali parc. št. (k. o. Trojane) / Present-day house or land plot no. Objava, poročila, vir / Publication, reports, sources 6 Štefan Brvar 1949 posestvo / property:Štefan Brvar (p. d. Stojc) Štefan Brvar Trojane 10 Ustni podatek / Personal information: M. Zupančič. 4 – 1949 posestvo / property: Franc Brvar Franc Brvar Trojane 7 Šašel 1954a, 160. Vir / Source: 18/2 6 Štefan Brvar 1950 posestvo / property:Štefan Brvar (p. d. Stojc) Štefan Brvar Trojane 10 Klemenc 1951, 124–129 3 Šašel (topografija / field survey) 1953 Zadružni dom Trojane 6 Šašel 1954a, 160 11 Šašel (topografija / field survey) 1953 (p. d. Matija) Ivan Izlakar Trojane 8 Šašel 1954a, 160 6 Štefan Brvar 1957 posestvo / property:Štefan Brvar (p. d. Stojc) Štefan Brvar Trojane 10 Ustni podatek / Personal information: M. Zupančič 6 Štefan Brvar 1957 – 1960 posestvo / property:Štefan Brvar (p. d. Stojc) Štefan Brvar Trojane 10 Ustni podatek / Personal information: M. Zupančič 16–18 Bolta 1959 trasa “stare” ceste / “old” road Ljubljana–Celje Bolta 1959; 1960; Zupančič 1979, 17. Vir / Source 21 10 Zupančič 1966 parcela / property: Valentina, Majda Hribar Zdravko Brvar Trojane 29 (parc. / plot 571/2) Cvikl Zupančič 1966; Zupančič 1966;Šašel 1966 9 Zupančič 1970 Ržunov vrt Anica Pungartnik, Majda Hribar med hišama Trojane 9 in 29 / between houses Trojane 9 and 29 Zupančič 1970; Zupančič 1971, 221; Zupančič 1979, 17. Viri / Sources: 1; 8; 15. 8 Zupančič 1975 parcela / plot: Zdravko Brvar parc. / plot 4/1, 4/2 Zupančič 1977; Zupančič 1979, 17; Zupan-čič et al. 2003, 16 13 Zupančič 1976 Hotel Trojane Trojane 27 Ustni podatek / Personal information: M. Zupančič 20 Ogrin 1996 parc. / plot 468/1, 469/1, 521/1, 518 Ogrin 1996 20 Mušič 1997 Mušič 1997 19 Kajfež 1997 trasa “stare” ceste / “old” road Ljubljana–Celje Kajfež 1997 12 Josipovič 2003 posestvo / property: Brvar parc. 9/1, 5/2, 5/3, 574/1 Josipovič 2003 13a Košir 2018 Robert Smrkolj Trojane 25 (parc. 560/5) Košir 2018 Tab. 1: Posegi v arheološke plasti na Trojanah od 19. stoletja do danes. Tab. 1: Excavations in Trojane from the 19th century until today. Sl. 2: Trojane. Raziskana in delno raziskana območja (1–13a). Glej poglavje Katalog arheoloških območij. Fig. 2: Trojane. Investigated and partly investigated areas (Areas 1–13a). See the Catalogue of archaeological areas. (Vir za podlago / Map source: DOF 2014_050 © Geodetska uprava RS) ŠTEVILČENJE ANTIČNIH OBJEKTOV Pomembno je opozoriti, da je Schmid – že med delom in iz nam neznanega razloga – preštevilčil dve od stavb, odkritih v prvi sezoni (deloma je to vidno v prvem dnevniku, leta 1941; vir 1). V našem besedilu smo prevzeli to naknadno preštevilčenje (= aktualno številčenje). Schmid je prvotno hišo I preštevilčil v III in prvotno hišo III (hišo s polkrožnim zaključkom) je preštevilčil v I.14 Schmidovi hiši II in IV (ter vse nadaljnje) sprememb v številčenju niso bile deležne, naše aktualno številčenje zanje je tako identično izvornemu. Za še pozneje odkrite stavbe (IX–XII) smo številčenje zvezno nadaljevali (prim. sl. 20; 22). Pri opisu posameznih virov in slik na problem oštevilčenja opozorimo. Antične stavbe V (aktualna št.; hiša s hipokavstom) Schmid ni nikjer ekplicitno opisal. V izkopavalni etapi leta leta 1942 je označil najdene tlorise od VI do VIII, zato sklepamo, da je hišo s skiciranim hipokavstom (odkrito sicer že leta 1936) upošteval kot peto (tudi glede na skico – vir 8; sl. 15). Arheološka dokumentacija je borna. Risbe treh natančneje izmerjenih objektov je za Schmida narisal Anton Farčnik (sl. 10; vir 3), ostale objekte poznamo samo iz skic v Schmidovih dnevnikih in dokumentaciji. Šašel je v graškem muzeju ob pregledu Schmidove dokumentacije prerisal Farčnikove risbe; ob terenskem pregledu leta 1961 je s pomočjo domačinov/informatorjev izdelal pozicijsko skico današnjih objektov, označil območja Schmidovih raziskav in dopisal komentarje (sl. 5).15 Šašel je leta 1961 uspel locirati ostanke stavbe (hiša IX; vir 13; glej območje 3 v nadaljevanju), ki naj bi jo že predhodno zaznal tudi Schmid. Mirina Zupančič je na Trojanah našla obrise temeljev več stavb,16 iz objav in dokumentacije (vir 25) lahko definiramo štiri: hišo I (= ponovno odkrita Schmidova hiša s polkrožnim zaključkom) in novo odkrite hiše X, XI, XII (prim. sl. 16; 20; 22). 14 Schmidovo preštevilčenje je v našem besedilu opazno npr. pri nekaterih reproduciranih skicah: dve hiši na eni od skic (sl. 14) nosita prvotno številčenje, na drugh skicah (prim. sl. 10; 13) je Schmid uporabil že spremenjeno številčenje. 15 Viri 15; 16 in 17. 16 Zupančič 1971. naselbina / settlement grob, grobiš~e / grave, cemetery napis (sekundarno najdiš~e) / terenski pregled / yeld survey cesta / road inscription (secondary position) Sl. 3: Arheološka območja na Trojanah (1–13a) in v okolici (14–20). Nahajališča vzidanih kamnitih spomenikov (21–24). Fig. 3: Archaeological areas in Trojane (1–13a) and its vicinity (14–20). Locations of built-in stone monuments (21–24). (Vir za podlago / Map source: DTK 25, 1998 © Geodetska uprava RS) 1–13a – Trojane (glej sl. 2 / see Fig. 2); 14, 15 – Učak; 16,17,19 – cesta / road Ljubljana–Celje; 18 – grobovi / graves; 20 – V Zideh; 21 – Šentožbolt; 22 – Sv. Mohor; 23 – V Zideh 4; 24 – Šentgotard Območje 1 (sl. 2: 1; 3: 1) Lokacija: Konjškove njive (danes Trojane 22). Zgodovina odkritij in raziskav: Pečnik je že leta 1889 v pismu Centralni komisiji na Dunaju pisal, da je na Konjškovih njivah našel žgane rim- ske grobove. Lokacijo je vrisal v karto arheoloških najdišč Radeč in Zagorja (sl. 4). Lokacija leži ob cesti, tik pred zahodnim robom rimske naselbine. Literatura in viri: Pečnik 1904, 131. – Vir 22; vir 23. Sl. 4: Izsek iz Pečnikove karte arheoloških najdišč Radeče in Zagorje (1889), z rimsko cesto in vrisanimi arheološkimi loka­cijami na Učaku,Trojanah in Podzidu. Simbol ob št. 4 je znak za rimsko »grobišče z žganimi pokopi« na območju 1. (Vir 23). Fig. 4: Section from Pečnik’s map of the archaeological sites at Radeče and Zagorje (1889) showing the Roman road and sites in Učak, Trojane and Podzid. The sign beside No. 4 indicates Roman cremation graves in Area 1. (Source 23). (Arhiv /Archives: Arhiv RS) Sl. 5: Trojane. Skica J. Šašla (1953). Označeno: potek rimske ceste, južni rob naselbine, območje raziskav Walterja Šmida (S) leta 1941 in 1942 (s križci – ×), zidovje pri hiši Trojane 8 (A). Skica je usmerjena približno proti severu. (Vir 15). Fig. 5: Trojane. Sketch by Jaroslav Šašel (1953) showing the supposed course of the Roman road, the southern edge of the settle­ment, the areas that Walter Schmid (S) excavated in 1941 and 1942 (×) and the walls near the house at Trojane 8 (A). The sketch is oriented roughly north. (Source 15). (Arhiv / Archives: IzA ZRC SAZU) Območje 2 (sl. 2: 2) Lokacija: Od Konjškovega posestva (danes Trojane 11), preko območja nekdanjega gasilskega doma (Trojane 4) do Zadružnega doma (Trojane 6). Zgodovina odkritij in raziskav: Od 1. do 21. oktobra 1942 je bil Schmid drugič na Trojanah. Našel je temelje treh antičnih hiš (VI, VII in VIII; sl. 6).17 Šašel je leta 1953, ob topografiji terena, po pripovedovanju domačinov na skici (sl. 5)18 označil severni del Trojan (med hišami št. 4 in 6) kot teren, ki ga je kopal Schmid tega leta. Opise stavb in skice povzemamo po Schmi­dovem dnevniku.19 Hiše VI–VIII (sl. 20; 22): – Hiša VI (dol. 39,10 m; šir. 13,60 m). Drobne najdbe:20 pečatna oljenka, žebelj, keramika zelene barve; ognjišče šir. 1 m, velika ogrevana soba, tubuli; maltni estrih, ruševine, apnenec, žganinski material. Objekt naj bi bil uničen v požaru.21 – Med stavbama VI in VII je v dnevniku narisano nekakšno popločenje oz. tlakovane poti (sl. 6). Sl. 6: Trojane, območje 2. Skica iz Schmidovega dnevnika (1942): pozicije hiš VI, VII in VIII. (Vir 2, str. 22). Fig. 6: Trojane, Area 2. Sketch from Walter Schmid’s field journal (1942) showing the locations of Houses I, VII, and VIII. (Source 2). (Arhiv /Archives: Universalmuseum Joanneum Graz) – Hiša VII (dol. 39 m; šir. 20,40 m) – po Schmidu naj bi šlo za hlev ali gospodarsko poslopje oz. skladišče za vozove. Zidovi široki 60 cm, približno 35 cm pod hodno površino. Ob severnem zidu je bil najden novec Vespazi­jana, zahodni zid je bil 1 m debel, 20–30 cm visok temelj zidu, ki je bil ponekod ohranjen samo v eni vrsti, najdene so bile številne črepinje.22 – Hiša VIII (dol. 27,70 m; šir. 17 m) – po Schmidu hlev. Severni zid 1,50 m širok, zahodni 1 m, vzhodni zid 80 cm, 35 cm pod nivojem se je pojavila laporna osnova, najdena je baza za steber oz. na pol prelomljen podpornik za streho. Geološka osnova: modri lapor; ognjišče in najdba dveh novcev in črepinj v bližini. Pravi, da je bila hiša uničena v požaru. Govori še o strešnikih in imbreksih ter novcu Nerona in Vespazijana.23 Viri: vir 2 (str. 14–24); vir 15. Območje 3 (sl. 2: 3) Lokacija: Zadružni dom (Trojane 6). Zgodovina odkritij in raziskav: – Hiša IX: Šašel je v času gradnje Zadružnega doma leta 1953 izrisal in umestil dva profila (sl. 7). Zemljišče je nagnjeno proti severu, naravna osnova je lapor. V treh profilih je Šašel definiral več kulturnih plasti, estrih in dve ruševinski plasti. Zaznal je močno žganinsko plast, ki je bila verjetno posledica nasilnega uničenja. Domačin Ivan Baloh je Šašlu omenil, da je že Schmid na tem mestu izkopal objekt kvadratne oblike (približne velikosti 15 × 15 m).24 Na 17 Vir 2. 18 Vir 15. 19 Vir 2. 20 Med arheološkim gradivom, ki je bilo leta 2013 predano v MMK, ni predmetov iz hiš VI do VIII. Opisi najdb so povzeti po Schmidovem terenskem dnevniku (vir 2). 21 Vir 2, str. 22. 22 Vir 2, str. 24. 23 Vir 2, str. 15–16. 24 Vir 13. Sl. 7: Trojane, območje 3. Profila gradbene jame leta 1953. Ostanki temeljev hiše IX (pomanjšano po Šašel 1954a, 161, 162, sl. 3a,b). Fig. 7: Trojane, Area 3. Cross sections of the excavation in 1953. The foundations of House IX (in reduced scale from Šašel 1954a, 161, 162, Fig. 3a,b). poskusu rekonstrukcije naselja smo objekt označili z rimsko številko IX (sl. 20; 22). Ostanek temelja zidu, ki ga je v profilih videl Šašel, je ležal na globini 1,5 m. Na globini 90 cm so odkrili tlak rimske ceste. Drobne najdbe predstavlja keramika, okrašena z metličenjem in odlomki tere sigilate, malta in rdeč ožgan stenski omet. Literatura in viri: Šašel 1954a, 160. – Vir 13. Območje 4 (sl. 2: 4) Lokacija: Trojane 7. Zgodovina odkritij in raziskav: Leta 1949 so samo na vzhodni polovici poslopja Trojane 7 (takratni lastnik Franc Brvar) odkopali več zidov, lukenj za bruna in drobne arheološke predmete. Severno od hiše Trojane 7, kjer se teren spušča v dolino Bolske, se je videl usek stare poti – domnevno rimske ceste. Na tem območju so bili občasno najdeni posamični rimski novci. Območje po doslej znanih najdbah predstavlja vzhodni rob rimske naselbine (sl. 22). Literatura in viri: Šašel 1954a, 160. – Vir 13, vir 18/2. Območje 5 (sl. 2: 5) Lokacija: Konjškovo posestvo, danes Trojane 11. Zgodovina odkritij in raziskav: – Lastnik gostilne Konjšek, tedanji župan, naj bi leta 1866 poklonil Deželnemu muzeju Avgustov srebrnik, ki je bil verjetno najden na njegovem posestvu.25 – Leta 1886 je na Trojanah kopal Pečnik,26 ki nikjer eksplicitno ne navaja točne lokacije izkopavanj. Simon Rutar piše, da je kopal na vrtovih Franca Konška (pri Konjšku) ter na vrtu Štefana Brvarja (pri Stojc).27 Tu je našel: “…ruševine hiš z lepimi freskami, kopalnice, več napisnih kamnov, novcev na tisoče, med njimi so tudi srebrni …”. – Leta 1961 je Konjšek urejal svoj gostinski in delovni prostor, pri čemer so pod staro mizarsko delavnico odkrili 2 m debel rimski kontrafor,28 ki ga je 1961 v prostor umestil Šašel (sl. 8). – V zadnjih letih, ko se je današnje gostinsko podjetje Trojane s parkirnimi prostori močno razširilo proti zahodu, je Milan Sagadin (ZVKDS, OE Kranj) našel lončene tubule, ki pričajo, da so tudi zahodno od Konjškove gostilne bili s centralno kurjavo ogrevani objekti29 (za lokacijo današnje hiše Trojane 11 prim. sl. 22). Gre verjetno za objekte južno od ceste – za zahodni rob rimskega Atransa. Literatura in viri: Deschmann 1866; Pečnik 1904; Orožen 1902, 20; Rutar 1890. – Vir 12; vir 14. Sl. 8: Trojane, območje 5. Skica J. Šašla: antični kontrafor (rdeča oznaka), odkrit leta 1961. Skica je usmerjena proti jugu. (Vir 14). Fig. 8: Trojane, Area 5. Sketch by Jaroslav Šašel showing the Roman retaining wall with a buttress (red) found in 1961. The sketch is oriented south. (Source 14). (Arhiv / Archives: IzA ZRC SAZU) 25 Deschmann 1866. 26 Pečnik 1904; Orožen 1902, 20. 27 Rutar 1890, 121–122. 28 Vira 12 in 14. 29 Ustni podatek Milan Sagadin (ZVKDS, OE Kranj). Območje 6 (sl. 2: 6) Lokacija: Brvarjevo posestvo (pri Stojc), danes Trojane 10. Zgodovina odkritij in raziskav: – Prva nestrokovna izkopavanja, brez objave najdb in rezultatov, so potekala leta 1846. Župnik Dornik iz Šent­gotarda je kopal na Stojčevem vrtu. Našel naj bi predvsem novce, zidovje, vmes nekaj zoglenelega žitnega zrna, in napisni kamen, ki je ostal V Zideh.30 – Leta 1874 je A. Müllner odkril na Trojanah šest napisnih kamnov,31 leta 1876 pa še dva32 – verjetno vse na Brvarjevem posestvu (kat. št. 6, 7, 9 in prvega od dveh fragmentov gradbenega napisa kat. št. 833). Vsaj dveh od teh kamnov ni bilo mogoče identificirati, za enega pa je že Müllner zapisal, da je izgubljen. –Prokop in Rutar omenjata novce,34 predvsem iz druge polovice 3. st., Dimitz pa opeke z žigom II. italske legije.35 – Leta 1886 je na Trojanah kopal Pečnik,36 vendar nikjer ne omenja natančne lokacije. Rutar piše, da je kopal na vrtovih Franca Konjška (pri Konjšku, glej območje 5) in na vrtu Štefana Brvarja (pri Stojc).37 Pečnik naj bi izkopal: “ ... rimsko kopališče z marmornim tlakom, pod drvarnico zoglenelo pšenico in ajdo, itd.”38 V pismu Dežmanu Pečnik piše o preiskovanju terena po njivah okoli Trojan in Učaka. Pravi, da so delavci našli lepe “malarije”, da se zidovi nahajajo “čez 1 klaftro”39 globoko, da so našli kos brona, ki tehta več kot 3 kg, in da je Stojc na svojem vrtu našel 8 cm velikega “malika”, ki ga je dal vodovodarju iz Motnika.40 – Hiša V: – Maja 1936 so kopali temelje za svinjak (parcela 5/2 k. o. Trojane) na parceli Štefana Brvarja, (p. d. Stojc, takrat Trojane 5). Našli so hipokavst, fragmente tegul z žigi, omet, keramiko in novce. Predmete hrani NMS.41 Ob prihodu Ložarja na terenu ni ostalo več dosti, le dva dobro ohranjena oboka (sl. 9).42 En lok je bil visok 0,70 m, drugi pa 0,60 m. Širina lokov je bila 0,60 m, kar ustreza dvema rimskima čevljema.43 Verjetno gre za ostanke prefurnijev hipokavsta. Objekt s hipokavstom je v prostor zelo nenatančno umestil Schmid (sl. 15; stavbe niso oštevilčene).44 Njena pozicija pa je dobro opisana pri Ložarju. Pripisali smo ji rimsko številko V (sl. 15; 20; 22). – Med letoma 1957 in 1960, ko so Brvarjevi gradili nov hlev, podatkov o novih najdbah ni.45 Pri kopanju temeljev za svinjak je bil odkrit še relief z grifonom, ki naj bi bil del nagrobne arhitekture (kat. št. 6).46 – Med letoma 1945 in 1949 so pri adaptaciji hiše Trojane št. 10 odkrili zidove in luknje za lesene stebre.47 – Ekskurzija arheologov in zgodovinarjev ljubljanske univerze je na svojem obisku Trojan 23. marca 1950 na zemljišču Jožefa–Štefana Brvarja kakšnih 50 cm pod rušo odkrila precejšen fragment večjega kamnitega spomenika. Gre za sestavni del (drugi fragment) že omenjenega gradbenega napisa – posvetilni napis Marka Avrelija in Lucija Vera (kat. št. 8).48 30 Müllner 1878; 1879, 81 ss. 31 Müllner 1874, 1–2. 32 Müllner 1878; 1879, 83. 33 Drugi fragment istega napisa je bil odkrit leta 1950, glej v nadaljevanju opisa območja 6. 34 Rutar 1890, 121; Prokop 1891. 35 Dimitz 1874, 56, op. 5. 36 Pečnik 1904; Orožen 1902, 20. 37 Rutar 1890. 38 Pečnik 1904; Rutar 1890, 121; Slovenec 1939; Jutro 1940. 39 1,896 m (Slovar slovenskega knjižnega jezika II, 324, Ljubljana 1975). 40 Vira 19 in 20. 41 Ložar 1937, 54; Slovenec 1939; Jutro 1941. Predmete hrani NMS: del pečatne oljenke FO(RTIS) (inv. št. R7130); – tegula s pečatom LEG II. (inv. št R7131); – tegula s pečatom VI VVV V (inv. št. R7132); – tegula z neprebranim pečatom (inv. št. R 7133); – kos stenskega ometa (inv. št R7134).42 Danes nista več ohranjena. 43 V članku neznanega avtorja “Zgodovinske Trojane” (Slovenec 1939) med drugim piše: “... Pri zidanju svinjaka so Stojčevi v Trojanah morali odnesti nekaj brega. V odkopu se je pokazala v oglje izpremenjena pšenica, znamenje da je ogenj uničil močno naselbino. Istotam so našli še tudi nekaj drugih predmetov, kakor igle, lončeno svetilko, ter neki kip brez glave; tudi dve plošči z napisi so našli leta 1862 na tem mestu. Ena se je izgubila neznano kam, drugo pa so položili na dno 14 m globokega vodnjaka, kjer je še danes. …”. Omenjeno je še, da je pred Stojčevo hišo Pečnik odkril kopališče, ki so ga leta 1936 pri gradnji nekega po­slopja ponovno odkrili. Isti članek je bil z enakim naslovom objavljen še v Buenos Airesu (Slovenski list 1939) ter v Clevelandu (Glasilo 1939). 44 Vir 8. 45 Ustni podatek M. Zupančič. 46 Zupančič et al. 2003, 12. Od leta 1967 je relief v MMK. 47 Ustni podatek M. Zupančič. 48 Klemenc 1951, 215–129; Šašel 1954b; RINMS, 306–323. Prvi fragment (ki ga je odkril Mülner) hrani MMK, drugi (najden leta 1950) pa je v PMC. Sl. 9: Trojane, območje 6. Oboka hipokavsta v antični hiši V. Fotografija je verjetno iz leta 1936. (Vir 24). Fig. 9: Trojane, Area 6. Two vaults of the hypocaust in House V. The photo was probably taken in 1936. (Source 24). (Arhiv /Arhives: Medobčinski muzej Kamnik) Literatura in viri: Dimitz 1874; Müllner 1878; 1879, 81 ss; Rutar 1890; Prokop 1891; Orožen 1902; Pečnik 1904; Ložar 1937, 54; Slovenec 1939; Jutro 1940; Jutro 1941; Klemenc 1951, 215–129; Šašel 1954b, 200–208; Bolta 1959, 130–131; Zupančič 1971, 215–222; RINMS, 306–323; Zupančič et al. 2003, 12. – Vir 8; vir 19; vir 20. Območje 7 (sl. 2: 7; 10–15) Lokacija: od Brvarjevega posestva (pri Stojc), danes Trojane 10 (prej št. 5) do hiše Trojane 7. Zgodovina odkritij in raziskav: – Pečnik je leta 1899 (vira 22 in 23) in 1904 (Pečnik 1904) poročal o rimskih stavbah v Trojanah (prim. sl. 4: točka št. 5) in vzidanih kamnih v neposredni okolici, ki dokazujejo “colno postajo Adrante”. Govori o ostankih stavb (s kopalnicami, slikanim ometom ipd. – o razkošnih stavbah). Leta 1886 je izkopal prostor z rdeče, črno in zeleno slikanim ometom in upodobitvijo rož, vendar ne omenja natančnejše lokacije.49 – Schmid je na tem delu Trojan raziskoval od 15. oktobra do 7. novembra 1941, ko je “zaradi deževja in mraza” izkopavanja zaključil.50 O najdbah in dognanjih je objavil članek v nemški različici v časopisu Marburger Zeitung,51 slovenska verzija pa je bila objavljena v Jutru.52 V obeh člankih je bila podana interpretacija izkopavanj, brez objave 49 Pečnik 1904 in vira 22 in 23. 50 Vir 1, str. 32–72. 51 Schmid 1941. 52 Članek v Jutru je takole povzel Schmidove interpretacije: “ … Na bregu, kamor precej strmo vodi stara rimska cesta, je stala pošta (mansio Atrantina), dolga 29,30 m, široka 16,35 m. … Strešna opeka, žigosana z žigom II. italske legije, pričuje o sosednji hiši, da je tam bila vojašnica majhnega oddelka (vexillatio) legije, ki je bila tu ustanovljena okrog leta 165. Ta je imela nalogo varovati italijansko državno mejo ter je stalno bivala na Ločici pri Polzeli. Leta 176 je bila legija premeščena v Albing ob Donavi, leta 190 v Lorch ob izlivu Aniže. Nad starejšim poslopjem, ki so ga Markomani zažgali leta 176 so legijonarji zgradili novo stavbo, dolgo 18,80 in široko 15,60. V njej je bilo mogoče dognati bivališče vojakov, kuhinjsko ognjišče, kovačnico z mehom, svetišče in oficiarske prostore poslikane rdeče in belo. Svoboden prostor je ločil vojašnico od sosednjega, pač najbolj impozantnega poslopja naselbine. To je bilo dolgo 39,90 m, široko 14,80 m in je delno pod današnjo cesto in gospodarskim poslopjem. Veliki pestro poslikani sobi, kjer se je dalo kuriti so bile priključene še tri večje sobe. Zidovje je bilo nenavadno močno, debelo po 1,65 m in je imelo mnogo globlje temelje kakor ostala poslopja. To je bila carinarnica. Imena uradnikov so pogostokrat naznačena z napisi. Kjer je carinarnica imela značaj trdnjave, je lahko v sili služila za obrambo, kakor podobna, masivno zidana stavbišča na Vrhniki in Ptuju.” (Jutro 1941). fotografij ali načrtov. V Schmidovi zapuščini se je ohranil terenski dnevnik, načrti in korespondenca s celjskim gradbenim inženirjem A. Farčnikom,53 ki je podrobneje izrisal le tri objekte (sl. 10).54 Šašel je ob topografiji terena leta 1953 po pripovedovanju domačinov skiciral Šmidove izkope (prim. sl. 5). Schmid je po lastni interpretaciji na Trojanah leta 1941 našel carinarnico (hiša IV), kasarno II. italske legije (hiša III/prvotno I), poštno postajo (hiša I/prvotno III; sl. 10; prim. sl. 20; 22) z gospodarskimi poslopji. Piše, da je bilo mogoče docela odkriti sledove dveh starorimskih hiš, ostanke tretje pa samo delno, ker so jih preplastile stanovanjska hiša in gospodarska poslopja.55 Opisi v nadaljevanju so povzeti po Schmidu, omenjene so njegove interpretacije, uporabljeno je aktualno številčenje hiš.56 – Hiša I (sl. 10–11; 14–16): po Schmidu pošta (dol. 29,30 m, šir. 16,35 m). Prostori 3, 5 in 6 – sobe za goste, 7 in 8 – koridor (širine 1,35 m); prostori 1, 4, 10, 11 – klet, 9 in 13 – kuhinja z ognjiščem in 14 – velika soba za goste. Tlak v prostoru 6 je bil iz kamnov (oblic) presekan z vkopom. Hiša je imela pod posameznimi sobami kleti, nivo sterilnega laporja je bil na globini 0,98 m. Najdenih je bilo več ognjišč, ostankov ometa, malte. V dnevniku je tudi skica polkrožnega zaključka (sl. 11),57 vendar nobenega komentarja na arhitekturo. 53 Za pojasnilo o Farčniku glej vir 3. 54 Vir 3. 55 Schmid 1941, 5 in Jutro 1941; Vira 6 in 9. 56 Vir 1, str. 41, 44, 64. Pojasnilo o Schmidovem številčenju in preštevilčenju glej v poglavju Številčenje antičnih objektov. 57 Vir 1, str. 44–48. Sl. 12: Trojane, območje 7. Odlomek marmornega stebrička iz prostora 8 v hiši III (premer je pribl. 7 cm). Fig. 12: Trojane, Area 7. Fragment of a small marble pillar from House III, Room 8 (diam. approx. 7 cm). (Medobčinski muzej Kamnik, inv. št. / Inv. No. MMK 3552. Foto / Photo: D. Valoh) V dnevniku Schmid omenja keramične črepinje, 2 ročaja amfore, 2 kosa posode, novec Trajana in Konstantina. Najdbe, ki so prišle v MMK, imajo na listkih stare terenske številke. Ker sklepamo, da so bili listki pridani najdbam v času izkopa, smo – zaradi že omenjenega Schmidovega preštevilčenja – najdbe, s terenskimi listki hiše III, pripisali hiši I (aktualno številčenje!). Najdbe: koščeno vreteno, koščena okrasna ploščica, bronast strigilis, železen stilus, lupine ostrig, odlomki sigilatnih posod, odlomki kuhinjskega posodja, skodele ter dva novca (as Domicijana in folis Constantina I.). – Hiša II (sl. 13–15): v Schmidovem dnevniku se pojavi samo na skicah.58 Širina objekta naj bi znašala 16,95 m. Umestitev, velikost oz. odmik od severne ceste in sosednjega objekta opisuje tudi Niko Kralj v pismu Schmidu.59 Med objektoma II in III je bila 6 m široka cezura in poploščenje/tlakovanje (sl. 13) ob severni fasadi hiše II.60 – Hiša III (sl. 10; 13; 15): po Schmidu kasarna za pomožne enote Druge italske legije (široka 15,60 in dolga 18,80 m). Prostor 1 – kuhinja z ognjiščem, 4 – kovačnica, 5 – svetišče, prostori 2, 3, 6, 7 – prostori za vojake, 8 in 9 – oficirske sobe. Zidovi so bili široki 40–45 cm, v prostoru 2 – hipokavst, je bilo najdeno kurišče oz. več ognjišč, stene so bile ometane z ometom, poslikane s freskami, estrihi so bili maltni, tlaki narejeni iz marmornih plošč. Bili sta vsaj dve gradbeni fazi, vmes žganinske plasti. Globina izkopa je bila med 139 in 142 cm.61 Med Schmidovimi arheološkim najdbami (s terenskimi listki za hišo I = aktualna hiša III) so: bronast pokrov pikside, železen kavelj, 16 odlomkov bronaste pločevine, surovci svinca, odlomka dveh oljenk, 2 odlomka stekla, odlomek steklene zapestnice, zelena steklena jagoda, školjke, polži, nekaj odlomkov keramičnih krožnikov in posod, 58 Med najdbami, ki so prišle v MMK z graškega UMJG ni bilo nobenega listka s terenskimi podatki za hišo II, zato iz tega objekta ni arheoloških predmetov.59 Vir 10. 60 Vir 1, str. 41. 61 Vir 1, str. 32–41. in arhtekturni člen – poševno nažlebljen marmorni stebriček62 (sl. 12). Stebriček močno spominja na odlomek s Panorame na Ptuju, datiran v 5. st.63 Nekatere najdbe iz hiše III so sakralnega značaja. Morda je bilo tukaj svetišče, kopališče in/ali pozneje cerkev? – Hiša IV (sl. 10; 15): po Schmidu carinarnica. 1 – ogrevan prostor, 2 – tlakovan prostor, 3, 4 – spremljajoča prostora. Objekt je meril 39,90 × 14,80 m. Plasti so si sledile od humusa, preko več ruševinskih plasti, plasti žganine, ilovice do globine 1,58 ali 1,66 m. Zidovi so bili široki od 1,35 m, po skici sodeč pa do 1,45 m. V prostorih so maltni estrih, marmorne plošče in podesti iz peščenjaka.64 Literatura in viri: Schmid 1941; Jutro 1941. Vir 1, str. 32–72; vir 2, str. 17; vir 6; vir 8, vir 9. Območje 8 ( sl. 2: 8) Lokacija: Trojane 30; parc. št. 4/1, 4/2 k. o. Trojane. Lastnik: Zdravko Brvar. Zgodovina odkritij in raziskav: – V letih 1956–1957 je lastnik Z. Brvar na parc. 4/1 (za lokacijo parcel glej sl. 22) ravnal teren in odrinil za 30 do 40 cm prsti, da bi naredil prostor za skladišče lesa ob žagi. Antično kulturno plast je uničil in odrinil proti jugu.65 – Hiša X: Leta 1975 je na tem območju zaščitno izkopavala Mirina Zupančič (MMK). Odkrito arhitekturo (oštevilčena je kot hiša X; sl. 16; 20; 22) je interpretirala kot rimsko sušilnico za keramiko (naj bi bila lesena). Na južni strani se je zaključila z dvema lončarskima pečema. Arheološke ostaline so bile uničene zaradi črne gradnje (Trojane 30). Drobne najdbe hrani MMK in so časovno opredeljene od 2. st. do prve polovice 4. st. Med gradbenim materialom temeljev je bil odkrit fragment kamna s posvetilnim napisom (kat. št. 10), v njenem zidu je bil vzidan odlomek marmornega kipa (kat. št. 11: komolec z delom nadlahti, visok 18 cm) ter polovica marmornega stebra in kosi marmornih plošč za oblogo – vse naj bi bilo močno ožgano. Tu so bili odkriti tudi odlomki stenskega ometa z večbarvno figuralno poslikavo. Hiša X leži tik ob Schmidovi hiši III, ob njeni južni steni in je edina pri kateri opazimo drugačno usmeritev (prim. sl. 20; 22). Literatura: Cvikl Zupančič 1966; Šašel 1959; Zupančič 1971, 221; 1977; 1979, 15–19; Zupančič et al. 2003, 16. Območje 9 (sl. 2: 9; 16) Lokacija: Ržunov vrt, med objektoma Trojane 9 in Trojane 29 (dan. parc. št. 573/1 k. o. Trojane). Lastnik: Anica Pungartnik. Zgodovina odkritij in raziskav: Prostor predstavlja nekdanji Ržunov vrt, ki je bil razdeljen. Na vzhodnem delu je bila zgrajena hiša Trojane 29 (lastnica Valentina Majda Hribar), na zahodu pa hiša Trojane 9 (Anica Pungartnik). Schmid je leta 1941 izkopaval na Ržunovem vrtu ter našel objekt s polkrožnim zaključkom (hiša I).66 Šašel poroča, da leži na prostoru stavbe ter na njivah južno od nje (do prvega preloma na pobočju, ki pada proti potoku Orehovica) v zemlji zidovje ter da se pogosto najdejo rimske starine.67 – Hiši XI in XII: Leta 1970 je M. Zupančič v okviru mladinskih raziskovalnih taborov na podlagi pričevanja domačinov o “pol­krožnem zidu” na vrtu izkopala tri sonde.68 Ponovno je odkrila temelje objekta z apsido (hiša I), ki ga je izkopal že Schmid leta 1941 (glej območje 7), in še temelje dveh novih antičnih stavb (hišo XI in del hiše XII; sl. 16; 20; 22). Sondiranja so dala veliko drobnih najdb,69 keramiko, dve bronasti in koščeno šivanko ter novce od druge po­lovice 1. st. do sredine 4. st. (najmlajši folis Konstantina II., kovan leta 337).70 62 Stebriček je bil odkrit 17. 10. 1941 v prostoru 8; za številčenje prostorov prim. (sl. 10: hiša z rdečo št. III). Dimenzije frag­menta: ohranjena višina 15,1 cm; premer 7,2 × 6,4 cm (inv. št. MMK 3552). 63 Knific 1991, 19, sl. 25. 64 Vir 1, str. 43, 63. 65 Ustni podatek M. Zupančič. 66 Vira 1 in 8. 67 Šašel 1954a, 160. Zupančičeva na več mestih navaja napačno številko parcele za lokacijo izkopa na parceli Valentine oz. Majde Hribar (Trojane 29) – parc. št. 4/2 k. o. Trojane je namreč locirana na parceli Zdravka Brvarja (Trojane 30). Včasih je težko razbrati, za kateri objekt gre in kdaj je bil v resnici izkopan. 69 Najdbe so v MMK. 70 Zupančič 1971, 221. Literatura in viri: Jutro 1941; Šašel 1954a, 160; Zupančič, 1970; 1971, 221; 1979, 17. – Vir 1; vir 8; vir 15. Območje: 10 (sl. 2: 10; 16) Lokacija: Trojane 29, 571/2 k. o. Trojane.71 Lastnik: Majda Hribar. Zgodovina odkritij in raziskav: Pri izkopu temeljev za hišo št. 29 so leta 1966 odkrili zidove in obvestili kamniški muzej. M. Zupančič je ob zaščitnem izkopavanju odkrila temelje antičnega objekta velikosti 7 × 12,80 m (hiša XII; sl. 16; 20; 22), s hipokavstom in kanalom. Južna polovica stavbe je bila zaradi oranja odstranjena. Po novcih Zupančičeva datira kompleks v čas 3. in 4. st. Tu naj bi šlo le za eno gradbeno fazo. V kulturni plasti je bilo veliko število odlomkov keramike.72 Literatura: Cvikl Zupančič 1966; Šašel 1959; Zupančič 1971, 221; 1977; 1979, 15–19; Zupančič et al. 2003. Območje 11 (sl. 2: 11) Lokacija: Trojane 8. Lastnik: Ivan Izlakar (p. d. Matija). 71 V objavah je parcela 571/2 pogosto napačno navedena kot parc. št. 4/2 k. o. Trojane (kar je sicer parcela Zdravka Brvarja, na kateri je bilo zaščitno izkopavanje leta 1975). 72 Cvikl-Zupančič 1966; Zupančič 1979, 17. Zgodovina odkritij in raziskav: Vzhodno od Trojan, južno od ceste Trojane–V Zideh, so bili v dolžini okoli 200 m (vzhodno od stavbe Trojane 8; za lokacijo prim. sl. 22) najdeni posamični rimski novci. Literatura: Šašel 1954a, 160. Območje 12 (sl. 2: 12) Lokacija: pod Stojcem (parc. št. 9/1, 5/2, 5/3, 5/4, 5/5, 574/1 k. o. Trojane). Lastnik: Brvarjevi. Zgodovina odkritij in raziskav: Leta 2003 so južno od Brvarjevega posestva (Trojane št. 10; za lokacijo prim. sl. 22), ki je pokalo in polzelo zaradi izgradnje trojanskega predora, izvajalci del začeli s sanacijo. Pri tem je bila utrjena stara in zgrajena nova dovozna pot do delovišča vrtin. Izvajalci so odstranjene arheološke plasti deponirali na sedem kupov ob novi dovozni poti. Pod vodstvom Draška Josipoviča (Magelan Skupina d. o. o., Kranj) je bila deponija presejana in pregledana. Najdena je bila kamnita plastika Herkula.73 Več kot polovico najdb predstavljajo navadne opeke, imbreksi, tegule, med njimi dve z žigom II. italske legije. Veliko je bilo keramičnega posodja, med katerimi prevladuje gradivo iz 3. in 4. st. Literatura: Josipovič 2003. Območje 13 (sl. 2: 13) Lokacija: Hotel Trojane (Trojane 27). Zgodovina odkritij in raziskav: Negativno. Leta 1941 naj bi po spominu domačinov območje s sistemom jarkov pregledal že Schmid, vendar ni našel ničesar. Pred gradnjo hotela naj bi to zemljišče – v vzhodnem delu naselja – leta 1975 sondirala M. Zupančič (MMK) vendar brez uspeha. Območje 13 zelo verjetno leži že zunaj vzhodnega roba rimske naselbine (za lokacijo stavbe prim. sl. 22). Viri: Ustni podatek: Mirina Zupančič in domačin Janez Ocvirk. Območje 13a (sl. 2: 13a) Lokacija: Trojane 25. Lastnik: Robert Smrkolj. Zgodovina odkritij in raziskav: Negativno. 2. avgusta leta 2018 je arheolog Uroš Košir (Avgusta d. o. o.), zaradi širitve stavbe proti vzhodu, izvajal arheološki nadzor ob gradnji. Antičnih ostalin niso našli. Območje 13a zelo verjetno leži že zunaj vzhodnega roba rimske naselbine. Literatura in viri: Košir 2018. Območje 14 (sl. 3: 14; 4) Lokacija: okolica Učaka. Zgodovina odkritij in raziskav: Učak je vzpetina na zahodnem robu trojanskega grebena, tik nad trojanskim prelazom. Leži med kmetijo Dobiček (Učak 14) in kmetijo Lukej (Učak 7).74 Leta 1846 naj bi kmetje v potoku Orehovica (potok se izliva v Me-dijo) pod Učakom našli dele bronastega kipa ter jih stalili. Leta 1889 je deček Anton Medvešček našel spodnji del konjskega gobca in del konjskega trebuha (prispelo v Deželni muzej).75 Rutar omenja tudi kamniti podstavek kipa z napisom, ki naj bi ga domačini uničili. Pečnik pa je lokacijo vrisal na karto arheoloških najdišč Radeče in Zagorje (sl. 4). Upodobljen naj bi bil cesar na konju, morda Mark Avrelij. Kip je po stilu datiran v 2. st.76 Literatura in viri: Rutar 1890; 1891, 197; Pečnik 1904, 131; Ložar 1937-1940; Šašel 1954a, 164; Šašel Kos 1999, 239; Istenič 2014, 97. – Vir 22; vir 23. 73 Najdbe hrani MMK. Višina plastike: 42,5 cm, šir. 12–14 cm, dol. 19,3 cm (kat. št. 12, inv. št. MMK 3713). 74 Šašel 1954a, 164. 75 Rutar 1890; 1891, 197; Pečnik 1904, 131. Ohranjeni in v inv. knjigo NMS vpisani so: fragment konjskih nozdrvi, fr. spod­nje čeljusti in štrije fr. konjskega trupa. NMS inv. št. 1811a,b ter 1812a–d. 76 Ložar 1937–1940; Šašel Kos 1999, 239; Istenič 2014, 97, sl. 105. Območje 15 (sl. 3: 15) Lokacija: Učak. Zgodovina odkritij in raziskav: – Oktobra leta 1942 je bil Schmid drugič na Trojanah, sočasno je sondiral območje hriba Učak. Naredil je 12 sond. Na vrhu je našel objekt, skoraj kvadratne oblike (stranica do 4,20 m)77 in ga interpretiral kot stolp oziroma utrdbo iz 3. in 4. st.78 Terenski dnevnik vsebuje skico z lokacijami sond ter opise plasti, globin in zidov (sl. 17).79 77 Viri 2, 4, 5 (korespondenca Farčnika in Schmida). 78 Vir 2. 79 Vir 2, str. 14–24. – Leta 1959 je na severovzhodnem vznožju griča, na parc. št. 641 in 647, k. o. Trojane, ob gradnji ceste sondiral Lojze Bolta (za Zavod za spomeniško varstvo LRS). Odkrili so antične zidove, tegule in keramiko. Literatura in viri: Šašel 1954a, 160–161; Bolta 1959, 130; 1960, 343. – Vir 2; vir 4; vir 5; vir 13. Območje 16 (sl. 3: 16) Lokacija: trasa “stare” ceste Ljubljana–Celje, zahodno od Trojan. Zgodovina odkritij in raziskav: Ob rekonstrukciji ceste Ljubljana–Celje so konec 50. in v začetku 60. let dvajsetega stoletja naleteli na traso rimske ceste. Na odseku C–56, Šentožbolt–Trojane, v profilu 136, je leta 1959 L. Bolta, PMC, na globini 90 cm našel rimsko cesto, ki je ponekod debela do 80 cm. Literatura in viri: Bolta 1959; 1960; Zupančič 1979, 17. – Vir 21. Območje 17 (sl. 3: 17) Lokacija: trasa “stare” ceste Ljubljana–Celje, zahodno od Trojan. Zgodovina odkritij in raziskav: Glej Območje 16. Rimska cesta je bila vidna tudi med profiloma 105 in 107 trase ceste Ljubljana–Celje. Literatura in viri: Bolta 1959; 1960; Zupančič 1979, 17. – Vir 21. Območje 18 (sl. 3: 18) Lokacija: trasa “stare” ceste Ljubljana–Celje, zahodno od Trojan. Zgodovina odkritij in raziskav: Glej Območje 16. Med profiloma 104 in 105 trase ceste Ljubljana–Celje je L. Bolta je od 6. 4. do 14. 8. 1959 izkopal tri antične grobove. Dva (verjetno žgana) grobova sta imela opečno konstrukcijo. V prvem so, poleg nekaj ožganih kosti, našli dvoročajni vrč in balzamarij. Drugi, v konstrukciji iz tegul, ni imel pridatkov. Tretji grob (skeletni) je bil v kamnitem sarkofagu, prazen, brez pridatkov. Poleg grobov je našel rimski zid, ki je bil povezan z grobiščem.80 Literatura in viri: Bolta 1959, 130–133; 1960. – Vir 21. Območje 19 (sl. 3: 19) Lokacija: Podzid; trasa “stare” ceste Ljubljana–Celje, vzhodno od Trojan. Zgodovina odkritij in raziskav: Leta 1997 je Tomislav Kajfež (za ZVKDS, OE Kranj) sondiral v dolini Bolske, pod Trojanami. Na parc. št. 831/5 k. o. Trojane je odkril rimsko cesto. Cesta je bila široka 6 m. Nasutje je bilo debelo od 20 do 25 cm. Literatura: Kajfež 1997; 2003, 267–268. Območje 20 (sl. 3: 20) Lokacija: V Zideh. Zgodovina odkritij in raziskav: Leta 1996 je Marija Ogrin (za ZVKDS, OE Kranj) s podpovršinskim pregledom raziskala severno pobočje tro­janskega grebena, pri naselju V Zideh (parc. št. 468/1, 469/1, 521/1 in 518 k. o. Trojane; sl. 18: 3), vendar ni odkrila naselbinskih ostankov. Leta 1997 je Branko Mušič (Oddelek za arheologijo, FF v Ljubljani) izvajal geofizikalne raziskave deloma na istem zemljišču – na obeh skrajnih koncih leta 1996 pregledanega pasu – in v nadaljevanju proti vzhodu in zahodu (sl. 18: 1 in 2). Z geoelektričnim kartiranjem so bile na zahodnem delu raziskanega zemljišča (sl. 18: sektor 2) odkrite pravilne oblike, ki naj ne bi bile posledica naravnih tvorb in bi lahko predstavljale ostanke arhitekture. Sledovi niso bili preverjeni z izkopavanji. Literatura: Ogrin 1996; Mušič 1997. 80 Od januarja leta 2017 je vse gradivo v hrambi v MMK (sarkofag inv. št. MMK 3701). Območja 21, 22, 23, 24 (sl. 3) Območja od št. 21 do 24 so sekundarna najdišča: točke na katerih so nekoč bili ali pa so še danes vzidani kamniti spomeniki, ki izvirajo iz antičnega Atransa. Glej katalog kamnitih spomenikov (kat. št. 13–23). Pomenljivo je, da gre za posvetilne napise, votivne oltarje in da nagrobnikov, z izjemno stele, posvečene Hilaru (CIL 5124), ni. KATALOG KAMNITIH SPOMENIKOV81 Večina teh spomenikov izvira iz antičnega Atransa. Nekateri so danes v sekundarni legi vzidani v objekte (zasebne hiše ali cerkve v bližnji okolici), drugi so pre­meščeni v muzeje, tretji pa izgubljeni. Za razlago v besedilu uporabljenih okrajšav glej Katalog arheoloških območij. Kat. št. 1 Najdišče: -. Votivni oltar. Atranti / Aug(ustae ?) sac(rum). / Fortunatus / C(ai) Antoni(i) Rufi / proc(uratoris) Aug(usti) ser(vus) vil(icus) / v(otum) s(olvit) l(ibens) m(erito). Druga polovica 2. st. Objave: CIL III 5117; Visočnik 2017, št. 423; HD060189; lupa 9520; EDCS-14500422. Hrani: Musei provinziali in Gorizia, inv. št. 5/53. Na tem mestu ne navajamo vseh objav napisov; od­ločili smo se samo za objavo v CIL, zadnjo objavo v Visoč­nik 2017 ter objave v epigrafskih bazah. Nikakor pa ne gre zanemariti še drugih pomembnih objav, predvsem v RINMS kakor tudi prevodov napisov slovenščino, ki so v glavnem na voljo v Šašel Kos 2004. (J. Ž.) CATALOGUE OF STONE MONUMENTS Most of these monuments originate from the Roman Atrans. Some of them are now in secondary position, built into buildings (private houses and churches in the vicinity), others have been transferred to museums, some are lost. Cat. No. 1 Unlocated. Votive altar. Atranti / Aug(ustae ?) sac(rum). / Fortunatus / C(ai) Antoni(i) Rufi / proc(uratoris) Aug(usti) ser(vus) vil(icus) / v(otum) s(olvit) l(ibens) m(erito). Second half of the 2nd century. References: CIL III 5117; Visočnik 2017, No. 423; HD060189; lupa 9520; EDCS-14500422. Kept in: Musei provinziali in Gorizia, Inv. no. 5/53. Kat. št. 2 Najdišče: -. Votivni oltar. Hecate / Augustae. / Aurelii, As- / clepiodo- / tus et Lucius,/ pro sal(ute) / sua et suor(um) / v(otum) s(olverunt) l(ibentes) m(erito). Druga polovica 2. st.–začetek 3. st. Objave: CIL III 5119; Visočnik 2017, št. 425; HD066799; lupa 9250; EDCS-14500424. Hrani: NMS, inv. št. L30. Kat. št. 3 Najdišče: -. Votivni oltar. I(ovi) O(ptimo) M(aximo). / C(aius) Antonius / Iu­lianus / proc(urator) p(ublici) p(ortorii) X. / V(otum) s(olvit) l(ibens) m(erito). Druga polovica 2. st. Objave: CIL III 5120; Visočnik 2017, št. 426; HD066798; lupa 9521; EDCS-14500425. Hramba: Poznan v 16. st., danes izgubljen. Kat. št. 4 Najdišče: -. Votivni oltar. Norei(a)e / August(ae) et / Honori / stat(ionis) Atrant(inae). / Bellicius et / Eutyches / c(ontra) sc(riptores) stat(ionis) / eiusdem / ex vot(o). Druga polovica 2. st. Objave: CIL III 5123; Visočnik 2017, št. 431; HD060190; lupa 9519; EDCS-14500428. Hrani: Musei provinziali in Gorizia, inv. št. 7/53. Kat. št. 5 Najdišče: območje 5. Nagrobna plošča. [---] / [---]OL[---] / [--- v]ix(it) an(nos) XX / [---] eiusdem / [---]I karissimo (!). 1.–3. st. Objave: CIL III 11678, Visočnik 2017, št. 437; HD066785; EDCS-14600174. Hramba: Danes izgubljen. Kat. št. 6 Najdišče: območje 6. Nagrobna stela? Relief krilatega grifona. Druga polovica 2. st. Objava: Zupančič et al. 2003, 12. Hrani: MMK, inv. št. 3704. Kat. št. 7 Najdišče: območje 6. Dva napisna kamna. Omemba: Müllner 1879, 259 ad 121. Cat. No. 2 Unlocated. Votive altar. Hecate / Augustae. / Aurelii, As- / clepiodo- / tus et Lucius,/ pro sal(ute) / sua et suor(um) / v(otum) s(olverunt) l(ibentes) m(erito). Second half of the 2nd century – beginning of the 3rd century. References: CIL III 5119; Visočnik 2017, No. 425; HD066799; lupa 9250; EDCS-14500424. Kept in: Narodni muzej Slovenije, Inv. No. L30. Cat. No. 3 Unlocated. Votive altar. I(ovi) O(ptimo) M(aximo). / C(aius) Antonius / Iulianus / proc(urator) p(ublici) p(ortorii) X. / V(otum) s(olvit) l(ibens) m(erito). Second half of the 2nd century. References: CIL III 5120; Visočnik 2017, No. 426; HD066798; lupa 9521; EDCS-14500425. Kept in: Known in the 16th century, now lost. Cat. No. 4 Unlocated. Votive altar. Norei(a)e / August(ae) et / Honori / stat(ionis) Atrant(inae). / Bellicius et / Eutyches / c(ontra) sc(riptores) stat(ionis) / eiusdem / ex vot(o). Second half of the 2nd century. References: CIL III 5123; Visočnik 2017, No. 431; HD060190; lupa 9519; EDCS-14500428. Kept in: Musei provinziali in Gorizia, Inv. no. 7/53. Cat. No. 5 Area 5. Tombstone. [---] / [---]OL[---] / [--- v]ix(it) an(nos) XX / [---] eiusdem / [---]I karissimo (!). 1st – 3rd century. References: CIL III 11678, Visočnik 2017, No. 437; HD066785; EDCS-14600174. Kept in: Lost. Cat. No. 6 Area 6. Funerary stele? Relief of a winged griffin. Second half of the 2nd century. Reference: Zupančič et al. 2003, 12. Kept in: Medobčinski muzej Kamnik, Inv. No. 3704 Cat. No. 7 Area 6. Two inscription stones. Mentioned in: Müllner 1879, 259 ad 121. Hramba: Eden je ostal na dnu vodnjaka pri hiši na Tro­janah (h. št. 10), drugi izgubljen. Kat. št. 8 Najdišče: območje 6. Cesarski gradbeni napis. [Imp(erator) Caes(ar) M(arcus) Aurelius Antonin] us Aug(ustus) / [Armeniac(us) Medicus Parthicus max(imus)] pontif(ex) / [maximus tribunic(iae) po­test(atis) XXII imp(erator) V c]o(n)s(ul) III p(ater) [p(atriae) e]t / [imp(erator) Caes(ar) L(ucius) Au­relius Verus Aug(ustus) Armenia]c(us) M[edicus] / [Parth(icus) max(imus) tribunic(ia) potest(atis) VIII imp(erator) V co(n)s(ul) III p(ater) p(atriae)] / [divi Antonini filii divi Hadriani] nepotes / [divi Traiani Parth(ici) pronep(otes) divi Ner]vae abnep(otes) / [--- ru]nt. Druga polovica leta 168. Objave: ILJug 382; CIL 11675; Visočnik 2017, št. 439; HD035121; lupa 5156; EDCS-14600171. Hranita: MMK, inv. št. MMK 3702 (prvi odlomek); PMC, inv. št. L 191 (drugi odlomek). Kat. št. 9 Najdišče: območje 6. Votivni oltar. I(ovi) O(ptimo) M(aximo) / Culminali. / Chresimus / Augg(ustorum) [...] / ------ Druga polovica 2. st.–začetek 3. st. Objave: CIL III 11673; Visočnik 2017, št. 427; HD066788; lupa 9252; EDCS-14600169. Hrani: NMS, inv. št. L 20. Kat. št. 10 Najdišče: območje 8. Posvetilni ali počastitveni napis. ------ / [---]+++[---] / [--- Sa]turn[in---? ---] / [pro] c(urator)(?) Au[g(usti)? ------] / ------ Druga polovica 2. st. Objave: Zupančič 1977; Zupančič et al. 2003, 7; Visočnik 2017, št. 436. Hrani: MMK, neinv. Kat. št. 11 Najdišče: območje 8. Odlomek kipa – komolec roke. 2. st. Objave: Zupančič 1977; Zupančič et al. 2003, 16. Hrani: MMK, neinv. Kat. št. 12 Najdišče: območje 12. Reliefna plastika – Herkul z gorjačo. Druga polovica 2. st.–začetek 3. st. One remained at the bottom of a well (house Trojane 10), the other one is lost. Cat. No. 8 Area 6. Imperial building inscription. [Imp(erator) Caes(ar) M(arcus) Aurelius Antonin] us Aug(ustus) / [Armeniac(us) Medicus Parthicus max(imus)] pontif(ex) / [maximus tribunic(iae) potest(atis) XXII imp(erator) V c]o(n)s(ul) III p(ater) [p(atriae) e]t / [imp(erator) Caes(ar) L(ucius) Au­relius Verus Aug(ustus) Armenia]c(us) M[edicus] / [Parth(icus) max(imus) tribunic(ia) potest(atis) VIII imp(erator) V co(n)s(ul) III p(ater) p(atriae)] / [divi Antonini filii divi Hadriani] nepotes / [divi Traiani Parth(ici) pronep(otes) divi Ner]vae abnep(otes) / [--- ru]nt. Second half of the year 168. References: ILJug 382; CIL 11675; Visočnik 2017, No. 439; HD035121; lupa 5156; EDCS-14600171. Kept in: Medobčinski muzej Kamnik, Inv. No. MMK 3702; Pokrajinski muzej Celje, Inv. No. L 191. Cat. No. 9 Area 6. Votive altar. I(ovi) O(ptimo) M(aximo) / Culminali. / Chresimus / Augg(ustorum) [...] / -----­ Second half of the 2nd century – beginning of the 3rd century. References: CIL III 11673; Visočnik 2017, No. 427; HD066788; lupa 9252; EDCS-14600169. Kept in: Narodni muzej Slovenije, Inv. No. L 20. Cat. No. 10 Area 8. Votive or honorary inscription. ------ / [---]+++[---] / [--- Sa]turn[in---? ---] / [pro] c(urator)(?) Au[g(usti)? ------] / -----­ Second half of the 2nd century. References: Zupančič 1977, 106–109; Zupančič et al. 2003, 7; Visočnik 2017, No. 436. Kept in: Medobčinski muzej Kamnik. Cat. No. 11 Area 8. Fragment of a statue – elbow. 2nd century. References: Zupančič 1977, 106–109; Zupančič et al. 2003, 16. Kept in: Medobčinski muzej Kamnik. Cat. No. 12 Area 12. Relief – Hercules with a club. Objava: Josipovič 2003, 6. Hrani: MMK, inv. št. MMK 3713. Kat. št. 13 Najdišče: območje 21 (Šentožbolt). Votivni oltar. Libero / Patri / sacr(um). / Abascantus / Antoni Rufi s(ervus) / scrut(ator?) v(otum) s(olvit) l(ibens) m(erito). Druga polovica 2. st. Objave: CIL III 5122; Visočnik 2017, št. 428; HD066797; lupa 9253; EDCS-14500427. Hrani: NMS, inv. št. L 27. Kat. št. 14 Najdišče: območje 21 (Šentožbolt). Votivni oltar. A[---] / Aurel(ius) [As]- / clepio[do]- / tus [---]. / V(oto) s(oluto) d(onum) d(edit). Druga polovica 2. st.–začetek 3. st. Objave: CIL III 11677; Visočnik 2017, št. 433; HD066786; EDCS-14600173. Hramba: izgubljen. Kat. št. 15 Najdišče: območje 21 (Šentožbolt). Votivni oltar. Marti Aug(usto). / Pro salute / Primigeni(i) / Augg(ustorum) nn(ostrorum) vil(ici) / stat(ionis) Atrantin(ae) / [---]. Druga polovica 2. st.–3. st. Objave: CIL III 11674; CIL III 13522; Visočnik 2017, št. 429; HD066443; lupa 9515; EDCS-14600170. Hramba: izgubljen. Kat. št. 16 Najdišče: območje 21 (Šentožbolt). Votivni oltar (?). Fragment votivnega oltarja. Müllner zapiše, da je bil brez napisa. Druga polovica 2. st.–3. st. (?) Objava: Visočnik 2017, št. 434. Hramba: izgubljen. Kat. št. 17 Najdišče: območje 22 (Sv. Mohor). Nagrobna plošča. ------ / [---]AC[---] / [---]NA [---] / [---]NUS [---] / 1.–4. st. Objave: ILJug 383; Visočnik 2017, št. 438. Hramba: vzidan v kropilnik v cerkvi sv. Mohorja in Fortunata. Second half of the 2nd century – beginning of the 3rd century. Reference: Josipovič 2003, 6. Kept in: Medobčinski muzej Kamnik, Inv. No. MMK 3713. Cat. No. 13 Area 21 (Šentožbolt). Votive altar. Libero / Patri / sacr(um). / Abascantus / Antoni Rufi s(ervus) / scrut(ator?) v(otum) s(olvit) l(ibens) m(erito). Second half of the 2nd century. References: CIL III 5122; Visočnik 2017, No. 428; HD066797; lupa 9253; EDCS-14500427. Kept in: Narodni muzej Slovenije, Inv. No. L 27. Cat. No. 14 Area 21 (Šentožbolt). Votive altar. A[---] / Aurel(ius) [As]- / clepio[do]- / tus [---]. / V(oto) s(oluto) d(onum) d(edit). Second half of the 2nd century – beginning of the 3rd century. References: CIL III 11677; Visočnik 2017, No. 433; HD066786; EDCS-14600173. Kept in: Lost. Cat. No. 15 Area 21 (Šentožbolt). Votive altar. Marti Aug(usto). / Pro salute / Primigeni(i) / Augg(ustorum) nn(ostrorum) vil(ici) / stat(ionis) Atrantin(ae) / [---]. Second half of the 2nd century – 3rd century. References: CIL III 11674; CIL III 13522; Visočnik 2017, No. 429; HD066443; lupa 9515; EDCS-14600170. Kept in: Lost. Cat. No. 16 Area 21 (Šentožbolt). Fragment of a votive altar. According to Müllner, there was no inscription. Second half of the 2nd century – 3rd century (?). Reference: Visočnik 2017, No. 434. Kept in: Lost. Cat. No. 17 Area 22 (Sv. Mohor). Tombstone. ------ / [---]AC[---] / [---]NA [---] / [---]NUS [---] / 1st – 4th century. References: ILJug 383; Visočnik 2017, No. 438. Kept in: Built into a holy water font in the church of Sts Hermagoras and Fortunatus. Kat. št. 18 Najdišče: območje 23 (V Zideh). Votivni oltar. ------ / [.?]vantit / Reg(ius?) / Regalis / Iulia / Aqui- . lin(a) / ex vot(o). 2. st.–3. st. Objave: CIL III 5118; CIL III 11671; Visočnik 2017, št. 432; HD066789; lupa 9255; EDCS-14500423. Hrani: NMS, inv. št. L 24. Kat. št. 19 Najdišče: območje 23 (V Zideh). Gradbeni napis. ------ / [..] +[...] +VS Qui[nt].an.s b(ene)f(iciarius) .. co(n)s(ularis) vac.? / leg(ionis) II Ital(icae) templum . vetustate conlabs- / um et in ruina conversum sump-tu s[uo ..?] / [restituit ---?]. Konec 2. st.–3. st. Objave: CIL 11676; Visočnik 2017, št. 440; HD035121; lupa 5156; EDCS-14600171. Hrani: NMS, inv. št. L 28. Kat. št. 20 Najdišče: območje 23 (V Zideh). Nagrobna stela. D(is) M(anibus) / Hilari, fili, qui / annos vixit III. / Secundianus / c(ontra)scr(iptor) et Rufilla / [par]entes filio / f(aciendum) c(uraverunt). 2. st.–3. st. Objave: CIL III 5124; Visočnik 2017, št. 435; HD045401; lupa 4165; EDCS-14500429. Hrani: MMK, inv. št. MMK 3703. Kat. št. 21 Najdišče: območje 23 (V Zideh). Del nagrobnika – lev. 2. ali 3. st. Objava: Zupančič et al. 2003, 13. Hrani: MMK, neinv. Kat. št. 22 Najdišče: območje 24 (Šentgotard 11 – župnišče). Votivni oltar. D(eo) I(nvicto) M(ithrae). / Eutyches / Iulior(um) / . c(onductorum) p(ortorii) p(ublici?) ser(vus) c(ontra) s(crip)t(or?) / stationis Boiod[u(rensis)] / ex vik(a­ . rio) Benigni vil(ici) / stat(ionis) Atrantin(ae) / aram cum signo / Lunae / ex voto posuit / pr(o)s(edente) (?) T(ito) Cla(udio) Senill(o). Druga polovica 2. st. Objave: CIL III 5121; Visočnik 2017, št. 430; HD066801; lupa 9254; EDCS-14500426. Hrani: NMS, inv. št. L 23. Cat. No. 18 Area 23 (V Zideh). Votive altar. ------ / [.?]vantit / Reg(ius?) / Regalis / Iulia / . Aquilin(a) / ex vot(o). 2nd century – 3rd century. References: CIL III 5118; CIL III 11671; Visočnik 2017, No. 432; HD066789; lupa 9255; EDCS-14500423. Kept in: Narodni muzej Slovenije, Inv. No. L 24. Cat. No. 19 Area 23 (V Zideh). Building inscription. ------ / [..] +[...] +VS Qui[nt].an.s b(ene)f(iciarius) .. co(n)s(ularis) vac.? / leg(ionis) II Ital(icae) templum . vetustate conlabs- / um et in ruina conversum sumptu s[uo ..?] / [restituit ---?]. End of the 2nd century – 3rd century. References: CIL 11676; Visočnik 2017, No. 440; HD035121; lupa 5156; EDCS-14600171. Kept in: Narodni muzej Slovenije, Inv. No. L 28. Cat. No. 20 Area 23 (V Zideh). Funerary stele. D(is) M(anibus) / Hilari, fili, qui / annos vixit III. / Secundianus / c(ontra)scr(iptor) et Rufilla / [par]entes filio / f(aciendum) c(uraverunt). 2nd century – 3rd century. References: CIL III 5124; Visočnik 2017, No. 435; HD045401; lupa 4165; EDCS-14500429. Kept in: Medobčinski muzej Kamnik, Inv. No. MMK 3703. Cat. No. 21 Area 23 (V Zideh). Part of a tombstone – lion. 2nd or 3rd century. Reference: Zupančič et al. 2003, 13. Kept in: Medobčinski muzej Kamnik. Cat. No. 22 Area 24 (Šentgotard 11). Votive altar. D(eo) I(nvicto) M(ithrae). / Eutyches / Iulior(um) / . c(onductorum) p(ortorii) p(ublici?) ser(vus) c(ontra) s(crip)t(or?) / stationis Boiod[u(rensis)] / ex vik(ario) . Benigni vil(ici) / stat(ionis) Atrantin(ae) / aram cum signo / Lunae / ex voto posuit / pr(o)s(edente) (?) T(ito) Cla(udio) Senill(o). Second half of the 2nd century. References: CIL III 5121; Visočnik 2017, No. 430; HD066801; lupa 9254; EDCS-14500426. Kept in: Narodni muzej Slovenije, Inv. No. L 23. Kat. št. 23 Najdišče: območje 24 (Šentgotard 11 – župnišče). Votivni oltar. Dianae Aug(ustae) / pro salute / Imperatorum / Septimi(i) Severi P(ii) / Pertinacis et / ------ Konec 2. st.–začetek 3. st. (najverjetneje med 197 in 211). Objave: CIL III 11672; Visočnik 2017, št. 424; HD066848; EDCS-14600168. Hramba: vzidan v župnišče, Šentgotard 11. (J. V.) DROBNE NAJDBE Izkopano arheološko gradivo, z izjemo desetih fragmentov, ki so v Narodnem muzeju Slovenije (NMS), je v depojih in na razstavah Medobčinskega muzeja Kamnik. Med antičnimi drobnimi najdbami prevladujejo odlomki posod (žganih v oksidacijskih in redukcijskih atmosferah), precej je posod z ročaji (vrči), veliko je krožnikov s premazi. Fragmenti keramičnih posod predstavljajo 62% vseh evidentiranih in inventariziranih najdb. Na drugem mestu po količinski zastopanosti so amfore s 13,3%; 7% je odlomkov fresk in 4,3% opeke (tab. 2). Najdbe imajo značilnosti bogatega civilnega dela naselbine. Drobni arheološki inventar je časovno opredeljen od 1. do 4. st. Vrsta najdbe / Find type Količina / Quantity Koščeni predmeti / Bone artefacts 6 Kosi marmorja oz. arhitekture / Pieces of marble or architectural elements 16 Novci / Coins 31 Lupine školjk in polžev / Shells or snails 33 Mozaične kocke / Mosaic tiles 55 Kosti / Bones 95 Neopredeljeno / Undefined 157 Steklo / Glass 159 Kovinski predmeti / Metal artefacts 201 Opeke / Bricks 238 Omet – freske / Plaster - frescoes 392 Amfore / Amphorae 743 Keramika / Pottery 3471 Skupaj / All 5597 Cat. No. 23 Area 24 (Šentgotard 11). Votive altar. Dianae Aug(ustae) / pro salute / Imperatorum / Septimi(i) Severi P(ii) / Pertinacis et / -----­ End of the 2nd century – beginning of the 3rd century (most likely between 197 and 211). References: CIL III 11672, Visočnik 2017, No. 424; HD066848; EDCS-14600168. Kept in: Built into the rectory – Šentgotard 11. (J. V.) SMALL FINDS The excavated archaeological material, with the exception of ten fragments that are kept in the Narodni muzej Slovenije, is in the depots and exhibitions of the Medobčinski muzej Kamnik. The prevalent types of small finds are fragments of vessels (fired in oxidizing and reducing atmospheres); there are many vessels with handles (jugs) and plates with slip. Fragments of ceramic vessels represent 62 % of all documented and inventoried finds. Amphorae take second place in quantity (13.3%); frescoes represent 7% of the material; and bricks 4.3% (Tab. 2). Small finds are dated between the 1st and 4th cen­turies. Tab. 2: Vrste rimskodobnega arheološkega gradiva, ki je bilo na Trojanah odkrito ob izkopavanjih: R. Ložarja leta 1936, W. Schmida 1941, M. Zupančičeve 1966, 1970 in 1975 ter D. Josipoviča 2003. Gradivo (z izjemo Ložarjevih izkopavanj) hrani MMK. Tab. 2: Types of archaeological finds excavated at Trojane: Ložar 1936, Schmid 1941, Zupančič 1966, 1970, 1975, and Josipovič 2003. All finds except those from Ložar’s excavations are stored in the Medobčinski muzej Kamnik. 16 15 14 9 88 6 2 1 18 16 14 12 10 8 6 4 2 0 Tab. 3: Novci, najdeni na Trojanah in registrirani v zbirki Numizmatičnega kabineta NMS. Tab. 3: Coins found at Trojane and registered in the database of the Numismatic Cabinet at the Narodni muzej Slovenije. NOVCI Numizmatični kabinet NMS ima v svoji računal­niški bazi evidentiranih 81 rimskih novcev82 s Trojan (tab. 3), od tega je danes skoraj polovica izgubljenih oziroma jih ni mogoče povsem zanesljivo pripisati Trojanam.83 Med registriranimi prevladujejo novci iz 3. in 4. st. Zaradi majhnega vzorca ni mogoče ugotavljati denarnega obtoka. Zlasti je premalo novcev iz 2. st., torej iz časa, ko naj bi Atrans postal carinska postaja. Valvasor omenja izjemno množico rimskih novcev, ki še “danes ležijo v zemlji” in da je osebno od domačinov iz vasi V Zideh nakupil “etliche tausend Stücken”.84 Da so “novci iz medenine, brona, srebra kot tudi iz elektruma in zlata in da jih je več tisoč” pa piše v peti knjigi.85 Iz trojanskih novcev je osnoval svojo več kot 8000 kosov obsegajočo zasebno numizmatično zbirko.86 Več kot 2000 novcev veliko zbirko, sestavljeno iz 24 zlatih, 606 srebrnih in 1728 bakrenih novcev, naj bi imel Jožef 82 Podatke je posredovala Alenka Miškec, Numizmatični kabinet NMS. 83 MMK s Trojan hrani 31 rimskih novcev; osem jih je pri izkopavanjih leta 1941 našel Schmid, 15 novcev je pri zaš-čitnih izkopavanjih leta 2003 našel D. Josipovič (2003). Ostali novci so z zaščitnih izkopavanj M. Zupančič. 84 Valvasor 1689a, 124–125. 85 Valvasor 1689b, 266–268. 86 Pegan 2016, 56. COINS There are 81 Roman coins14 from Trojane (Tab. 3) in the database of the Numismatic Cabinet of Narodni muzej Slovenije. Almost half of them are now lost or cannot be attributed to Trojane with certainty.15 The most numer­ous among the registered coins are those from the 3rd and 4th centuries. Small sample size prevents an estimate of monetary circulation. In particular, there are not enough coins from the 2nd century, i.e. from the time when Atrans supposedly became a customs station. Valvasor mentions an exceptional number of Roman coins, which “are still lying in the ground” and claims that he personally bought from the locals of the village V Zideh “etliche tausend Stücken”.16 In his fifth book he writes that “the coins are made of brass, bronze, silver, as well as electrum and gold. There are several thousands of them”.17 The coins from Trojane were the basis for his personal numismatic collection of more 14 Information from Alenka Miškec, Narodni muzej Slo­venije. 15 There are 31 Roman coins from Trojane in the Medobčinski muzej Kamnik; eight coins have been excavated in 1941 by Walter Schmid, and 15 coins in the rescue excava­tion in 2003 led by Draško Josipovič (Josipovič 2003). The remaining coins are from the rescue excavations conducted by Mirina Zupančič. 16 Valvasor 1689a, 124–125. 17 Valvasor 1689b, 266–268. Repežič,87 francoski principal, ki je v letih 1810–1814 služboval na Trojanah.88 Šentgotardski župnik Dornik je leta 1846 med ostalim na Stojčevem (sl. 2: območje 6) našel tudi 14 novcev (od Nera do Gracijana).89 Prokop (1891) omenja številne novice iz druge polovice 3. st. (največ cesarja Proba) in mlajše, do Konstantina, ki naj bi jih poleg ostalega leta 1886 izkopal Pečnik.90 Na tisoče novcev iz “colne postaje na meji …” omenja Pečnik.91 Novčne najdbe opisujejo tudi Ložar,92 Šašel,93 Stražar94 in drugi. Najdeni so bili ob cestah, na vrtovih po Trojanah in bližnji okolici. Samo grobi seštevek opredeljivih novcev je precej več kot 10.500 novcev, kar je pomenljivo veliko. NASELJE (sl. 19–22) CESTA IN GROBIŠČE Popotniki, ki so zapuščali Deseto regijo Italije so iz smeri današnje Blagovice zavili proti Šentožboltu (proti vzhodu), šli mimo, po Pečniku, “mnogo rimskih grobov”.95 Ostankov niti lokacij ne poznamo. Vzpe­njajočim po trojanskem klancu se je dvignil pogled na mogočen kip jezdeca na konju (domnevno kip Marka Avrelija), ki je stal na območju Učaka (sl. 3: 14).96 Rimska cesta je (po Pečniku,97 Rutarju98 in Schmidu99) vodila južno okoli hriba Učak (sl. 4). Tik pred naselbino je bila na zahodu nekropola. Pečnik100 je na Konjškovih njivah (sl. 2: 1) našel nekaj rimskih grobov, Bolta pa na trasi “stare ceste” (sl. 3: 18) tri: enega v sarkofagu, dva pa v opečni grobni konstrukciji.101 Grobovi z območij 1 in 87 Stražar 1985, 84. 88 Slovenec 1939; Slovenski list 1939, 5–6. 89 Müllner 1878; 1879, 81. 90 Prokop 1891. 91 Pečnik 1904. 92 Ložar (1937) poroča o najdbi novcev leta 1936, pri Stojc. 93 Leta 1949 so pri Brvarju odkopali več rimskih zidov, luknje za bruna, druge starine … sem in tja tudi rimske nov­ce. Za hišo Trojane št. 7 in severno od ceste Trojane–Zide se najdejo posamični rimski novci, drugega ne (vir 18/2). Novci so bili redko najdeni od hiše št. 8 proti Zidem (Šašel 1954a, 160). Cestar je večkrat našel novce na klancu med Trojanami in vasjo V Zideh (Šašel 1954a, 164). Najdejo se tudi na strmi­ni na južni strani vasi (vir 18/3). Iz zbirke Stare, Šmartno pri Litiji, je v NMS 12 novcev: Vespazijan, Septimij Sever, Gotik, Galijen, Severina in Constantin I. (Šašel 1954a, 163). 94 Stražar 1985. 95 Pečnik 1904. 96 Pečnik 1904, 131; Rutar 1890, 121; Istenič 2014, 97, sl. 105. 97 Vira 22 in 23. 98 Rutar 1890, 121. 99 Vir 1, vir 2. 100 Vira 22 in 23. 101 Bolta 1959, 131; 1960. than 8000 pieces.18 Jožef Repežič,19 a French principal, who served in Trojane in the years 1810–1814, suppos­edly had a 2000-piece collection of coins, which included 24 gold, 606 silver, and 1728 copper coins.20 In 1846, 14 coins (from Nero to Gratian) were discovered among other finds in Stojčevo (Fig. 2: area 6) by Dornik, parish priest in Šentgotard.21 Some coins were mentioned in 1891 by Prokop; they were predominantly from the sec­ond half of the 3rd century (mostly Emperor Probus) and to the time of Constantine.22 Thousands of coins from “the customs station on the border …” are mentioned by Pečnik.23 Ložar,24 Šašel,25 Stražar26 and others also mentioned finds of coins, which were discovered along the roads and in the gardens of Trojane and its closer surroundings. A rough sum of all numerically identifi­able coins is much more than 10,500 coins, which is a significant amount. SETTLEMENT (Fig. 19–22) ROAD AND CEMETERY Travellers leaving Regio X of Italy went from presentBlagovica towards Šentožbolt and, according to Pečnik, passed by “many Roman graves”.27 Their remains or loca­tions are not known. Ascending towards Trojane, their eyes met with a stately statue of a horseman (presumably Marcus Aurelius), which stood at Učak (Fig. 3: 14).28 Ac­cording to Pečnik,29 Rutar,30 and Schmid,31 the Roman 18 Pegan 2016, 56. 19 Stražar 1985, 84. 20 Slovenec 1939; Slovenski list 1939, 5–6. 21 Müllner 1878; 1879, 81. 22 Prokop 1891. 23 Pečnik 1904. 24 Ložar (1937) reports a find of coins at Stojc in 1936. 25 “In 1949, several Roman walls were excavated in Brvar‘s property, as well as postholes, other antiquities … occasionally even Roman coins. Behind the house no. 7 and north of the Trojane–Zide road, individual Roman coins can be found, but nothing else...” (Source 18/2). In some rare cases, coins were found in the area between the house no. 8 and V Zideh (Šašel 1954a, 160). A road mender discov­ered several coins on the slope between Trojane and V Zideh (Šašel 1954a, 164). Coins can be also found on the steep slope south of the village (Source 18/3). The Narodni muzej Slo­venije keeps 12 coins from the Stare collection from Šmartno pri Litiji: Vespasian, Septimius Severus, Gothicus, Gallienus, Severina, and Constantine I (Šašel 1954a, 163). 26 Stražar 1985. 27 Pečnik 1904. 28 Pečnik 1904, 131; Rutar 1890, 121; Istenič 2014, 97, fig. 105. 29 Sources 22 and 23. 30 Rutar 1890, 121. 31 Sources 1 and 2. 18 verjetno tvorijo grobišče, ki je pripadalo antičnemu Atransu. Drugje grobovi doslej niso bili okriti. Naselbinski plato je bil omejen s strmimi pobočji, ki padajo proti severu (potok Bolska) in jugu (Orehovica). Via publica je ohranjena na dveh mestih na Troja­nah (sl. 3: 16, 17) in vzhodno v Podzidu (sl 3: 19). Široka je bila 6 m.102 Pečnik je leta 1889 skiciral cesto skozi antični Atrans (sl. 4) in po njem sta jo povzela Schmid (vir 1, vir 2) in Šašel (vir 5, vir 7). Schmid je skiciral potek rimske ceste na vzhodnem delu antične naselbine (sl. 15), ki naj bi šla za hišo Trojane 7 (sl. 22), kjer je bil v šestdesetih letih še viden vsek domnevno rimske trase.103 Tlak rimske ceste je pri antični hiši IX (sl. 2: območje 3; sl. 5; 7) ugotovil tudi Šašel. Najverjetnejši potek rimske ceste skozi Atrans je zarisan na sliki 22. Glavni krak viae publicae je zavil proti severu, proti Celeji – po strmem pobočju proti trasi “stare ceste” (sl. 19) v dolini Bolske. Via vicinalis pa je šla verjetno s Trojanskega platoja naravnost proti vzhodu – proti današnjemu zaselku V Zideh in naprej proti Zasavju. Z vprašanjem poteka te ceste se nismo podrobneje ukvarjali.104 102 Bolta 1959, 131; 1960; Kajfež 1997; 2003. 103 Šašel 1954a, 160. Vir 18/4. Za več o rimskih cestah na tem območju glej: Šašel 1975 in Pirkmarjer 1985. road followed the southern side of the Učak hill, “right” (i.e. east) of the “old road” and turned into the Roman settlement of Atrans (Fig. 4). There was a necropolis to the west, just before the settlement. Several Roman graves were discovered in Konjšek’s fields (Fig. 2: 1) by Pečnik,32 and three graves in the course of the “old road” by Bolta (Fig. 3: 18): one with a sarcophagus and two with a con­struction made of bricks.33 The settlement plateau was bounded by steep slopes descending towards the north (the creek of Bolska) and south (the creek of Orehovica). The 6 m wide via publica survived in two locations near Trojane (Fig. 3: 16, 17), and in Podzid (Fig. 3: 19).34 Jernej Pečnik drew the road through Atrans in 1889 (Fig. 4), followed by Šmid (Sources 1, 2) and Šašel (Sources 5, 7). Schmid sketched the course of the Roman road in the eastern part of the Roman set­tlement (Fig. 15), presumably behind the house Trojane 7 (Fig. 20), where the cut of the supposed Roman road was still visible in the 1960s.35 Pavement of the Roman road was identified by Šašel near the House IX (Figs. 2: 3; 5; 7). The most likely course of the Roman road through Atrans is drawn in the plan (Fig. 3: 19; 22). 32 Sources 22 and 23. 33 The inhumation burial in a stone sarcophagus was empty, without grave goods. The second grave contained some burnt bones, a two-handled jug and a balsamarium. The third grave was just a construction made of tegulae and without grave goods (Bolta 1959, 131; 1960). 34 Bolta 1959, 131; 1960, 343; Kajfež 1997; 2003. 35 Šašel 1954 a, 160; Source 18/4. X stavba / building stavba s hipokavstom/ometom/freskami / building with hypocaust/plaster/fresco domnevni zid / presumed wall 50 m domnevni potek ceste / presumed direction of the road Sl. 20: Trojane, območja 2–3, 6–10. Hiše I–XI. Poskus rekonstrukcije s pomočjo skic W. Schmida (A. Farčnika), J. Šašla ter načrta M. Zupančič. Stopnja natančnosti je različna. M. = 1:1000. Fig. 20: Trojane, Areas 2–3, 6–10. Houses I–XI. Reconstruction according to the sketches by Walter Schmid (Anton Farčnik) and Jaroslav Šašel, and the plan by Mirina Zupančič. Th e degree of accuracy varies. Scale 1:1000. ZGRADBE Iz pregledane dokumentacije in literature smo uspeli izluščiti dvanajst tlorisov antičnih stavb, ki jih z veliko zanesljivostjo lahko lociramo in rekonstruiramo obseg antične naselbine. Skozi Atrans je tekla via publica in prerezala naselbino na severno in južno polovico (sl. 20; 22). Na severnem delu ceste so bile postavljene največje stavbe. Bile so zidane, deloma iz kamenja, deloma lesene. Objekti južnega dela so bili grajeni iz kamna, vezanega z malto. Imele so klet in vsaj eno nadstropje. Stene so bile praviloma poslikane s kvalitetnimi freskami (hiše I, III, IX, X; sl. 21). Prevladujejo geometrijski oz. cvet­lični motivi. Hipokavsti ali tubuli so omenjeni v hišah VI in IX severno od ceste in v hišah III, V in XII južno od ceste. Veliko je prostorov z ognjišči (po Schmidu). Verjetno je kateri izmed najdenih kompleksov pripadal kopališču (sl. 20). Objekti so bili prekriti z opečno kritino, zidovi pa predvsem v hiši št. IV zelo debeli: od 1,45 do 1,60 m (!). Razdalje med stavbami so ponekod 6 m, spet drugje le 1,35 m. Globine izkopa segajo tudi do 1,66 m. Večkrat se v stavbah omenjata dva estriha/tlaka, kar pomeni, da so bile prezidane, dozidane v več fazah. Glede na vse doslej analizirane vire (epigrafske in arheološke) lahko trdimo, da so prebivalci Atransa bivali v lepo opremljenih, s hipokavstom ogrevanih sobanah. Če je v hiši IV bila carina s carinskim uradom, morebiti domovanjem uslužbencev in celo taberno, kot piše Schmid v opisu prostorov, potem je bilo v zahodnih stavbah (hiši VII in VIII) morda skladišče. V pročelja carinskega objekta in/ali v bližnja svetišča so bili verjetno vzidani številni kamniti spomeniki (arhitekturni členi – grifon, posvetilni cesarski napis, …) ali pa so stali pred pročelji impozantnih stavb. A via vicinalis must have surely led through the set­tlement of V Zideh, although the question of the course of the Roman road to the Zasavje region has not been studied in detail.36 BUILDINGS By studying the documentation and literature, the authors were able to identify 12 Roman buildings that can be located with great probability (Fig. 20; 22). A via publica ran through Atrans and cut the settlement in the northern and southern half. The largest buildings were on the northern side of the road. The buildings were masonry constructions; part stone and part wood. The buildings in the southern part were made of stone and bound with mortar. They had a cellar and at least two storeys. The walls were usu­ally painted with fine frescoes (Houses I, III, IX, X; Fig. 21). The rooms of many houses (III, V, VI, IX, XII) were heated with a hypocaust, and many of the rooms had a fireplace (according to Schmid). It is likely that some of the discovered buildings belonged to a bath-house (Fig. 20). The buildings were covered with brick tiles and the walls were very thick, especially in the House IV: from 1.45 to 1.60 m (!). The distance between the houses is sometimes 6 m and sometimes only 1.35 m. The depths of the foundations reach up to 1.66 m. Two layers of pavement are usually mentioned in the buildings, which means they were rebuilt, constructed in several phases. Based on all sources (epigraphic and archaeological), the residents of the Roman Atrans lived in well-furnished hypocaust-heated rooms. If House IV was a customs office, perhaps the residence of the officials, and even a taberna, as is the opinion of Schmid in his description of the rooms, then Houses VII and VIII on the west might 36 More on the Roman roads in the area: Šašel 1975 and Pirkmarjer 1985. Sterilno osnovo terena predstavlja mornarsko moder lapor (po Schmidu),105 medtem ko Zupančičeva opozarja, da so ga uporabljali tudi za niveliranje – izrav­navo terena – in da ni nujno, da pod njim ni starejših kulturnih plasti.106 Južni rob naselbine je bil verjetno utrjen s škarpo – kontrafor pod Konjškom (območje 5, vir 12) – da bi preprečili drsenje objektov proti potoku Orehovica (sl. 8).107 Na 555 m n. m. poteka naravni prelom, ki je omejeval poselitev proti jugu.108 Glede na ostanek treh hiš (sl. 20; 22; hiše X, XI in XII), ki jih je izkopala Zupančičeva, je bila terasa verjetno pozidana prav do roba. Pomenljive so omembe žganinskih plasti, zoglene­lega žita, kar bi lahko povezovali s skladišči, carinskimi zalogami in velikim požarom (ali več požari), po katerem sta Mark Avrelij in Lucij Ver dala obnoviti pomembno stavbo (ali svetišče) na katero je bil vzidan cesarski napis (kat. št. 8). Lokacija napisa je znana iz leta 1874: območje 6, kamor smo umestili antično hišo V (sl. 2: 6). (J. Ž.) EPIGRAFSKI PODATKI Antične Trojane so bile kot mejna postojanka Norika vključene v davčni sistem publicum portorii Illyrici et ripae Thraciae, ki se je kasneje preimenoval v publicum portorii vectigalis X publica in je zajemal ogromno območje: Deseta regija Italije, Recija, Norik, Dalmacija (Ilirik), obe Panoniji, obe Meziji, vse Dacije in Trakijo.109 Na mejni postaji Atrans na Trojanah so pobirali portorium med Italijo in celotnim severovzhod­nim imperijem.110 Njeno vzporednico na drugi strani meje (v Italiji) bi lahko iskali v postaji Ad publicanos.111 Atrans je bila carinska postaja, Ad publicanos pa je lahko bila mansio ali statio. Mimo slednje bi torej morali tisti, ki so Italijo zapuščali; postaja Atrans pa je bila tudi prva carinska postaja za tiste, ki so vstopali v Norik z juga. Trojane so tako bile ena izmed šestih mejnih noriških carinskih postaj (Boiodurum – statio Boiodurensis; Pons Aeni – statio Enensis; v okolici Brixna/Bressanone; Loncium; 105 Vir 1. 106 Zupančič 1979, 17. 107 Vir 14. 108 Vir 15. 109 Za več o postajah tega davčnega teritorija glej pri Or­sted 1985, 255–258; De Laet 1949, 175–245 (o carinskih da­jatvah v obdonavskih provincah). 110 Orsted 1985, 283; prim. še Tatscheva 1996. 111 De Laet (1949, 188) je postajo lociral v kraj Gradišče pri Podpeči; Orsted (1989, 180–188) na Vače; Šašel (1975, 89) je za najbolj sprejemljivo lokacijo predpostavljal območje Krašnje in Lukovice. Prim še RINMS 105, 316–317. Ana Ples­tenjak meni, da bi lahko bila ta postaja v Blagovici (Plestenjak 2020, v tej knjigi. have been storehouses. Several monuments and archi­tectural elements – e.g. a griffin, an imperial dedication inscription – were probably built into the façades of the customs building and/or the nearby sanctuaries, or stood in front of imposing buildings. While the base layer is navy blue marl (according to Schmid),37 Zupančič points out that the same stone was also used for levelling the ground.38 The southern edge of the settlement was probably reinforced with a retain­ing wall and a buttress under Konjšek’s property in order to prevent the buildings from sliding down towards the Orehovica (Fig. 8).39 On the elevation of 555 m above sea level, the settlement was bounded to the south by a natural break in the terrain.40 Judging by the remains of three houses (Fig. 20; Houses X, XI and XII) excavated by Mirina Zupančič, the buildings most likely covered the terrace right to the edge. Notable are the mentions of burnt layers and charred cereal grains, which could be associated with storehouses, customs supplies, and a big fire (or fires), after which Mar­cus Aurelius and Lucius Verus had an important house (or sanctuary) with an imperial inscription (Cat. No. 8) rebuilt. The location of the inscription is known from the year 1874 – area 6 or House V (Fig. 2: 6).41 (J. Ž.) EPIGRAPHIC DATA As a station on the border of Noricum, Atrans was incorporated in the publicum portorii Illyrici et ripae Thraciae tax system, later renamed as publicum portorii vectigalis X publica. The publicum portorii encompassed a huge area: Regio X of Italy, Raetia, Noricum, Dalmatia (Illyricum), both Pannonias, both Moesias, all Dacias, and Thrace.42 At the border station in Trojane, portorium between Italy and the entire northeastern empire was collected.43 Its parallel on the other side of the border (in Italy) could have been the Ad publicanos station44 – 37 Source 1. 38 Zupančič 1979, 17. 39 Source 14. 40 Source 15. 41 From there it was taken to the house V Zideh 4 – Fig. 3: 23. The upper half is today in the Medobčinski muzej Kam­nik and the lower half in the Pokrajinski muzej Celje. 42 For further information on the stations of this tax ter­ritory see Orsted 1985, 255–258; De Laet 1949, 175–245 (the cited pages are dedicated to customs duties in the Danube provinces). 43 Orsted 1985, 283; cf. also Tatscheva 1996, 177–182. 44 De Laet 1949, 188 located the station in the village of Gradišče pri Podpeči; Orsted 1989, 180-188 in Vače; Šašel 1975, 89 saw as the most acceptable location the area of Krašnja and Lukovica; cf. also RINMS 105, 316–317. Ana Plestenjak thinks that this station could have been in Bla­ Larix – statio Bilachiniensis). Osnovna priložnost, ob kateri so portorium pobirali, je bil vsak premik med deželami z različnim pravnim statusom (npr. iz provinc v Italijo, iz barbarskih dežel v province).112 Kot pričajo napisi, je v Atransu najverjetneje od druge polovice 2. st. delovala carinska postaja (kat. št. 4, 15, 20, 22). Izpričani so zakupniki carinskega območja (kat. št. 22) na tej postaji in cesarski prokuratorji, ki so jo nadzorovali (kat. št. 1, 3, 10). Kot zakupniki na Trojanah so zapisani Juliji (kat. št. 22): Titus Iulius Capito, Caius Iulius Epaphroditus in Iulius Ianuarius.113 Po markoman­skih vojnah se je upravljanje carinskih postaj spremenilo. Te namreč niso bile več v zakupu zasebnikov, ampak so jih nadzorovali cesarski prokuratoji. Na Trojanah sta izpričana dva: Gaj Antonij Rufij (kat. št. 1, 13) in Gaj Antonij Julijan (kat. št. 3). Na napisih s Trojan najdemo uradnike, ki so služ­bovali na statio Atrantina: kontraskriptorje, skrutatorje, vilike in vikarje (kat. št. 1, 4, 13, 15, 20, 22). Vilicus stationis je bil eden izmed vilikov suženjskega statusa, ki so jih dodelili za zbiranje in upravo neposrednih davkov.114 Vicarius je bil med imperialni sužnji terminus technicus za pomočnike, zastopnike oz. pooblaščence ali oskrbnike. Scrutator predstavlja davčnega uradnika (revizorja), ki je preverjal deklaracije trgovcev ter pre­iskoval blago in osebe. Imeli so pravico preiskati prtljago in ocariniti vse, kar ni bilo nujno za na pot; le vojaki bi naj bili tega oproščeni. Scrutatores so bili rangirani nad contrascriptores,115 ki so opravljali finančno službo, pri kateri je šlo predvsem za kontroliranje in preverjanje (prevajamo jih tudi kot kontrolorje). Izpričana je sa-mo na postajah, ki so bile locirane na meji z Italijo.116 Viliki in ostali nižji uradniki na davčnih postajah so bili praviloma suženjskega statusa. Funkcijo zakupnika davčnega sistema (procurator oz. conductor) pa so glede na tridelno imensko formulo z imeni latinskega izvora opravljali rimski državljani, verjetno italskega izvora.117 O obstoju beneficiarske postaje govori gradbeni napis (kat. št. 19), kjer je omenjen beneficiarij 2. italske legije. Cesarja Mark Avrelij in Lucij Ver sta dala postaviti oz. popraviti pomembno stavbo v drugi polovici leta 168 (kat. št. 8). 112 Orsted 1985, 251. 113 Orsted (1985, 321–327) razloži povezanost Julijev (na osnovi preostalih napisov, na katerih so omenjeni), čas njih­ovega delovanja, predstavi pa tudi potencialne vezi z drugimi conductores, ki so svoje naloge opravljali sočasno. Zelo hipo­tetično poskusi sestaviti tudi njihove kariere. 114 Prim. Carlsen 1995, kjer avtor natančno razdela funk-cije vilikov; o njihovi vlogi v davčnem sistemu glej predvsem str. 46–55. Prim. še De Laet 1949, 380. RE Suppl. VII 1940, 1202–1203. Poleg izrecno zapi­sanih kontrolorjev se zdi verjetno, da je bil na fragmentarno ohranjenem napisu (kat. št. 10) izpričan še eden (prim. CIL III 4720: Titus Iulius Saturninus). 116 Orsted 1985, 288. Prim. še: De Laet 1949, 381. 117 Visočnik 2014, 284. Atrans was a customs station, while Ad publicanos could have been a mansio or a statio. If Ad publicanos was crossed by those who were leaving Italy, Atrans was the first customs station for those who were entering Noricum from the south. Atrans was thus one of the six border customs stations of Noricum (Boiodurum – statio Boiodurensis; Pons Aeni – statio Enensis; vicinity of Brixen/Bressanone; Loncium; Larix – statio Bilachiniensis). The basic opportunity to collect portorium was every movement from or to a region with a different legal status (e.g. from the provinces into Italy, from the barbaric lands into the provinces).45 Inscriptions testify to the fact that there was a cus­toms station in Atrans probably since the second half of the 2nd century on (Cat. Nos. 4, 15, 20, 22). There is evidence of those who contracted the collecting of the customs tax (Cat. No. 22) at this station, and of the im­perial procurators supervising it (Cat. Nos. 1, 3, 10). As conductores in Atrans are mentioned the Iulii (Cat. No. 22): they include Titus Iulius Capito, Caius Iulius Epaph­roditus, and Iulius Ianuarius.46 After the Marcomannic Wars, the management of customs stations changed: instead of being assigned to private conductores, they were now supervised by imperial procurators. Two of them are attested in Atrans: Gaius Antonius Rufius (Cat. Nos. 1, 13) and Gaius Antonius Iulianus (3). Inscriptions from Trojane mention officials who served in statio Atrantina: contrascriptores, scrutatores, vilici, and vicarii (Cat. Nos. 1, 4, 13, 15, 20, 22). Vilicus stationis was a vilicus with the status of a slave, assigned to collecting and managing direct taxes.47 Among imperial slaves, vicarius was a terminus technicus for assistants, representatives or agents, and keepers. Scrutator was a tax official (auditor) who revised merchants’ declarations and inspected goods and persons. They had the right to in­spect luggage and to charge customs duties for everything that was not essential for travelling; only soldiers were 48 exempt. Scrutatores were ranked above contrascriptores,who performed financial services, mostly controls and verifications (the term is also translated as controllers). There is evidence of this office only in the stations located govica (Plestenjak 2020, in this book).45 Orsted 1985, 251. 46 Orsted (1985, 321–327) explains the connection be­tween the Iulii (based on other inscriptions mentioning them), the time of their service, and presents potential ties with other conductores who performed their duties at the same time. He attempts a highly hypothetical reconstruction of their careers. 47 Cf. Carlsen 1995, where the functions of the vilici are analysed in detail; for their role in the tax system see in par­ticular pp. 46-55; cf. also De Laet 1949, 380. 48 RE Suppl. VII 1940, 1202–1203. In addition to the ex­pressly mentioned controllers it seems likely that the frag­mentarily surviving inscription (Cat. No. 10) mentions one more (cf. CIL III 4720: Titus Iulius Saturninus). Na obstoj enega ali več svetišč kažejo oltarji, dati­rani v drugo polovico 2. ali na začetek 3. st., posvečeni različnim božanstvom, rimskim in orientalnim. Izpri-čani so kulti Jupitra (kat. št. 3), Jupitra Kulminalisa (kat. št. 9), Marsa (kat. št. 15), Diane (kat. št. 23), Hekate (kat. št. 2), Libera Patra (kat. št. 13), Atransa (kat. št. 1), najverjetneje lokalnega božanstva (I)vantit (kat. št. 18), Mitre (kat. št. 22) in Noreje (kat. št. 4). Z gotovostjo lahko trdimo, da je bil najden samo en nagrobni spomenik (kat. št. 20): nagrobna stela za dečka Hilara, ki sta mu jo dala postaviti starša Sekun­dijan in Rufila. Tudi oče Sekundijan sodi med trojanske kontraskriptorje. Najverjetneje pa lahko govorimo še o treh fragmentih nagrobnih napisov (kat. št. 5, kat. št. 17 in kat. št. 21). Prvi je danes izgubljen. (J. V.) Zahvale Pri nastajanju zahtevnega pregleda izkopavanj in raz­ličnih posegov na Trojanah ter poskusu delne rekonstrukcije preteklosti ene pomembnejših mejnih postojank antike pri nas, so, tako in drugače, pomagali številni kolegi. Vsem, iskrena hvala. Brez vas in vašega prispevka, bi tega dela ne bilo mogoče predstaviti. Dolgoletna direktorica kamniška muzeja in razi­skovalka Trojan v šestdesetih in sedemdesetih letih dvajsetega stoletja Mirina Zupančič je prispevala dragocene podatke, tudi čas s terenskim ogledom Trojan in risbe z geolociranimi objekti – izdelal jih je Mark Zupančič. Milan Sagadin je dodal dokumentacijo, ki jo hrani ZVKDS, OE Kranj, in s terenskim ogledom pomagal locirati nekatere poselitvene točke. Dragan Božič je pomagal prebrati terenski dnevnik Walterja Schmida. Pri transkripciji Schmidovega dnevnika iz leta 1941 je sodelo­vala Irina Mitov (v letu 2017 je bila zaposlena v MMK). An-dreja Dolenc Vičič je ponudila vpogled v zapuščino Jaroslava Šašla, in arhiv IzA ZRC SAZU. Mateja Belak je vztrajno in potrpežljivo risala karte, svetovala in bila vseskozi v moralno podporo. Velika zasluga in zahvala velja Jani Horvat, ki je pomagala z literaturo, nasveti in korekturami, da je pričujoč prispevek ugledal luč sveta. Aleš Lazar (Geoservis d.o.o.) se je potrudil z umestitvijo antičnih objektov v prostor. Geodet­sko karto iz leta 1937 je prek Jurija Mlinarja (MOP) in Aleša Seliškarja posredovala Majda Mavec (GURS). Alenka Miškec je posredovala podatke o novčnih najdbah s Trojan, ki so za­beleženi v bazi Numizmatičnega kabineta NMS. Helena Bras Kernel je preko Janke Istenič omogočila pregled najdb, ki so v hrambi v NMS. Jure Krajšek pa je gradivo, ki ga je leta 1959 na Trojanah izkopal Lojze Bolta, izročil v trajno hrambo MMK. Izjemen prispevek k poznavanju Trojan in vpogled v čase po Schmidu je ponudil domačin Janez Ocvirk. Svojo zgodbo in spomine na preteklost Trojan, kot sta jo doživela prek svojih prednikov, sta pripovedovala: Zdenka (rojena Brvar) in njen mož Lado Goričan. Brez vseh naštetih in še koga, ki je nenamerno izpadel iz opisanega seznama, tega dela ne bi bilo mogoče napisati. Vsem iz srca hvala. on the border with Italy.49 Vilici and other lower officials in tax stations were usually slaves. The function of the contractor of the tax system (procurator or conductor) was, judging by the three-name formula with names of Latin origin, performed by Roman citizens, probably of Italian origin.50 The existence of a beneficiary station is attested by a building inscription which mentions a beneficiary of the legio II Italica (Cat. No. 19). In the second half of the year 168, emperors Marcus Aurelius and Lucius Verus had an important building constructed or repaired (Cat. No. 8). The existence of one or several sanctuaries is indi­cated by altars, dated to the second half of the 2nd century or the beginning of the 3rd century, and dedicated to vari­ous deities, either Roman or Oriental. There is evidence of the cults of Jupiter (Cat. No. 3), Jupiter Culminalis (Cat. No. 9), Mars (Cat. No. 15), Diana (Cat. No. 23), Hecate (Cat. No. 2), Liber Pater (Cat. No. 13), Atrans (Cat. No. 1), probably local deity (I)vantit (Cat. No. 18), Mithras (Cat. No. 22), and Noreia (Cat. No. 4). It can be claimed with certainty that only one funer­ary monument has been found (Cat. No. 20): a funerary stele for Hilarus, a boy, by his parents Secundianus and Rufilla. The father Secundianus was one of the contras-criptores at Trojane. Two more fragments of funerary inscriptions are deemed to have existed: Cat. No. 5 and Cat. No. 17, which are now lost. (J.V.) Translation: Meta Osredkar 49 Orsted 1985, 288; cf. also De Laet 1949, 381. 50 Visočnik 2014, 284. PISNI VIRI IN DOKUMENTACIJA [WRITTEN SOURCES AND DOCUMENTATION] UVOD Za potrebe projekta o rimskih naseljih (Trojane) je bila na Inštitutu za arheologijo ZRC SAZU pregledana zapuščina Jaroslava Šašla, kopije arhiva Centralne komisije na Dunaju (k. k. Central-Commission zur Erforschung und Erhaltung der kunst- und historischen Denkmale, originale hrani Arhiv Republike Slovenije, kopije arhiv IzA ZRC SAZU) ter kopije pisem Jerneja Pečnika Karlu Dežmanu (originale hrani ARS, kopije arhiv Iza ZRC SAZU). Universalmuseum Joanneum Graz v svojem arhivu hrani dokumentacijo Walterja Schmida, ključnega izkopa­valca Trojan. Raziskoval jih je leta 1941 in 1942. Njegovi dnevniki, pisma, nekaj fotografij in drugih dokumentov je bilo skeniranih in digitaliziranih v sklopu projekta InterArch–Steirmark (sofinanciran je bil z EU sredstvi) – progra-ma čezmejnega sodelovanja Slovenija–Avstrija 2007–2013. Rezultati so bili publicirani v katalogu, ki ga je uredila Barbara Porod, V novi luči / Ans Licht gebract (Porod et al. 2013). Arhivski dokumenti in fotografije predmetov so bili objavljeni na spremljajočem DVD-ju ter postali javno dostopni. Arheološko gradivo s Trojan, Ptuja in drugih slovenskih najdišč je bilo vrnjeno Republiki Sloveniji (glej vir 11) in je hranjeno v pristojnih muzejih. Kopije nekaterih Schmidovih originalnih dokumentov se nahajajo tudi v arhivu IzA ZRC SAZU in so del zapuščine Jaroslava Šašla. Šašel je v arhivu graškega Joanneuma očitno pregledal Schmidovo zapuščino in nekaj dokumentov/načrtov ali skic kopiral, fotografiral ali prerisal ter komentiral. Dokumentacijo izkopavanj MMK (risbe z geolociranimi objekti, izdelal jih je Mark Zupančič) začasno hrani Mirina Zupančič; dokumentacijo najnovejših izkopavanj pa ZVKDS, OE Kranj in izkopavalci. V naše besedilo o Trojanah so vključeni samo dokumenti iz Seznama virov. Za razlago okrajšav, ki so uporabljene v seznamu, glej v uvodu v poglavju Katalog arheoloških območij. SEZNAM VIROV Vir 1 Walter Schmid (Sl. 11; 13; 14) – Terenski dnevnik, Atrans 1941 (rokopis v nemščini; izčrpane strani 32–48, 63–66). Dnevnik izkopavanj W. Schmida leta 1941 na Trojanah. Med besedilom so skice izkopanih antičnih objektov (hiše I, II, III in IV)* opisi plasti, globine izkopov, debeline zidov ipd. Hrani: arhiv UMJG (W. Schmid). Pojasnilo: * Schmid je izkopane temelje hiš sprva oštevilčil od zahoda proti vzhodu, naraščajoče: zahodni takrat odkriti objekt je nosil št. I – skrajno vzhodni (s polkrožnim zaključkom) pa št. III. Pozneje je sledilo delno preštevilčenje (glej poglavje Katalog arheoloških območij). Vir 2 Walter Schmid (Sl. 6; 17) – Terenski dnevnik, Atrans-Libna-Neg* 1942 (rokopis v nemščini; izčrpane strani 14–24). Dnevnik izkopavanj W. Schmida leta 1942 na Trojanah. Med besedilom so skice izkopanih antičnih objektov, opisi plasti, globine izkopov, debeline zidov ipd. Pomemben je podatek, da je Schmid tudi leta 1942 izkopaval na Trojanah, kjer je našel obrise še treh objektov (hiše št. VI, VII in VIII),** in da je na bližnjem Učaku izkopal 12 sond. Hrani: arhiv UMJG (W. Schmid). Pojasnili: * Neg pomeni Negova ( v dnevniku so opisane še Schmidove raziskave v drugih krajih). ** Antične stavbe VI–VIII smo locirali s pomočjo informatorja Janeza Ocvirka, domačina, ki je razložil zgo­ dovino in pozicijo za lociranje izkopov pomembnega gasilskega doma (Trojane 4). Vir 3 Anton Farčnik* (Sl. 10) – Načrt. Tlorisi treh rimskodobnih objektov, ki jih je leta 1941 na Trojanah izkopal W. Schmid (milimetrski papir; format A3; M. = 1: 200). Avtor načrta je A. Farčnik, tlorise hiš (I, III in IV) je izrisal po Schmidovem naročilu. Načrt je verjetno nastal v letu 1942, po izkopavanjih jeseni 1941. Posamezni prostori v notranjosti stavb so oštevilčeni (arabske številke), navedene so debeline zidov, ponekod vrste tlakov. Na originalno Farčnikovo risbo je rdeče rimske številke ob hišah dopisal že Schmid.** Hrani: arhiv UMJG (W. Schmid). Pojasnili: * Anton Farčnik je bil uslužbenec cestnega gradbenega urada Štajerskega okrožja s sedežem v Celju in je s Schmidom sodeloval pri izrisu načrta tlorisov stavb s Trojan in sosednje vzpetine Učak. ** Rdeče rimske št. (I, III in IV) predstavljajo preštevilčenje hiš (glej poglavje Katalog arheoloških območij). Vir 4 Anton Farčnik (korespondenca Farčnik – Schmidu) – Pismo A. Farčnika (Celje, 24. 12. 1942; rokopis) v katerem W. Schmidu sporoča, da je narisal trdnjavo pravo­kotne oblike na Učaku ter štiri antične stavbe na drugi terasi na Rifniku. S Trojanami bo počakal, dokler mu Schmid ne pošlje topografskega načrta. Hrani: arhiv UMJG (W. Schmid). Vir 5 Walter Schmid (korespondenca Schmid – Farčniku) – Pismo W. Schmida A. Farčniku (Graz, 17. 2. 1943; tipkopis). Schmid piše, da je prosil Franca Kralja iz Trojan št. 5, da bi za Farčnika kopiral katastrski načrt Trojan. Sprašuje ga, če je narisal tlorise objektov s Trojan v merilu 1:200. Prosi ga za nekaj popravkov na načrtu Učaka. Hrani: arhiv UMJG (W. Schmid). Vir 6 Walter Schmid – Besedilo z naslovom Atrans. Erforschung einer römischen Gränzstation zwischen Italien und Noricum am Tro­janapaß (format A4; tipkopis; 2 strani besedila in 5 podnapisov za slikovno gradivo). Verjetno priprava za objave* (brez slikovnega gradiva) v časopisih Jutro in Marburger Zeitung. Hrani: arhiv UMJG (W. Schmid). Pojasnilo: * Glej Jutro 1941 in Schmid 1941. Vir 7 Konjšek (korespondenca družina Konjšek – Schmidu) (Sl. 19) – Razglednica, ki so jo Konjškovi poslali Walterju Schmidu v Gradec (1. 5. 1943). Pogled na Trojane z vzhoda. Pošiljajo mu lepe pozdrave in željo, da se kaj oglasi pri njih. Hrani: arhiv UMJG (W. Schmid). Pojasnilo: Na skrajno levem robu fotografije se vidi cesta, ki bi lahko predstavljala ostanek trase stare rimske ceste, ki se je na vzhodu relativno strmo spustila v dolino reke Bolske in približno na traso “stare” ceste, ob kateri so bili ostanki rimske dokazani – prim. sl. 3: 19 (Kajfež 1997) – pri gostilni Potrbin (V Zideh 10). Potek ceste za hišo Trojane 7 proti potoku Bolska je predvideval tudi Šašel (1954, 160), že prej pa npr. tudi Pečnik (glej vir 23). Vir 8 Walter Schmid (Sl. 15) – Pozicijska skica razmestitve antičnih objektov izkopanih leta 1941 na Trojanah. Priloga v Schmidovi terenski dokumentaciji. Stavbe niso oštevilčene.* Usmeritev proti jugu. Hrani: arhiv UMJG (W. Schmid). Pojasnilo: * Gre za stavbe, ki jih je leta 1941 izkopal Schmid (aktualno številčenje I–IV). Na skici je umeščena tudi stavba s hipokavstom (skicirani oboki!), najdena že leta 1936 (že pred Schmidovimi izkopavanji; pri aktualnem številčenju je to hiša V). Vir 9 Walter Schmid – Poročilo, Bericht über die Ergebnisse der Grabung im Jahre 1941 und 1942. An die Unterabteilung II der Reichssertthlterei Steiermark, Graz, 25. 3. 1943 (tipkopis; format A4). Schmid poroča, kje na območju slovenske Šta­jerske je kopal, kdaj in kaj je našel. Na Trojanah “… je našel kasarno II. ital. legije, carinarnico, pošto z gospodarskimi pomožnimi objekti …”, kot piše, skupno 6 stavb in poznoantično utrdbo na Učaku. Hrani: arhiv UMJG (W. Schmid). Vir 10 Niko Kralj* (korespondenca Kralj – Schmidu) – Pismo Kralja W. Schmidu (Trojane, 18. 11. 1941; rokopis; list format A4). Kralj piše, da je razdalja med Schmidovima objektoma 1 in 2 le 1,30 in ne 1,35 m, kot je pomotoma prej sam izmeril in, da debelina južnega zidu objekta 2 znaša 75 cm. Iz skice je razvidno, da sega južno pročelje objekta 2 za 3,60 m južneje od linije objekta 1. Kralj je pismo podpisal z: od vašega Nikita. Hrani: arhiv UMJG (W. Schmid). Pojasnilo: * Niko Kralj (1920–2013), znani slovenski arhitekt, industijski oblikovalec znemenitega stola Rex. Doma je bil v Zavrhu pri Trojanah. Očtino je sodeloval pri izkopavanjih antičnega Atransa oz. je pomagal Schmidu z izmerami ali skicami zidov. Vir 11 Korespondenca R Slovenija – R Avstrija – Pismo o nameri med Republiko Slovenijo in Republiko Avstrijo o vračilu arheoloških predmetov, ki so bili izkopani na območju Slovenije v času nemške okupacije; poslano Ministrstvu za kulturo RS (št. 6241-2/20012– MIZKS/1) 20. junija leta 2012. Materialna zapuščina W. Schmida s Trojan je bila prevzeta po izteku razstave V novi luči v Pokrajinskem muzeju Murska Sobota, 18. avgusta 2014 in je od takrat v trajni hrambi v MMK. Skupno je inventariziranih 140 predmetov ali njihovih delov, med njimi 8 novcev (inv. št. MMK 3500.1–3572). Hrani: arhiv MMK. Vir 12 Jaroslav Šašel – Poročilo o novi gradnji na Trojanah, Ljubljana, 12. 5. 1961 (tipkopis; list format A4). Šašel poroča o izkopu pri Konjškovih na Trojanah: zaradi širitve gostinskega objekta, postavitve mizarske delavnice in izgradnje škarpe. Ob delihza škarpo je bil odkrit antični kontrafor, debeline 2 m. Dne 12. 5. 1961 sta si teren ogledala Šašel in Lojze Bolta. Šašel je predlagal sklic pristojnih na najdišču, nadzor nad gradbenimi deli in prekinitev del, če bi bilo to treba. Opisuje tudi, da je bil sakrofag, ki ga je bil izkopal leto poprej, postavljen za pošto v Trojanah in da se je začelo njegovo razbijanje. Hrani: arhiv IzA ZRC SAZU: AO, Šašel; [ARH 1/1]). Vir 13 Jaroslav Šašel – Poročilo o obisku Trojan; Ljubljana, 14. 5. 1961, poslano Zavodu za Spomeniško varstvo LRS (tipkopis; list format A4). Šašel poroča, da je bil namen obiska Trojan (14. 5. 1961) srečanje in pogovor z delavcem Ivanom Balo­hom iz Učaka 13, kovaškim vajencem, ki je leta 1942 pomagal Schmidu izkopavati na Učaku. Hrani: arhiv IzA ZRC SAZU: AO, Šašel; [ARH 1/2]). Pojasnilo: Baloh je Šašlu omenil tudi kvadratno stavbo velikosti 15 × 15 m na lokaciji Zadružnega doma z ognjiš-čem v sredini in več sobami, ki jo je Schmid kopal leto poprej, torej leta 1941. Po aktualnem številčenju je to hiša IX. Vir 14 Jaroslav Šašel (Sl. 8) – Skica osrednjega dela Trojan leta 1961 (svinčnik; format A5). Nanaša se na Šašlovo poročilo dne 14. 5. 1961 (= vir 12). Vrisani so takratni objekti: Stojčevo, nova mizarska delavnica, gostilna Konjšek, stara mizarska delavnica, bodoče prenočitvene kapacitete, parkirni prostor in škarpa , ki jo je za potrebe nove delavnice gradil Konjšek (na Konjškovem –Trojane 11). Na škarpi je rdeče označen ob delih odkrit rimski kontrafor. Skica je usmerjena proti jugu. Hrani: arhiv IzA ZRC SAZU: AO, Šašel; [ARH 1/3]). Vir 15 Jaroslav Šašel (Sl. 5) – Skica, nastala je leta 1961 (svinčnik; format A5). Nanaša se na Šašlovo poročilo 14. 5. 1961 (= vir 12). Ski-cirana je takratna situacija s hišami na Trojanah; označena so območja Schmidovih izkopavanj leta 1941 in 1942. Šašel je s križci (x) označil zemljišče, ki ga je Schmid izkopal ali sondiral z jarki (S). Vrisan je še domnevni potek rimske ceste in južni rob naselbine. Skicirane so še druge lokacije domnevnega zidovja (A, in pri hiši [Trojane] 8) – odkritega že po Schmidovih izkopavanjih. Skica je usmerjena približno proti severu. Šašel je lokacije umestil s pomočjo domačina Jožefa Brvarja. Hrani: arhiv IzA ZRC SAZU: AO, Šašel; [ARH 1/4]). Vir 16 Jaroslav Šašel – Načrt: fotografija* (črno-bela; vel. 22,5 × 17,2 cm); sestavljena iz posnetkov dveh dokumentov iz Schmidove zapuščine v arhivu UMJG (vir 3 in vir 8; oba se nanašata na Schmidova izkopavanja leta 1941 na Trojanah). Hrani: arhiv IzA ZRC SAZU: AO, Šašel; [ARH 2/5]). Pojasnilo: * Šašel je v UMJG skupaj fotografiral originalni Farčnikov načrt antičnih objektov (tlorise hiš I, III in IV = vir 3) in skico razmestitve izkopanih antičnih objektov (vir 8). Vir 17 Jaroslav Šašel – Načrt (paus papir; vel. 69 × 50 cm), sestavljen iz kopij več delov drugih dokumentov. Nanaša se na na Schmi­dova izkopavanja leta 1941 na Trojanah. Ob strani so napisani Šašlovi komentarji. Hrani: arhiv IzA ZRC SAZU: AO, Šašel; [ARH 5/1]). Pojasnilo: Šašel je v UMJG v prerisal originalni Farčnikov načrt antičnih objektov (tlorise hiš I, III in IV = vir 3) in dodal še druge risbe, ki jih je našel v Schmidovi zapuščini (pozicijska razmestitev antičnih objektov po Schmidu – vir 15 in detaj hiše I, vir 23 – skica poteka rimske ceste skozi Atrans s komentarjem: “Šmidov preris iz Pečnikovega rk v NM”, pismo Kralja Schmidu s skico odmika stavbe I od objekta II – vir 10 ter skici izjemno debelih zidov, ki pa se jih ni dalo umestiti v generalni načrt!). Vir 18 Jaroslav Šašel – Kartotečni listki (tipkopisi in rokopisi; format A5). Šašlova kartoteka za posamezne lokacije najdb na Trojanah in okolici. Večino podatkov je pridobil ob topografskih pregledih leta 1953. Hrani: arhiv IzA ZRC SAZU: AO, Šašel; [ARH 1/7–1/18]). Pojasnilo: Ugotovitve in spoznanja s topografije Trojan je Šašel objavil (1954a). V besedilu smo uporabili: Vir 18/1 – kartotečni list Trojane (2 str.). – Opis dozdajšnjih najdb in izkopavanj, omemb, literature za Trojane (ARH 1/7). Vir 18/2 – kartotečni list Trojane h. št. 7 (last Brvar Franc). Šašel piše, da so “leta 1949 našli več rimskih zidov, lukenj za bruna in drobnarije materialne kulture… Na njivah nad “staro cesto” (to je vsek prejkone srednjeveške danes opuščene ceste, ki se na južnem pobočju Bolske, vzhodno za omenjeno hišo, drobro razloči), vzhodno od hiše št. 7 ter severno od ceste Trojane-Št.Gotard, najdejo včasih posamične rimske novce, drugega nič.” (ARH 1/12). Vir 18/3 – kartotečni list V Zideh in Podzid (2 str.). – Omenja, da domačini ne pomnijo, da bi V Zideh kdaj v preteklosti naleteli na kakšne zidove – vsi kamniti spomeniki so bili vzidani v hišo št. 4 – pri Devsu. Podzid – vidni ostanki dolgega zidu pod cerkvijo sv. Mohorja in Fortunata (ARH 1/15). Vir 18/4 – kartotečni list Trojane (ceste) – 2 str. Šašel popisuje že potek prazgodovinske jantarne poti preko Trojan, ki po njegovem v prazgodovini ni bila toliko v rabi. Opisuje potek rimske ceste preko Trojan, ki naj bi na Trojane prišla mimo Učaka. Pri kmetiji p. d. Cvet je cesta usmerjena proti Šentožboltu po južnem bregu Radomlje. Proti vzhodu do: Št. Gotarda-Zaplane-Preselda-Ostgrovice-Tabora in naprej (ARH 1/18). Vir 19 Jernej Pečnik (korespondenca Pečnik – Dežmanu) – Pismo Pečnika Karlu Dežmanu, Krško, 17. 4. 1886 (rokopis; format A4). Hrani: ARS: SI AS 854, Dežman Dragotin, 1786–1895. (Pregledana je bila kopija iz arhiva IzA ZRC SAZU, TA 321). Vir 20 Jernej Pečnik (korespondenca Pečnik – Dežmanu) – Pismo Pečnika Karlu Dežmanu, Krško, 24. 4. 1887 (rokopis). Hrani: ARS: SI AS 854, Dežman Dragotin, 1786–1895. (Pregledana je bila kopija iz arhiva IzA ZRC SAZU, TA 321). Vir 21 Geodetski načrti (kopije), leto 1959 Geodetski projekt trase ceste Ljubljana–Celje, C–56, odsek Št. Ožbolt–Trojane (papir, vel. 388 × 30 cm). Na načrtu so vrisani profili* ceste. Hrani: Uprava za ceste LRS, Cesta I/10 Maribor–Trst, odsek Trojane; glavni projekt. Situacija 1:1000. ing. AvanzoLeo, Willenport Ludvik, ing. Jerin Albin, januar 1959. (Pregledana je bila kopija iz arhiva IzA ZRC SAZU: AO, Šašel). Pojasnilo: * Profile je opisal in objavi Lojze Bolta (1959; 1960) – vir za lociranje dveh ostankov rimske ceste in treh grobov pri Trojanah (prim. sl. 3: območja 16, 14, 18). Vir 22 Jernej Pečnik (korespondenca Pečnik – Centralni komisiji na Dunaju) Pismo Pečnika Centralni komisiji, april 1889. Hrani: ARS, fond AS 1100, C. kr. spomeniški urad, K 61/15. (Pregledana je bila kopija iz arhiva IzA ZRC SAZU). Vir 23 Jernej Pečnik (Sl. 4) Pečnikova karta arheoloških najdišč Radeče in Zagorje (Ratschach und Sagor) in opis, leta 1889 (risba in roko-pis). Označena so prazgodovinska in rimska najdišča, tudi za Trojane in okolico. Hrani: ARS, fond AS 965, Pečnik Jernej, fasc. 3. (Pregledana je bila kopija iz arhiva IzA ZRC SAZU). Vir 24 (Sl. 9) Fotografija, nastala leta 1936. V ozadju sta vidna oboka hipokavsta, izkopanega leta 1936 na Trojanah, ob gradnjitemeljev za svinjak na posestvu Štefana Brvarja (p. d. Stojc, takrat hiša št. 5; danes Trojane 10). Hrani: dokumentacija MMK. Pojasnilo: Gre za stavbo V (aktualno številčenje). Na fotografiji sta brata; prvi z leve je verjetno Štefan ali Ciril, drugi pa Jernej Brvar. (Pri identifikaciji je pomagala nečakinja Zdenka Brvar, poročena Goričan). Vir 25 Mirina Zupančič (Sl. 16) Načrt (datoteka v formatu DWG; nastanek 21. 1. 2017). Kompozitni načrt, izdelan s pomočjo terenske dokumentacije, nastale ob zaščitnih izkopavanjih Mirine Zupančič (MMK) v letih 1966, 1970 in 1975 na Trojanah. Umestitev antičnih objektov je izvedel Mark Zupančič. Pozicionirani so objekti, ki so bili izkopani (aktualno številčenje: hiše X, XI in XII) ali ponovno odkriti (hiša I; Schmidov izkop leta 1941). Objekti so umeščeni v prostor, poudarjene so za lokacijo pomembne današnje stavbe in dodane hišne številke. Usmeritev proti severu. M. = 1:1000 (Podlaga: DOF, © GURS, 2017). Hrani: MMK. (J. Ž.) Okrajšave / Abbreviations CIL = Corpus inscriptionum Latinarum. EDCS = Epigraphik-Datenbank Clauss / Slaby (Service provi­der / skrbnik: Manfred Clauss) [http://db.edcs.eu/epigr/ epi_de.php]. HD = EDH, Epigraphische Datenbank Heidelberg (Service pro­vider / skrbnik: Heidelberger Akademie der Wissenschaften [http://edh-www.adw.uni-heidelberg. de/home?&lang=de]. ILJug = A. et J. Šašel, Inscriptiones latinae quae in Iugoslavia inter annos MCMXL et MCMLX repertae et editae sunt. – Situla 5, Ljubljana 1963. lupa = UBI ERAT LUPA – F. und O. Harl, www.ubi-erat-lupa. org (Bilddatenbank zu antiken Steindenkmälern). RINMS = M. Šašel Kos, The Roman Inscriptions in the National Museum of Slovenia / Lapidarij Narodnega muzeja Slovenije. – Situla 35, Ljubljana 1997. BEZLAJ, F. 1958, Predslovanski ostanki v slovenščini. – Naša sodobnost 8–9, 673–693. BOLTA, L. 1959, Trojane. – V / In: Arheološki pregled 1, 130–133. BOLTA, L. 1960, Trojane. – Varstvo spomenikov 7, 343. CARLSEN, J. 1995, Vilici and Roman Estate Managers until AD 284. – Analecta Romana instituti Danici. Suppl. 24. CVIKL ZUPANČIČ, M. 1966, Trojane. – Varstvo spomenikov 11 (1967), 130. DE LAET, S. J. 1949, Portorium. Étude sur l' organisation douaniere chez les Romains, sortout a l' époque du Haut--Empire. – Brugge. DESCHMAN, C. 1866, Münzen. Verzeichniss der seit 1. Mai 1862 bis Ende April 1866 eingegangenen Geschencke und der sonstigen Erwerbungen des krainischen Landesmu­seums. – Mitteilungen des Musealvereins für Krain 1, 274. DIMITZ, A. 1874, Geschichte Krains 1, Ljubljana. DOBÓ, Á. 1940, Publicum Portorium Illyrici. – Dissertationes Pannonicae, Ser. 2, Fasc. 16. – Budapest. Glasilo 1939 = s. a., Zgodovinske Trojane, Glasilo K. S. K. jed-note, letnik 25, št. 52, 1939 (27.12.1939), str. 6, Cleveland. HITZINGER, P. 1855, Starozgodovinski pomenki. Rimska cesta od Emone na Celejo. – Kmetijske in rokodelske novice, letnik 13, št. 64, (11. 8. 1855), str. 254. HORVAT, J. 1999, Roman Provincial Archaeology in Slovenia Following the Year 1965: Settlement and Small Finds / Rimska provincialna arheologija v Sloveniji po letu 1965: poselitvena slika in drobna materialna kultura. – Arheološki vestnik 50, 215–257. ISTENIČ, J. 2014, Rimske zgodbe s stičišča svetov. - Ljubljana. JOSIPOVIČ, D. 2003, Poročilo o arheoloških delih na lokaciji Trojane, na trasi AC Blagovica–Vransko, predor Trojane. Kranj, 25.11.2003. – Poročilo / Report; Arhiv ZVKDS, OE Kranj (neobjavljeno / unpublished). Jutro 1940 = s. a. (“Trojanec”), Od Atransa do Trojan. – Jutro. Ponedeljska izdaja, leto 11, št. 4, [“Jutro” XXI., št. 16a], 1940 (22. 1. 1940), str. 8, Ljubljana. Jutro 1941 = s. a., Rimska obmejna postojanka v Trojanah. Nova odkritja vseučil. prof. dr. Valterja Schmida. – Jutro, leto XX, št. 276 (25. 11. 1941), str. 3, Ljubljana. KAJFEŽ, T. 1997, Poročilo, Zaščitno arheološko sondiranje na lokacije “Trojane cesta”, Ljubljana, julij 1997. – Poročilo / Report; Arhiv ZVKDS, OE Kranj (neobjavljeno / unpu­blished). KAJFEŽ, T. 2003, Trojane cesta. – V: B. Djurić et al., Zemlja pod vašimi nogam. Arheologija na avtocestah Slovenije. Vodnik po najdiščih, 267–268, Ljubljana (= The Earth Beneath Your Feet. Archaeology on the Motorways in Slovenia. Guide to Sites, Ljubljana 2004). KLEMENC, J. 1951, Trije novi napisi iz Celja in okolice. – Arheološki vestnik 2, 124–133. KLEMENC, J. 1955, Zgodovina Emone. – V / In: Zgodovina Ljubljane 1. Geologija in arheologija, 331–426, Ljubljana. KNIFIC, T. 1991, Arheologija o prvih stoletjih krščanstva na Slovenskem. – V / in: T. Knific, M. Sagadin, Pismo brez pi-save. Arheologija o prvih stoletjih krščanstva na Slovenskem / Carta sine litteris. The Archaeology of the first Centuries of Christianity in Slovenia, 11–32, Ljubljana. KOS, M. 1975, s. v. Trojane. – V / In: Gradivo za historično topo­grafijo Slovenije (za Kranjsko do leta 1500) 2, 648, Ljubljana. KOŠIR, U. 2018, Poročilo o arheoloških raziskavah ob gradnji na Trojanah (parc. št. 560/5 k. o. Trojane). – Poročilo / Report; Arhiv ZVKDS, OE Kranj (neobjavljeno / unpublished). LOŽAR, R. 1937, Rimska ogrevalna naprava na Trojanah. – Glasnik muzejskega društva za Slovenijo 18, 54–57. LOŽAR, R. 1937–1940, Ostanki bronaste konjske statue s Trojan – Prispevek k poznavanju monumentalne rimske bronaste plastike. – Vjesnik hrvatskog arheološkog društva N. S. 18–21, 353–367. MÜLLNER, A. 1878 – V / In: Mitteilungen des Musealvereines für Krain NF IV, str. LXXXVI. MÜLLNER, A. 1879, Emona. Archäeologische Studien aus Krain. – Laibach. MÜLLNER, A. 1900, Typische Formen aus den archäologischen Sammlungen des krainischen Landersmuseums “Rudol-finum” in Laibach in photographischen Reproductionen. – Laibach. MUŠIČ, B. 1997, Trojane – V Zideh: geofizikalna raziskava. Poročilo. Ljubljana. – Poročilo / Report; Arhiv ZVKDS, OE Kranj, Center za preventivno arheologijo (neobjavljeno / unpublished). OGRIN, M. 1996, Poročilo o opravljenem intrasite podpovršin­skem pregledu na delu Zavarovanega arheološkega območja V Zideh – Trojane. – Poročilo / Report; Arhiv ZVKDS, OE Kranj (neobjavljeno / unpublished). OROŽEN, F. 1902, Vojvodina Kranjska 4. – Ljubljana. OROŽEN ADAMIČ, M., D. PERKO, D. KLADNIK et al. 1995, Trojane. – V / In: Krajevni leksikon Slovenije, Ljubljana, 397. ORSTED, P. 1985, Roman imperial economy and Romanization. A study in Roman imperial administration and the public lease system in the Danubian provinces from the first to the third century A.D. – Copenhagen. PEČNIK, J. 1904, Prazgodovinska najdišča na Kranjskem. – Izvestja muzejskega društva za Kranjsko 14, 130–131. PEGAN, E. M. 2016. – Prispevki k zgodovini numizmatike v Sloveniji. Ob 75-letnici numizmatičnega društva Slovenije. – V / In: Slovensko šolstvo v luči medaljerstva od 16. stoletja naprej in nagrajevanje šolskih uspehov, Ljubljana. PIRKMAJER, D. 1985, Rimska cesta Emona–Celeia. Odsek itinerarske ceste Aquileia–Donava. – Celjski zbornik 1985, 159–176, Celje. PLESTENJAK, A. 2020, Blagovica. – V/In: J. Horvat, I. Lazar, A. Gaspari (ur. / eds.), Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Arhaeologici Sloveniae 40, 231–247. POROD, B. (ur. / ed.) et al. 2013, V novi luči. Arheološka dediš-čina slovenske Štajerske iz Univerzalnega muzeja Joanneum/ Ans Licht gebracht. Archäologisches Erbe der Štajerska aus dem Universalmuseum Joanneum. Katalog. – Graz. PROKOP, 1891, Notizen. – Mittheilungen der k. k. Central- Commission zur Erforschung und Erhaltung der Bau­denkmale 17, 239. RE = Pauly-Wissowa-Kroll-Mittelhaus-Ziegler, Realencyclo­pädie der classischen Altertumswissenschaft. RECHFELD, F. J. 1948, Copia eines Manuscriptes des Valentin Vodnik. Itinerarium 1808. – Mittheilungen des historischen Vereines für Krain, 87–93. REPANŠEK, L. 2016, Keltska dediščina v toponimiji jugovzhod­nega alpskega prostora (Celtic Legacy in the Toponymy of South-Eastern Alps). - Linguistica et philologica 33. RUTAR, S. 1890, Prazgodovinska in rimska izkopavanja na Slovenskem l. 1889. – Letopis Matice Slovenske za leto 1889, 117–130. RUTAR, S. 1891, Prazgodovinske in rimske izkopine po Slo­venskem 1. 1890. – Letopis Matice Slovenske za leto 1890, 182–209. Slovenec 1939 = s. a., Zgodovinske Trojane. – Slovenec, letnik 67, št. 231, 1939 (8. 10. 1939), str. 10, Ljubljana. Slovenski list 1939 = s. a., Zgodovinske Trojane. – Slovenski list, letnik 10, št. 134, 1939 (25. 8. 1939), str. 5–6, Buenos Aires. SCHMID, W. 1941, Atrans – eine römische Grenzstation. Aus­grabungen auf der Höhe des untersteierischen Trojanapas­ses. – Marburger Zeitung 297/298, (22./23. nov. 1941), str. 5. SNOJ, M. 2009, s. v. Trojane. – V / In: M. Snoj, Etimološki slovar slovenskih zemljepisnih imen, 439–440, Ljubljana. STRAŽAR, S. 1985, Atrans – Trojane pomembno arheološko najdišče. – V / In: Črni graben. Od Prevoj do Trojan, 84–91, Lukovica. ŠAŠEL. J. 1954a, Arheološko–topografske novosti s področja Colatia in Atransa. – Arheološki vestnik 5, 154–166. ŠAŠEL, J. 1954b, Add. ad CIL III 11675 Atrans, Noricum. – Živa antika 4, 200–208 (= id., Opera selecta, Situla 30, 1992, 227–223).ŠAŠEL, J. 1959, Prebujeni Atrans. – Ljubljanski dnevnik, letnik 9, št. 200 (29. 8. 1959), str. 3. ŠAŠEL, J. 1975, Rimske ceste v Sloveniji. – V / In: Arheološka najdišča Slovenije, 74–99, Ljubljana.ŠAŠEL, J., L. BOLTA 1975, Trojane. – V / In: Arheološka naj­dišča Slovenije, 267–268, Ljubljana. ŠAŠEL KOS, M. 1999, Cestni postaji Atrans in Pretorij La-tobikov. – V / In: B. Aubelj (ur. / ed.), Zakladi tisočletij. Zgodovina Slovenije od neandertalcev do Slovanov, 238–240, Ljubljana. ŠAŠEL KOS, M. 2004, Lapidarij Narodnega muzeja Slovenije: rimski spomeniki, vodnik. - Ljubljana. TATSCHEVA, M. 1996, Neues über Publicum Portorii Illyrici et Ripae Thraciae. – In: Rimski limes na srednjem i donjem Dunavu / Roman Limes on the Middle and Lower Danube, Đerdapske sveske, pos. izd. 2 / Cahiers des Portes de Fer, Monogr. 2, 177–182, Beograd. TORKAR, S. 2012, Razpoznavanje slovenskih zemljepisnih imen. – Slavistična revija 60/4, 693–707. VALVASOR, J. V. 1689a, Die Ehre des Herzogthums Crain, 2. izdaja, knjiga II., Rudolfswerth 1877, Laybach–Nürnberg. VALVASOR, J. V. 1689b, Die Ehre des Herzogthums Crain, 2. izdaja, knjiga V., Rudolfswerth 1877, Laybach, Nürnberg. VISOČNIK, J. 2014, Foreigners in the area of Celeia. – Classica et Christiana 9/1, 275– 298. VISOČNIK, J. 2017, The Roman inscriptions from Celeia and its ager. – Ljubljana, Celje. ZUPANČIČ M. 1966, Rimska hiša na Trojanah. – Kamniški občan, leto V, št. 4, julij 1966, str. 10. ZUPANČIČ, M. 1970, Letošnja arheološka izkopavanja na Trojanah. – Kamniški občan, leto IX, št. 11/12, 20. 11. 1970, str. 13. ZUPANČIČ, M. 1971, Atrans – Trojane. Mladinski raziskovalni tabori 1970. – Ljubljana, 215–222. ZUPANČIČ, M. 1977, Atrans – nov napis / Atrans – recente scoperta di un‘epigrafe. – Arheološki vestnik 28, 1977, 106–109. ZUPANČIČ, M. 1979, Arheološki najdišči Trojane in Mengeš. – V / In: Zbornik občine Domžale, 15–22, Domžale. ZUPANČIČ, M. et al. 2003 = M. Zupančič, M. Lovenjak, M. Lesar, Z. Torkar 2003, Lapidarij kamniškega muzeja. – Kamnik. Janja Železnikar Medobčinski muzej Kamnik Muzejska pot 3 SI-1241 Kamnik janja.zeleznikar@gmail.com Julijana Visočnik Nadškofijski arhiv Ljubljana Krekov trg 1 SI-1000 Ljubljana julijana332@yahoo.com Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 295–304 ŠEMPETER V SAVINJSKI DOLINI Irena LAZAR Izvleček Antično ime naselbine in njen status nista znana, niti njena točna lokacija. O njenem obstoju sklepamo po obsežni rimski nekropoli. Skromni podatki o ostankih rimske arhitekture izvirajo z območja severno od magistralne ceste. Iz oko-lice Šempetra so rimski ostanki znani iz Zgornjih Grušovelj, Podloga, Dobrteše vasi, Novega Kloštra in Gotovelj, vendar nobeno od najdišč ni bilo raziskano. V Šempetru je bila raziskana nekropola s 96 grobovi in ostanki marmornih grobnic, ki pričajo o visokem družbenem statusu in funkcijah posameznikov. Grobišče, ki je ležalo tik ob cesti Emona–Celeia, je bilo uničeno v poplavi v 3. st., mlajše grobišče je nastalo severno. Hipoteza, da je na območju Šempetra morda ležala postaja Ad Medias, ni preverjena. Ključne besede: Norik, Šempeter v Savinjski dolini, Ad Medias, rimska doba, naselbina, vicus, grobišče, marmorna grobnica, cesta Emona–Celeia, Jupiter Dolihenus Abstract The Roman name of the settlement and its administrative status are not known. Its existence is supposed on the basis of an extensive Roman necropolis. Very modest data about Roman architecture are known from the area north of the current main road. The Roman remains in the area around Šempeter were attested in Zgornje Grušovlje, Podlog, Dobrteša vas, Novi Klošter and Gotovlje, but none of these sites were researched. The Roman necropolis with 96 graves and remains of the marble tombs and plots was excavated in Šempeter and gives evidence about the high social status and administrative functions of individuals. The necropolis was adjacent to the road Emona-Celeia and was destroyed during a 3rd century flood; later, another necropolis was positioned to the north. Hypothesis about the Roman station Ad Medias to be positioned at Šempeter remains to be verified. Keywords: Noricum, Šempeter v Savinjski dolini, Ad Medias, Roman period, settlement, vicus, necropolis, marble tomb, road Emona-Celeia, Iuppiter Dolichenus https://doi.org/10.3986/9789610502586_15 Sl. 1: Šempeter v Savinjski dolini z vrisanimi lokacijami raziskanih območij in posamičnih najdb: 1 – rimska nekropola (vzhodni, in situ ohranjeni del); 2 – rimska cesta; 3 – rimska nekropola (v poplavi uničen del); 4a – poznorimska nekropola; 4b – pozno­rimska nekropola; 5 – najdišče posvetilnega oltarja Jupitru Dolihenskemu; 6 – ostanki zidov, naselbina? Fig. 1: Šempeter v Savinjski dolini with marked locations of the researched areas and individual finds: 1 – Roman necropolis (eastern part, preserved in situ); 2 – Roman road; 3 – Roman necropolis (part destructed in a flood ); 4a – late Roman necropo­lis; 4b – late Roman necropolis; 5 – find spot of the votive altar for Iuppiter Dolichenus; 6 – architectural remains, settlement? ANTIČNO IME IN LEGA Šempeter v Savinjski dolini se je sredi 19. stoletja imenoval Sv. Peter ob Savinji, po vojni nekaj časa tudi Št. Peter. Leži 12 km zahodno od Celja (Municipium Clau­dium Celeia). Mimo je že v prazgodovini tekla trgovska pot, t. i. jantarna cesta, v rimski dobi pa via publica Emona–Celeia (Šašel 1975, 76; Lazar 2006b, 160). Južno je skoraj neposredno ob rimski cesti tekla reka Savinja. Ob poplavi v 3. stoletju si je reka vrezala novo strugo in Savinja še danes teče precej bolj južno, izven naselja (Kolšek 1961, 147). Antično ime naselbine ni znano, niti njena točna lokacija. O obstoju naselbine sklepamo po obsežni rim-ski nekropoli. Edini podatki o ostankih rimske stavbne arhitekture izvirajo z območja severno od magistralne ceste, kjer so v kleti stavbe Šempeter št. 22 (danes Juhar­tova ulica 29) naleteli na temelje, ki so jih opredelili kot del stavbe (sl. 1: 6; Kolšek 1983; 1997, 15). Glede na razdaljo verjetno hipotezo, da bi postajo Ad Medias, po 13 rimskih milj oddaljeno od naselja Atrans in od Celeje, iskali v Šempetru, je zapisala Vera Kolšek (1983, 88). Iz okolice Šempetra so rimski ostanki znani iz Zgornjih Grušovelj, Podloga, Dobrteše vasi, Novega Kloštra in Gotovelj, vendar nobeno od teh najdišč ni bilo sistematično raziskano (Kolšek 1959, 118; 1997, 8–9). THE ANCIENT NAME AND POSITION OF THE SETTLEMENT Šempeter v Savinjski dolini was named Sv. Peter ob Savinji (St Peter at Savinja River) in the 19th century; after WWI, it was referred to as St Peter for a while. Lying some 12 km west of Celje (Municipium Claudium Celeia), it is close to the prehistoric amber route; in Roman period, it lay on the state via publicaEmona–Celeia (Šašel 1975, 76; Lazar 2006b, 160), south of which flew the Savinja River in its near vicinity. During the flood in the 3rd century, the river cut a new riverbed; Savinja still runs much more to the south, outside the settlement (Kolšek 1961, 147). The ancient name of the settlement is not known, nor its exact location. Its existence is surmised from the extensive near-by Roman necropolis. The only remains of Roman architecture were located north of the main road, in the basement of building Šempeter No. 22 (the present day 29 Juhartova ulica St.), where foundations were found that were defined as part of a building (Fig. 1: 6; Kolšek 1983; 1997, 15). The plausible hypothesis that this could have been the Ad Medias station, 13 Roman miles away from both Atrans and Celeia, was first suggested by Vera Kolšek (1983, 88). Around Šempeter, Roman remains have been found in Zgornje Grušovlje, Podlog, Dobrteša vas, Novi Klošter and Gotovlje, none of which were systematically explored (Kolšek 1959, 118; 1997, 8–9). ZGODOVINA RAZISKAV Prve arheološke raziskave v Šempetru segajo v leto 1947, ko so na vzhodnem robu takratnega naselja ob kopanju proda na južni strani ceste naleteli na zidove in keramiko (sl. 1: tik vzhodno od območja 1); ob odkritju so jih interpretirali kot ostanke objekta (parc. št. 742, k. o. Šempeter; Perc 1948). Leta 1952 je bila pri zemeljskih delih slučajno od­krita rimska nekropola. V vrtu družine Wolf so naleteli na kip sedeče ženske, reliefa satira z nimfo in arhitek­turne člene (sl. 1: znotraj območja 3; parc. št. 731, k. o. Šempeter). Slovenska akademija znanosti in umetnosti je še istega leta prevzela izkopavanja, ki so trajala do leta 1956 (sl. 1: 3; Klemenc, Kolšek, Petru 1972).1 Leta 1964 so sledile raziskave Pokrajinskega muzeja Celje na vzhodni strani naselja, prav tako južno od ceste Ljublja­na–Celje, kjer so odkrili vzhodni del rimske nekropole (parc. št. 741, 742, 743, k. o. Šempeter; Kolšek 1977). Odkritih je bilo 96 grobov, ostanki grobnih parcel (sl. 1: 1; 2) in rimska cesta Emona–Celeia (sl. 1: 2). Pokazalo se je, da so omenjeni odkopani ostanki iz leta 1947 (sl. 1: vzhodno od območja 1) del rimske nekropole. Leta 1956 so blizu železniške postaje naleteli na mlajše grobišče, odkrili so štiri grobove iz 4. st. (sl. 1: 4a), v neposredni bližini pa nato leta 1978 še marmorne nagrobnike (sl. 1: 4b) (Kolšek 1983, 83; 1986). PRAZGODOVINA Prazgodovinski grobovi iz halštatskega obdobja so bili odkriti na vzhodnem delu rimske nekropole, ležali so pod rimskimi grobovi (prim. sl. 1: 1; Kolšek 1977, 11, tab. 30, 31). Na južnem robu Savinjske doline, dobra dvakilometra od Šempetra, v Grižah in Šeščah, poznamo prazgodovinske gomile, pripadajoča prazgodovinska naselbina leži morda na Lajnarjevem hribu ali na Homu (Bolta 1959, 108). VIRI ZA RIMSKO DOBO Rezultati raziskav rimske nekropole so objavljeni v monografijah Antične grobnice v Šempetru (Klemenc, Kolšek, Petru 1972) in Vzhodni del antične nekropo­le v Šempetru (Kolšek 1977). Študijo o ikonografiji šempetrskih spomenikov je prispeval J. Kastelic (1998) v knjigi Simbolika mitov na rimskih nagrobnih spomenikih, pomembne dopolnitve pa v zadnjem času K. Šmid (2013; 2014). G. Kremer (2001) je šempetrske grobnice vklju-čila v delo o tipologiji noriških nagrobnih spomenikov, zadnji prispevek k rekonstrukciji šempetrskih grobnic 1 Izkopavanja, ki jih je vodil prof. Josip Klemenc, so po­tekala v sodelovanju s Pokrajinskim muzejem Celje in Arhe­ološkim seminarjem Univerze v Ljubljani. RESEARCH HISTORY First archaeological research was conducted in 1947, when walls and ceramics were uncovered during gravel-digging on the southern side of the road (Fig. 1: east of area 1); they were interpreted as remains of a building (plotNr. 742, cadastral municipality Šempeter; Perc 1948). In 1952 the Roman necropolis was accidentally dis­covered during some earthworks. A sculpture of a sitting woman was discovered in the garden of the Wolf family, two reliefs of a satyr and a nymph and some architectural elements (sl. 1: 3; plot Nr. 731, cadastral municipalityŠempeter). That year, the excavations were taken over by the Slovenian Academy of Sciences and Arts, and lasted until 1956 (Klemenc, Kolšek, Petru 1972).1 In 1964 further research was conducted by the Regional Museum Celje at the eastern side of the settlement south of the Ljubljana–Celje road where the eastern part of the necropolis was uncovered (plot Nr. 741, 742, 743,cadastral municipality Šempeter; Kolšek 1977). 96 graves were found, remains of grave plots and the Roman road Emona–Celeia (Figs. 1: 1,2; 2). It became obvious that the remains excavated in 1947 were part of the Roman necropolis (Fig. 1: east of area 1). In 1956, an earlier graveyard was uncovered close the railway station; four graves from the 4th century were found (Fig. 1: 4a), followed by the discovery of near-by marble tombstones in 1978 (Fig. 1: 4b; Kolšek 1983, 83; 1986). PREHISTORY Prehistoric graves from the Hallstatt period were discovered in the eastern part of the Roman necropolis; they lay under the Roman graves (Fig. 1: 1; Kolšek 1977, 11, Pls. 30, 31). On the southern edge of the Savinja Rivervalley, in Griže and Šešče some two kilometres fromŠempeter, there are prehistoric barrows; the relevant prehistoric settlement lay perhaps on the Lajnarjev hrib or on Hom (Bolta 1959, 108). SOURCES ON THE ROMAN PERIOD Results of research conducted on the Roman necropolis were published in the monographs Antične grobnice v Šempetru (Klemenc, Kolšek, Petru 1972) and Vzhodni del antične nekropole v Šempetru (Kolšek 1977). Kastelic contributed a study on the iconography of the Šempeter monuments in his book Simbolika mitov na rimskih nagrobnih spomenikih (1998); they were recentlyimportantly expanded by Šmid (2013; 2014). Kremer (2001) included the Šempeter tombs in her work on 1 Excavations, directed by Josip Klemenc, were con­ducted in collaboration of the Regional Museum Celje and Archaeological seminar from the Ljubljana University. je delo A. Maver (2004), ki je rekonstruirala t. i. arkad-no grobnico. Zgodovina raziskav s seznamom razprav in literature je objavljena v članku Rimska nekropola v Šempetru – zgodovina raziskav ter za zgodnje obdobje v Arheoloških najdiščih Slovenije (Lazar 2006a; Bolta 1975). Posvetilna napisa in napisi z nagrobnih spomeni­kov so vključeni v najnovejšo objavo napisov celejskega agra (Visočnik 2017a), z vso zbrano starejšo literaturo. Numizmatične najdbe je zbral Peter Kos in so vklju-čene v Die Fundmünzen der römischen Zeit in Slowenien II (FMRSl II 363/1 in FMRSl II 363/2). TOPOGRAFIJA Rimska cesta Emona–Atrans–Celeia (sl. 1: 2), ki je tekla po dolini, je v Šempetru narekovala postavitev in razvoj naselja in tudi nekropole. Grobišče, ki je, sodeč po grobovih z vzhodnega dela nekropole, nastalo sredi 1. st. in trajalo do velike povodnji konec 3. st. (Kolšek 1961, 150), se je razprostiralo samo na severni strani ceste (sl. 1: 1), saj je na jugu takrat tekla reka Savinja. Po uničenju večjega dela grobišča (sl. 1: 3) konec 3. st. se razvije poznorimska nekropola bolj severno (sl. 1: 4a,b; Kolšek 1983), kar kaže, da je tudi naselje po poplaviživelo naprej. Ostanki stavb v samem Šempetru so zelo skromni (že omenjeni zidovi na območju sl. 1: 6) in vprašanje je, če jih ne bi morali iskati bolj na zahodnem delu današnjega naselja. CESTA Pri izkopavanjih vzhodnega dela nekropole so v letih 1964–1965 in 1967 vzdolž celotne odkopane nekro-pole (pribl. 270 m) odkrili ostanke ceste Emona–Celeia (sl. 1: 2). Cestišče je ležalo le 10–15 cm pod površino in je bilo dokaj slabo ohranjeno. Vozni del ceste je meril 6 m, skupaj z obcestnima jarkoma pa je bila cesta široka 9 m (Kolšek 1977, 10). Ohranila se je le 20–30 cm debela plast nasutja iz kompaktnega drobnega in trdega gramo­za, pod njim pa večje oblice. Zahodno od Šempetra je bila trasa rimske ceste Emona-Celeia dokumentirana na Ilovici pri Vranskem, na vzhodu pa v Levcu, ob zahodni nekropoli rimske Celeje (Lazar 2006b). GROBIŠČE Rimsko grobišče v Šempetru je ležalo vzdolž sever-ne strani ceste Emona (Ljubljana)–Atrans (Trojane)–Ce­leia (Celje), na južni strani pa je tekla reka Savinja. Ob izkopavanjih v letih 1952–1956 je bil raziskan predel nekdanje struge Savinje, s prodom in peskom zasuto rečno korito, v katerem so ležali deli podrtih grobnic, typologies of tombs in Noricum. The last contribution on the reconstruction of Šempeter tombs was authored by Maver who reconstructed the so-called arcade tomb (2004). Research history with all relevant sources was published in the article Rimska nekropola v Šempetru – zgodovina raziskav (Lazar 2006a; see also Bolta 1975). Two votive inscriptions and funerary inscriptions were included in the latest publication of inscriptions from the Celeia ager (Visočnik 2017a), including all relevant literature. Numismatic finds were were collected by Peter Kos and included in the Die Fundmünzen der römischen Zeit in Slowenien II (FMRSl II 363/1 and FMRSl II 363/2). TOPOGRAPHY The Emona–Atrans–Celeia road (Fig. 1: 2), which ran along the valley, must have been a decisive factor in the planning of the Šempeter necropolis and quite possibly the settlement as well. The burial ground which, judging by the graves from the eastern part of the necropolis, took shape in the mid-1st century and stayed in use until the great flood in the late 3rd century (Kolšek 1961, 150), occupied only the northern side of the road (Fig. 1: 1), since the Savinja River flew on the southern side. After the damage to the necropolis at the end of the 3rd century (Fig. 1: 3), the late Roman necropolis took shape further to the north (Fig. 1: 4a,b; Kolšek 1983), indicating that the settlement must have lived on. Its remains are very scarce in Šempeter itself (some walls in the area Fig. 1: 6), so maybe they should be looked for further to the west. THE ROAD During the 1964–65 and 1967 excavations remains of the Emona–Celeia road were discovered all along the necropolis (approx. 270 m; Fig. 1: 2). The poorly preserved road laid only about 10–15 cm below the surface. The driving portion of the road was 6 m wide; together with the side ditches the whole road was 9 m wide (Kolšek 1977, 10). Only the 20–30 cm thick layer of compact gravel was preserved and some larger peb­bles underneath it. The remains of the Emona–Celeia road west of Šempeter were documented in Ilovica near Vransko and to the east in Levec, alongside the western necropolis of Roman Celeia (Lazar 2006b). THE NECROPOLIS The Roman necropolis in Šempeter lay along the northern side of the Emona (Ljubljana)–Atrans (Trojane)–Celeia (Celje) road; along south side flew Sl. 2: Šempeter v Savinjski dolini. Vzhodni del nekropole z rimsko cesto (najdišči: sl. 1: 1 in 2). Območje je urejeno kot arheološki park. Pogled proti zahodu. Fig. 2: Šempeter v Savinjski dolini. The eastern part of the Roman necropolis with the Roman road (sites fig. 1: 1 and 2). The area is settled as an archaeological park. View to the west. (Fototeka Pokrajinski muzej Celje / Archive of the Regional Museum Celje) pepelnic, stel idr. (sl. 1: 3).2 Profili so pokazali, da je Savinja na tem delu delala manjši okljuk. Izkopavanja vzhodnega dela nekropole (1964 in dalje) pa so odkrila del grobišča, ki ni bil v celoti uničen ob poplavi (sl. 1: 1; 2). Grobovi in grobnice na ohranjenem vzhodnem delu so bili nanizani tesno drug ob drugem v smeri proti uničenim grobnicam. Nekatere grobne parcele so imele zidan osrednji grob, ostali grobovi znotraj parcele so bili preprosto vkopani v zemljo in prekriti s tlakom iz manjših rečnih prodnikov. Na ostanke tlaka so naleteli na več grobnih parcelah in med posameznimi grobovi; kaže, da je bilo grobišče v celoti tlakovano s prodniki (Kolšek 1977, pril. 1). Ostanki grobnic so se ohranili le v temeljih različnih oblik in velikosti, nekateri so bili plitvi in dokaj slabo grajeni, masivni temelji pa so segali 60 do 100 cm v globino. Ob nekaterih temeljih se je ohranila tudi ograja grobne parcele in na površini med temeljem in ograjo so ležale preprostejše grobne jame nepravilnih oblik. Nekatere so imele s kamenjem obloženo steno groba in so segale do 120 cm globoko. V takih grobovih je bil inventar v celoti ohranjen. Na dveh mestih so bili na vzhodnem delu nekro-pole ohranjeni tudi ostanki grmade (ustrina). To je bila plitva ovalna jama tik groba, dolga približno 150 cm, polna žganine in razbitih kosov keramike (Kolšek 1977). 2 Na 2500 m2 raziskane površine so odkrili 604 kamnite kose oz. odlomke uničenih grobnic, pretežno iz belega mar-morja. the Savinja River. During the 1952–1956 excavations, a portion of the former Savinja riverbed was researched, It had been filled with gravel and stones in which parts of demolished tombs, ash chests, stelae etc. were located (Fig. 1: 3).2 Cross sections indicated that the river made a slight turn here. The excavations of the eastern part of the necropolis (1964 and later) uncovered a portion of the burial ground, which survived the flood (Figs. 1: 1; 2). Graves and tombs were racked closely to­gether in the direction towards the demolished tombs. Some grave plots had a walled central grave, while other graves within the plot were simply dug into the ground and covered with a pavement made of small river cobbles. Since remains of paving were discovered in several burial plots, it looks like the whole burial ground could have been paved with pebbles (Kolšek 1977, App. 1). The remains of the tombs survived only in the foundations of various shapes and sizes, some of which were shallow and poorly built, while the more massive foundations ranged from 60 to 100 cm deep. With some foundations, the walls of grave enclosures also survived, and on the surface between the foun­dation and the wall there were shallower, irregularly shaped grave pits. Some were walled with stones and extended up to 120 cm deep. In these graves, the inven­tory completely survived. 2 On the excavated area measuring 2500 m2 some 604 stone fragments were found, i. e. elements of destroyed tombs, mostly made of white marble. Grobišče je bilo dolgo več kot kilometer, zaradi strnjenih stanovanjskih poslopij na zahodni strani ni v celoti raziskano, prav tako pa ne na vzhodni strani. Ob poplavi konec 3. st. (Kolšek 1983) je bil velik zahodni del grobišča uničen. Kaže, da so v pozni antiki poko­pavali na novi nekropoli, ki je ležala severno od stare. S te poznamo le štiri grobove in ostanke nagrobnikov. Od leta 1964 so štiri rekonstruirane marmorne grobnice postavljene v neposredni bližini odkritja, temelj grobnice Spektacijev je ohranjen in situ (sl. 1: znotraj območja 3). Leta 1978 so dodali prezentacijo ostankov rimske ceste v dolžini 300 m ter rekonstruira­nimi tlorisi grobov in grobnih parcel ob njej (sl. 1: 1,2; 2; Kolšek 1997). Rimska nekropola je vpisana v register kulturne dediščine kot arheološki spomenik državnega pomena od leta 2003. STATUS NASELJA Status naselja na osnovi literarnih in epigrafskih virov ni poznan, glede na bližino municipija Celeia lahko domnevamo, da je šlo za vicus, ki je upravno spadal v mestno območje Celeje (Municipium Claudium Celeia) in bil del rimske province Norik (Noricum), od vlade cesarja Dioklecijana dalje pa je spadal v Sredozemski Norik (Noricum Mediterraneum). DRUŽBA Na epigrafskih spomenikih Šempetra se pojavljajo imena italskih trgovskih družin (Eniji, Spektaciji; sl. 3), večinoma priseljenih iz severne Italije (Visočnik 2007; 2017b). O njihovem družbenem statusu in statusu posa­meznikov lahko sklepamo iz bogato okrašenih grobnih edikul in napisov, na katerih so omenjene pomembne administrativne funkcije v upravi celejskega agra. Na-grobni relief (sl. 4) grobnice Spektacijev z upodobitvijo Tutorije Avite in 12-letnega nečaka (Gaius Rusticius Tutor) govori o njuni vlogi v Izidinem kultu (Glaser 1997, 130; Visočnik 2011, 42). Nagrobni in posvetilni napisi iz Šempetra izpriču­jejo funkcije duovirov, edilov, svečenikov in dekurionov, katerih nosilci so funkcijo opravljali v upravi municipija Celeja. Gaius Vindonius Successus je opravljal funkcijo edila v Celeji (sl. 5; Visočnik 2017a, 409, št. 415); grob­nica je datirana na konec 1. oz. v 2. st. (Kastelic 1998, 222), Visočnikova pa jo postavlja v 2. in na začetek 3. st. (Visočnik 2017a, 410). Gaius Spectatius Priscianus in Gaius Spectatitus Finitus sta v Celeji opravljala funkcijo duovirov oz. županov; grobnica je datirana v sredino 2. oz. na začetek 3. st. (Visočnik 2017a, 406, št. 413). Enako funkcijo je opravljal tudi Bellicius Victor, ki je bil tudi član mestnega sveta – decurio v Celeji; njegova nagrobna In the eastern part of the necropolis, parts of burn­ing stakes (ustrina) were discovered in two places – a shallow, 1.5 m long and oval shaped pit right next to the grave, which was filled with burnt debris and pieces of ceramics (Kolšek 1977). The burial ground was over a kilometre long; how­ever, due to the densely packed living quarters it wasn’t fully explored nor on the western nor the eastern side. During the 3rd century flood (Kolšek 1983), a large por­tion of the burial ground was ruined. It looks like the new necropolis was used during the late Roman period, which lay to the north. Only a few graves and tombstones are known from the latter. Since 1964, four reconstructed marble tombs were erected in the close vicinity of where they had been discovered. The ground plan of the Spectatii tomb is pre­served in situ (Fig. 1: 3). In 1978, a 300-m presentation of the Roman road was added with reconstructions of graves and grave plots foundations (Figs. 1: 1,2; 2; Kolšek 1997). The Roman necropolis was registered as an archaeological monument of national importance in 2003. SETTLEMENT STATUS The status of the settlement is not discernible in contemporary literary and epigraphic sources. Since the municipium Celeia was not very far away, it is safe to as­sume that this could have been a vicus under the admin­istration of Celeia (Municipium Claudium Celeia) and was thus a part of the province of Noricum and later the Noricum Mediterraneum during the reign of Diocletian. SOCIETY On the epigraphic monuments of Šempeter names of Italic merchant families appear (the Ennii, the Spectatii; Fig. 3) that mostly came from northern Italy (Visočnik 2007; 2017b). Their social status can be surmised from the richly adorned grave aediculae and inscriptions, in which important administrative functions in the administration of the Celeia ager are numbered. The grave relief (Fig. 4) of the Spectatii tomb with the depiction of Tutoria Avita and her 12-year old nephew (Gaius Rusticius Tutor) speaks of their role in the Isis cult (Glaser 1997, 130; Visočnik 2011, 42). The tomb inscriptions from Šempeter speak of the functions of duumviri, aediles, priests and decuriones, performed by individuals in the administration of the Celeia municipium. Gaius Vindonius Successus was an aedile in Celeia (Fig. 5; Visočnik 2017a, 409, no. 415); the tomb dates from the end of the 1st or the early 2nd century (Kastelic 1998, 222); Visočnik places it in the 2nd or early 3rd century (2017a, 410). Gaius Spectatius Priscianus and Gaius Spectatitus Finitus were duumviri Sl. 3: Šempeter v Savinjski dolini, grobišče (najdišče: sl. 1: 3). Rekonstruirana družinska grobnica Spektacijev. Gaj Spektacij Priscijan in Gaj Spektacij Finit sta opravljala v Celeji funkcijo duovirov. Fig. 3: Šempeter v Savinjski dolini, necropolis (site fig. 1: 3). Reconstructed marble tomb of the Spectatii family. Gaius Spectatius Priscianus and Gaius Spectatius Finitus acted as duoviri in nearby Celeia. (Foto / Photo: Tomaž Lauko; fototeka / archive PM Celje) stela na poznorimski nekropoli (sl. 1: 4b) je datirana na konec 3. oziroma v 4. st. (Visočnik 2017a, 405, št. 412). Na posvetilnem oltarju (sl. 1: 5; 6) za Jupitra Doli­henskega, postavljenem tudi v čast cesarja Makrina in sina Diadumeniana, nastopa trojica svečenikov Jupitra Dolihenskega – Aurelius Aquila, Aurelius Bassus in Varinus, ki so poskrbeli za obnovo svetišča, uničenega v požaru (Visočnik 2017a, 404; Kolšek 1986, 371–374). Napis je datiran med letoma 217 in junijem 218. Sl. 4: Šempeter v Savinjski dolini, grobišče (najdišče: sl. 1: 3). Rekonstruirana družinska grobnica Gaja Spektacija Sekun­dijana. Relief Tutorije Avite in 12-letnega nečaka. Oblačilo kaže, da je bila Tutorija Avita verjetno Izidina svečenica, nečak pa ima na glavi značilen čop, ki označuje t. i. Horove dečke, posvečene v začetno stopnjo Izidinega kulta. Fig. 4: Šempeter v Savinjski dolini, necropolis (site: Fig. 1: 3). Reconstructed family tomb of Gaius Spectatius Secundianus. Relief of Tutoria Avita and her 12-years old nephew. Her dress indicates that she was probably active as an Isis priestess while her nephew wears typical pony tail on his head that represents so-called Horusknaben, boys who were initiated in the first rank of the Isis cult. (Foto / Photo: Viktor Berk; fototeka / archive PM Celje) or mayors of Celeia; their tomb dates from the mid-2nd or early 3rd century (Visočnik 2017a, 406, no. 413). The same function was held by Bellicius Victor who was also a decurio; his grave stele in the late Roman necropolis (Fig. 1: 4b) dates from the the end of the 3rd or the 4th century (Visočnik 2017a, 405, no. 412). The votive altar for Jupiter Dolichenus (Fig. 1: 5; 6), built also in honour of Emperor Macrinus and his son Diadumenianus, features three Jupiter’s priests – Aure­lius Aquila, Aurelius Bassus and Varinus, responsible for the restoration of the temple ruined in a fire (Visočnik 2017a, 404; Kolšek 1986, 371–374). The inscription dates from the period between 217 and June 218. ETNIČNA PRIPADNOST Rimljani svojo etnično pripadnost na eni strani izpričujejo z imeni in imenskimi formulami polno­pravnih rimskih državljanov in tudi reliefnimi upo­dobitvami na nagrobnih spomenikih (Kastelic 1998; Visočnik 2007). Avtohtono prebivalstvo in keltsko tradicijo pre­poznamo posebej v ženskih upodobitvah na reliefih (npr. z grobnice družine Enijev; sl. 7), keltskih imenih (Visočnik 2017b) in tudi v grobnih pridatkih, kjer se lo-kalna lončarska tradicija keltskega oz. prazgodovinskega obdobja prepleta z rimskimi importi (primer npr. grob 27 z vzhodnega dela nekropole; sl. 8). Omeniti velja tudi lokalno, drugje neznano keltsko božanstvo Carvonia, verjetno povezano z naravo, katere čaščenje dokumentira posvetilni oltar (CIL III 5115 (+ p. 1827); Visočnik 2017a, 402, št. 410)3, najden še pred letom 1832 v Dobrteši vasi (Šempeter v Savinjski dolini).4 3 Nagrobnik je bil vzidan v poštno poslopje. 4 Zaradi rasti in širitve sodobnega Šempetra v Savinjski dolini je od leta 1999 Dobrteša vas del naselja (Uradni list RS 88/1999). Sl. 6: Šempeter v Savinjski dolini (najdišče: sl. 1: 5). Posvetilni oltar za Jupitra Dolihenskega je dala postaviti trojica sveče­nikov tega božanstva – Aurelius Aquila, Aurelius Bassus in Varinus leta 217 oz. 218. Fig. 6: Šempeter v Savinjski dolini (site: Fig. 1: 5). Votive altar for Iuppiter Dolichenus was erected by three priests of his cult – Aurelius Aquila, Aurelius Bassus and Varinus in the years 217 or 218. (Foto / Photo: Ortolf Harl, ubi-erat-lupa) ETHNICITY Romans professed their ethnicity with names and name formulae of fully-fledged Roman citizens, as well as the relief depictions on tombstones (Kastelic 1998; Visočnik 2007). Autochthonous residents and their Celtic traditions are recognizable in female depictions and reliefs (Fig. 7), Celtic names (the tombstone of the Ennii family; Visočnik 2017b) and grave artefacts, where local Celtic or prehis­toric pottery traditions mix with Roman imports (i.e. the grave no 27 from the eastern part of the necropolis; Fig. 8). Another important aspect is the local Celtic deity of Carvonia not worshiped elsewhere, which might have been related to nature. Its worshipping is documented in a votive altar (CIL III 5115; Visočnik 2017a, 402, no. 410),3 found prior to the year 1832 in Dobrteša vas (Šempeter v Savinjski dolini).4 Translation: Gregor Pobežin, Irena Lazar 3 The altar was bulit-in in a post-office building. 4 Due to the growth and development of modern Šempe­ter, the village Dobrteša vas was included in the settlement in 1999 (Uradni list RS 88/1999). Sl. 7: Šempeter v Savinjski dolini, grobišče (najdišče: sl. 1: 3). Portreti z rekonstruirane družinske grobnice Enijev. Ennia Oppidana je upodobljena v značilni keltski noši. Fig. 7: Šempeter v Savinjski dolini, necropolis (site: Fig. 1: 3). Portraits from the reconstructed family tomb of the Ennii fam­ily. Ennia Oppidana is represented in a typical Celtic costume. (Foto / Photo: Viktor Berk; fototeka / archive PM Celje) Sl. 8: Šempeter v Savinjski dolini. Grob 27 z vzhodnega dela nekropole (najdišče: sl. 1: 1) s pridatki odraža keltsko lončarsko tradicijo, ki se prepleta s fino namizno keramiko, uvoženo iz severnoitalskih delavnic. 1. do 2. stoletje. Fig. 8: Šempeter v Savinjski dolini. Grave 27 from the eastern part of the necropolis (site: Fig. 1: 1) reflects with its grave goods the local Celtic pottery tradition which is interwoven with the Roman fine tableware, imported from the northern Italian work­shops. 1st–2nd century. (Foto / Photo: Tomaž Lauko; fototeka / archive PM Celje) Okrajšave CIL = Corpus inscriptionum Latinarum. FMRSl II = P. Kos, Die Fundmünzen der römischen Zeit in Slowenien II. – Berlin 1988. BOLTA, L. 1959, Najstarejša zgodovina Spodnje Savinjske doline. – Savinjski zbornik 1, 103–113. BOLTA, L. 1975, Šempeter v Savinjski dolini. – V / In: Arhe­ološka najdišča Slovenije, 290–291, Ljubljana. GLASER, F. 1997, Isisverehrerinen in Noricum. – V / In: B. Djurić, I. Lazar (ur. / eds.), Akten des IV. internationalen Kolloquiums über Probleme des provinzialrömischen Kunst­schaffens, Situla 36, 127–130. KASTELIC, J. 1998, Simbolika mitov na rimskih nagrobnihspomenikih. Šempeter v Savinjski dolini. – Ljubljana. KLEMENC, J., V. KOLŠEK, P. PETRU 1972, Antične grobnice vŠempetru (Antike Grabmonumente in Šempeter). – Katalogi in monografije 9. KOLŠEK, V. 1959, Savinjska dolina v rimski dobi. – Savinjski zbornik 1, 118–131. KOLŠEK, V. 1961, Dva miljnika iz Celja. – Arheološki vestnik 11–12, 147–151. KOLŠEK, V. 1977, Vzhodni del antične nekropole v Šempetru(Die östliche Antike Nekropole in Šempeter im Savinjatal). – Katalogi in monografije 14.KOLŠEK. V. 1983, Nova arheološka odkritja v Šempetru v Savinjski dolini. – Savinjski zbornik 5, 83–88. KOLŠEK, V. 1986, Epigrafski in reliefni paberki na Celjskem. – Arheološki vestnik 37, 371–384. KOLŠEK, V. 1997, Rimska nekropola v Šempetru. Vodnik /Römische Nekropole im Šempeter. Führer. – Celje. KREMER, G. 2001, Antike Grabbauten in Noricum. – Öster­reichisches Archäologisches Institut, Sonderschriften 36. LAZAR, I. 2006a, The Roman Necropolis in Šempeter: The History of Research. – V / In: M. Kokole, B. Murovec, M. Šašel Kos, M. Talbot (ur. / eds.), Mediterranean myths from classical antiquity to the eighteenth century / Mediteranski miti od antike do 18. stoletja, 57–72, Ljubljana. LAZAR, I. 2006b, Rimska cesta Atrans-Celeia. – V / In: Avtocestna srečanja sodobnosti s preteklostjo, Rast. Revija za literaturo, kulturo in družbena vprašanja, letnik 17, št. 2 (104), 160–164. MAVER, A. 2004, The Arcade Tomb in Šempeter, Slovenia – anattempt at a reconstruction / Arkadna grobnica v Šempetru v Savinjski dolini. – Arheološki vestnik 55, 343–414. PERC, M. 1948, Odkritje temeljev stare stavbe v Št. Petru v Savinjski dolini. – Celjski tednik I/4, 6. ŠAŠEL, J. 1975, Rimske Ceste v Sloveniji. – V / In: Arheološka najdišča Slovenije, 74–99, Ljubljana. ŠMID, K. 2013, Zur Kampfszenerelief am Grabmonument derSpectatii in Šempeter (Prizor boja na grobnici Spektatijev vŠempetru). – Zbornik za umetnostno zgodovino 49, 13–26. ŠMID, K. 2014, Fragment pripovednega reliefa na grobnici Spektatijev v Šempetru. Epizoda zgodbe o Orestu in Ifigeniji na Tavridi ali Toantov dokončni poraz na otoku Sminta? – Acta historiae artis Slovenica 19/1, 7–16. VISOČNIK, J. 2007, Jezikovne značilnosti napisov antične Ce-leje z okolico kot vir za preučevanje romanizacije celejskega prostora. – Disertacija / PhD thesis, Oddelek za klasično filologijo, Filozofska fakulteta Univerze v Ljubljani (neo­bjavljeno / unpublished). VISOČNIK, J. 2011, “Crudeles parentes” on the funerary stele for Aurelius Secundinus and his family (CIL III 5246). – V / In: I. Lazar (ur. / ed.), Religion in public and private sphere. Acta of the 4th International Colloquium The Autonomous Towns of Noricum and Pannonia, Annales Mediterranei, 41–48, Koper. VISOČNIK, J. 2017a, The Roman inscriptions from Celeia and its ager. – Celje, Ljubljana. VISOČNIK, J. 2017b, Jezikovne značilnosti napisov Celeje in njene okolice (The Language Characteristic of Inscriptions Found at and around Celeia). – Keria 9/2, 7–23. Irena Lazar Univerza na Primorskem Fakulteta za humanistične študije Titov trg 5 SI-6000 Koper irena.lazar@fhs.upr.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 305–324 COLATIO – STARI TRG PRI SLOVENJ GRADCU Saša DJURA JELENKO Izvleček Kolaciona (Stari trg pri Slovenj Gradcu) leži na jugozahodnem obrobju Slovenjgraške kotline ob nekdanji rimski cesti. Odkril in delno raziskal jo je Hans Winkler med letoma 1909 in 1912. Kolaciona je nižinska naselbina, poseljena vse od 1. do začetka 5. st. (ostanki svetišča, stanovanjski del in grobišče). Sredi obzidanih grobnih parcel so bile postavljene razkošne grobnice, ki so pripadale podeželski eliti. Najbolje je raziskana grobnica I, v njej so v tretji četrtini 1. st. pokopali moškega in dve ženski. Iz maloštevilnih ohranjenih napisov izvemo, da je bilo prebivalstvo staroselsko, keltizirano, ni pa zanemarljiv niti delež latinskih imen. Ključne besede: Norik, Stari trg pri Slovenj Gradcu, Kolaciona, rimska doba, naselbina, poštna postaja, svetišče, posvetilni oltarji, grobišče, mrtvaška postelja, sella curulis Abstract Colatio (Stari trg pri Slovenj Gradcu) lies on the northwestern outskirts of the Slovenj Gradec basin along the Roman road. It was discovered and partly explored by Hans Winkler between 1909 and 1912. It is a lowland settlement that was inhabited as early as the 1st century and remained populated until the early 5th century (temple remains, residential area, and cemetery). Luxurious tombs for the rural elite were erected in the centre of walled funerary plots. Tomb I revealed the burial remains of a man and two women from the third quarter of the 1st century. The few survived inscriptions indicate the indigenous Celtic population; however, the Latin names are also present. Keywords: Noricum, Stari trg pri Slovenj Gradcu, Colatio, Roman period, settlement, postal station, temple, altars, cemetery, deathbed, sella curulis https://doi.org/10.3986/9789610502586_16 Sl. 1: Najpomembnejša arheološka najdišča v Mislinjski dolini. Fig. 1: The most important archaeological sites in the Mislinja valley. 1a –Legen (gomilno grobišče / tumulus cemetery); 1b – Gradišče (naselbina / settlement); 2 – grajski hrib; 3 – Vrhe; 4 – Mislinjska dobrava; 5 – Colatio; 6 – Zgornje Dovže; 7 – Spodnje Dovže; 8 – Puščava, 9 - sv. Jurij. (Kartografski vir / Map source: http://gis.arso.gov.si/evode/profile.aspx?id=atlas_voda_Lidar@Arso) GEOGRAFSKA LEGA Mislinjska dolina je ujeta med Pohorje na vzhodni in Karavanke na zahodni strani. Južni del te doline predstavlja Slovenjgraška kotlina, ki je dolga 12 in široka 4 km (sl. 1). Sodi med najhladnejša območja v celotni Sloveniji z značilno temperaturno inverzijo. Širše območje izkopanin v Starem trgu leži na izlivnem območju Suhadolnice v Mislinjo. GEOGRAPHICAL LOCATION The Mislinja valley is positioned between the Pohorje Massif on the eastern side and the Karavanke Mountains on the western side. The southern part of the valley consists of the Slovenj Gradec basin, which is 12 km long and 4 km wide (Fig. 1). It is one of the coldest areas in Slovenia, with a typical temperature inversion. The archaeological area of Stari trg is situated at the conflu­ence of the Suhadolnica tributary with the Mislinja River. Sl. 2: Stari trg pri Slovenj Gradcu z grajskim hribom in cerkvijo sv. Pankracija. Pogled proti zahodu. Fig. 2: Stari trg pri Slovenj Gradcu. Aerial photo of Stari trg with Castle Hill and the church of St Pancras. View to the west. (Foto / Photo: P. Juvan) Mislinjska dolina je v rimskem času spadala v provinco Norik. Za časa cesarja Klavdija (41–54) je nastalo več avtonomnih mest, med njimi tudi Virunum (na Gosposvetskem polju) in Celeia (Celje). Med tema municipijema so v sredini 1. st. zgradili rimsko cesto,1 ki je v kraje med Pohorjem in Karavankami pripeljala rimsko vojsko, pa tudi trgovce in obrtnike. Iz starosel­ske keltizirane naselbine,2 se je pod grajskim hribom v Starem trgu razvila poštna postaja Colatio / Kolaciona (sl. 2). V drugi polovici 1. st. je urbanizacija južnega Norika doživela pomemben razcvet, razvijalo pa se je tudi življenje na podeželju. ZGODOVINA RAZISKAV (Sl. 3) Zasluga za lokalizacijo Kolacione gre slovenj­graškemu notarju Hansu Winklerju, ki je v letih od 1909 do 1912 pod vznožjem grajskega hriba delno raziskal svetišče, na starotrških njivah ostanke stano­vanjskih hiš in nekoliko južneje od današnje vasi še pripadajoče grobišče (sl. 3).3 Od leta 1911 je skupaj 1 Miljnik lupa 8047. 2 Trenutno znane le sporadične najdbe. Številna keltska imena domačega prebivalstva kažejo, da je bila Kolaciona poseljena že od prazgodovine dalje (Djura Jelenko, Visočnik 2006). Egger 1914, 59 ss [IV. Colatio (Altenmarkt bei Win- The Mislinja valley was a part of the Roman province of Noricum. During the time of Claudius (AD 41-54), a number of autonomous towns were founded, including Virunum and Celeia. In the mid-1st century, a Roman road was built between these two municipalities, 1 which brought the Roman army as well as traders and craftsmen into the area between Pohorje and Karavanke. The postal station Colatio developed beneath Castle Hill in Stari trg from an autochthonous Celtic settlement (Fig. 2).2 The urbanization of southern Noricum accelerated in the second half of the 1st century, while life in rural areas continued to develop as well. RESEARCH HISTORY (Fig. 3) The locating of Colatio can be credited to Hans Winkler, a notary from Slovenj Gradec who partly researched the temple at the foot of Castle Hill, the re­mains of Roman buildings at Stari trg fields and a Roman cemetery between 1909 and 1912 (Fig. 3).3 From 1911 1 Boundary stone lupa 8047. 2 To date, only stray pre-Roman finds are known from the area. The numerous Celtic names of autochthonous pop­ulation indicate the settlement continuity from prehistory on (Djura Jelenko, Visočnik 2006). 3 Egger 1914 [IV. Colatio (Altenmarkt bei Windisch-Graz)]; Pahič 1968, 51; Pahič 1975, 277; Djura Jelenko 2004. Sl. 3: Stari trg pri Slovenj Gradcu. Območje rimske naselbine in grobišč z etapami raziskav. Fig. 3: Stari trg pri Slovenj Gradcu. Roman settlement and cemetery area with the years of research. (Prirejeno in dopolnjeno po / Adapted from: Djura Jelenko 2012,19. Izvedba / Elaborated by: E. Koraca) z Walterjem Schmidom raziskal še poznoantično in zgodnjesrednjeveško grobišče na Puščavi (sl. 1: 8), ki je del grebena grajskega hriba, za cerkvijo sv. Pankracija.4 Konec 70-let so se raziskave nadaljevale v zahodnem delu južnega grobišča.5 Odkrita je bila večja obzidana grobnica z žganimi grobovi okoli nje, v plasteh nad grobovi pa tudi mlajši predmeti iz časa od 3. pa vse do začetka 5. st (sl. 3). V 90-ih letih 20. stoletja se je težišče raziskav premaknilo na starotrške njive oz. naselbinski disch-Graz)]; Pahič 1968, 51; Pahič 1975; Djura Jelenko 2004.4 Pleterski, Belak 2002. 5 Raziskave: Zavod za varstvo kulturne dediščine Slove- on, Hans Winkler and Walter Schmid excavated the late Antique and early Medieval cemetery at Puščava (Fig. 1: 8), located on the ridge behind the church of St Pancras (Castle Hill).4 In the late 1970s, excavations continued in the western part of the Roman cemetery.5 A large walled tomb with surrounding cremation graves was discovered. The layers covering the graves yielded finds from the 3rd to the early 5th century (Figs. 3; 11). The settlement area in Stari trg fields continued to be researched in the last decade of the 20th century (Figs. 3; 4).6 During the years 4 Pleterski, Belak 2002. 5 ZVKDS, OE Maribor; excavations in 1977, 1978 and nije, OE Maribor; 1977, 1978 in 1982 (Strmčnik Gulič 1981; 1982 (Strmčnik Gulič 1981; 1984; 1995). 1984; 1995). 6 Koroški pokrajinski muzej, excavations in 1993 (Bala­ Sl. 4: Stari trg pri Slovenj Gradcu. Osrednji del rimske naselbine. Lokacije objektov z etapami raziskav. Fig. 4: Stari trg pri Slovenj Gradcu. Building foundations in the central Roman settlement area with marked years of research. (Kartografski vir / Map source: DOF 5 ©Geodetska uprava RS. Izvedba / Elaborated by: E. Koraca) del (sl. 3; 4).6 Območje zgodnjerimskega grobišča je bilo raziskano pred večjimi gradbenimi posegi v letih 2004–2005 in 2006–2007 (sl. 3).7 Odkrita sta bila dva skeletna poznorimska grobova (2004; sl. 11: VI,VII) in bogata naselbinska plast (2006) iz 3. in 4. st. (sl. 3). V neposredni bližini svetišča, na dvorišču bivše Karnerjeve gostilne, sta bili odkriti vsaj dve fazi poselit­ve (2010–2011; sl. 3; 4).8 Zadnje raziskave so pokazale 6 Raziskave: Koroški pokrajinski muzej (v nadaljevanju KPM); 1993/Balaban, 1996/Balaban-Pur in 1999/Areh-Den­kov (Djura Jelenko 2001; Britovšek 2007). 7 Raziskave: KPM; 2004–2005/Krožišče in 2006–2007/ Korobenz (Djura Jelenko, Kumprej Gorjanc 2009).8 Raziskave: KPM; 2009–2011/sprehajalna pot in 2010– 2004/2005 and 2006/2007, the area of the early Roman cemetery was subject to significant construction interven­tions preceded by rescue excavations (Fig. 3).7 Two late Roman inhumation graves were discovered (2004; Fig. 11: VI,VII) as well as a settlement layer dating from the 3rd to the 4th century (2006; Fig. 3). In the courtyard of the former Karner Inn, which lies in the immediate vicinity of the temple, at least two ban), 1996 (Balaban-Pur) and 1999 (Areh-Denkov). Djura Jelenko 2001; Britovšek 2007. 7 Koroški pokrajinski muzej, excavations in 2004/2005 (roundabout) and 2006/2007 (Korobenz). Djura Jelenko, Kumprej Gorjanc 2009. skromne sledove poselitve še na območju med naselbino in grobiščem (2015; sl. 3).9 PRAZGODOVINA Najstarejše najdbe izvirajo s Tisnika nad Gornjim Doličem.10 Na območju Mislinjske doline je bilo odkritih več kamnitih sekir, vse so slučajne najdbe.11 Lončene najdbe iz pozne in mlajše bronaste (Bd D1–2 in Ha A1) ter starejše železne dobe (Ha B) je Winkler odkril med sondiranji na grajskem hribu (sl. 1: 2; 2).12 Na Vrheh so leta 1909 domnevno odkrili zakladno najdbo iz Ha B/C obdobja (sl. 1: 3).13 Na Gradišču, vzhodno od Slovenj Gradca, na samem začetku legenske terase, ki se vzpenja proti Pohorju, je še neraziskana starejšeželeznodobna naselbina (sl. 1: 1b). Bolje poznano je pripadajoče go-milno grobišče v Florijanovem gozdu na Legnu (sl. 1: 1a) iz 8. in 7. st. pr. Kr.14 Novejše raziskave so razkrile poselitev v času med 1300 in 700 pr. Kr. še v Mislinjski dobravi v Starem trgu (sl. 1: 4).15 Iz latenskega obdobja izvirajo le sporadične najdbe.16 RIMSKA DOBA NASELBINA (Sl. 3–11; 18-20) Trenutno znan obseg nižinske naselbine v Kola-cioni znaša približno 650 × 250 m. Na skrajnem seve­rozahodnem delu leži forum z ostanki svetišča, jugo­vzhodneje naselbina (sl. 3; 4) in približno 500 m proti jugu še zgodnjerimsko grobišče, na območju katerega nastane v pozni rimski dobi domnevno obrtniški del naselbine (sl. 11). Večjo uganko predstavlja cesta, ki je vodila v na­selbino in iz nje. Položaj Winklerjevih grobnic (prim. sl. 3: 1911–1912; 11) kaže, da bi jo smeli pričakovati v smeri vzhod–zahod, tudi lokacija nagrobnega spome­nika z liktorjem na južni strani grobnice I (sl. 12–13) to potrjuje. Winkler omenja cesto, ki bi jo naj odkril 2011/Falež) (Djura Jelenko, Kajzer Cafnik 2011).9 Raziskave: Vesna Merc, 2015. 10 Djura Jelenko 2010, 65. 11 Sekire so del Sokličeve zbirke v KPM: Djura Jelenko et al. 2016, 63. 12 Djura Jelenko 2004, 70, t. 43–51. 13 Teržan 1990, 371; Djura Jelenko 2004, 76, sl. 46. 14 Strmčnik Gulič 1979; Djura Jelenko et al. 2016. 15 Zračni posnetki in topografski pregled pri graščini Hartenstein (KPM 1999–2002). Na skrajnem zahodnem delu Mislinjske dobrave je arheološke raziskave leta 2015 izvedel Samo Hvalec. 16 Pahič, 1975; Pleterski, Belak 2002; Djura Jelenko 2001, 182; Djura Jelenko 2016, 99. settlement phases were discovered (2010–2011; Fig. 3–4).8 The latest research has shown modest traces of occupa­tion in the area between the settlement and the cemetery (2015; Fig. 3).9 PREHISTORY The earliest finds originate from Tisnik.10 A number of stone axes were discovered incidentally in the Mislinja valley, which now form part of the Soklič collection.11 The Late Bronze Age, as well as the Early Iron Age pottery, was discovered by Winkler during the test trenching at Castle Hill (Fig. 1: 2; 2).12 A hoard from the Ha B/C period was allegedly uncovered in Vrhe in 1909 (Fig. 1: 3).13 A large and unexplored Early Iron Age settlement is located at Gradišče, east of Slovenj Gradec, at the very beginning of the Legen terrace which rises toward Pohorje (Fig. 1: 1b). A corresponding tumulus cemetery from the 8th and 7th century BC in Florian’s forest at Legen has been more thoroughly researched (Fig. 1: 1a).14 Recent studies have revealed settlement traces from the period between 1300 and 700 BC at Mislinjska dobrava in Stari trg (Fig. 1: 4).15 Only stray finds can be dated into the La Tene period.16 ROMAN PERIOD SETTLEMENT (Figs. 3–11; 18-20) The currently known extent of the Colatio lowland settlement is approx. 650 × 250 m. A forum with a temple is located in the northwestern part (Figs. 3; 4). The early Roman cemetery is situated ca. 500 m south of the settle­ment. Its area was occupied by artisanal buildings in the late Roman period (Fig. 11). The location of the road leading to and from the settlement is questionable. The placement of Winkler’s tombs indicates the east-west direction of the road (Fig. 3: 1911–1912; 11), which is corroborated by the position of 8 Koroški pokrajinski muzej, excavations in 2009-2011 (footpath) and 2010-2011 (Falež). Djura Jelenko, Kajzer Caf­nik 2013. 9 Archeological research was conducted by Vesna Merc in 2015. 10 Djura Jelenko 2010, 65. 11 Djura Jelenko et al. 2016, 63. 12 Djura Jelenko 2004, 70, pl. 43–51. 13 Teržan 1990, 371; Djura Jelenko 2004, 76, fig. 46. 14 Strmčnik Gulič 1979; Djura Jelenko et al. 2016. 15 Aerial view and a topographic survey at the Harten-stein Manor (KPM 1999–2002). The archaeological research was carried out by Samo Hvalec (2015) in the far-western part of Mislinjska dobrava. 16 Pahič, 1975, 277; Pleterski, Belak 2002; Djura Jelenko 2001, 182; Djura Jelenko et al. 2016, 99. Sl. 5: Stari trg pri Slovenj Gradcu. Forum. Prirejeno in dopolnjeno po originalnem Winklerjevem načrtu (prim. Djura Jelenko 2004, sl. 6). (Izvedba / Elaborated by: E. Koraca) Fig. 5: Stari trg pri Slovenj Gradcu. Forum. According to Winkler’s original ground plan (cf. Djura Jelenko 2004, Fig. 6). oktobra 191317 tik pred ovinkom za Šancel (na območju današnjega krožišča; sl. 3); Strmčnikova za raziskave leta 1977 navaja nabit gramoz, ki bi naj ustrezal rimskemu cestišču.18 Druga cesta je omenjena še na Winklerjevem načrtu območja foruma (sl. 5). Djura Jelenko 2004, sl. 34. Na drugi strani te ceste je KPM leta 2009 raziskoval ob gradnji sprehajalne poti. O kakšnem cestnem nasutju ni bilo niti sledu. Lahko, da so jo poškodovala večkratna dela na sedanji cesti, ki deloma pre­kriva Winklerjevo traso. Podobno tlakovano cesto z manj­šimi prodniki, kot jo navaja Winkler, smo odkrili leta 2012 pri gostilni Karner. Gre za novodobno cestišče, tlakovano iz oblic. Ležalo je tik pod asfaltno prevleko. 18 Arheološke raziskave krožišča v letih 2004–2005 (sl. 3) nadaljevanja te trase v smeri proti naselbini niso potrdile. the tombstone with a lictor on the southern side of Tomb I (Figs. 12–13). Winkler reportedly discovered a road in October 1913,17 just before the turn for Šancel (modern roundabout; Fig. 3). In 1977 Mira Strmčnik mentioned a compact gravel layer that might derive from the Roman 17 Djura Jelenko 2004, fig. 34. During a footpath con­struction in 2009, Koroški pokrajinski muzej researched the other side of the Winkler’s road but no traces of a roadbed were found. It may have been damaged by multiple con­struction works on the current road, which partly overlaps Winkler’s route. A similarly paved road with small pebbles, as Winkler notes, was discovered at the Karner Inn in 2012. It was a modern road, which lay just below the asphalt cover. Sl. 6: Stari trg pri Slovenj Gradcu. Odlomki kamnitih spomenikov z območja foruma. A, B, F – posvetilni oltarji; C – korintski kapitel; G – votivna plošča; E, D – votivni figuri (lokacije na sl. 5). Fig. 6: Stari trg pri Slovenj Gradcu. Forum. Altars (A, B, F), Corinthian capital (C), votive slab (G), and votive figures (E, D); (locations on Fig. 5). (Foto / Photo: T. Jeseničnik) Rimsko svetišče je bilo krito z opečnato kritino (sl. 5).19 Južna stranica obodnega zidu je bila dolga do-brih 16 m. Na forumu, severozahodno od cele, so našli arhitekturne elemente stebrišča, temelje nekaterih javnih stavb ter številne posvetilne oltarje (sl. 6).20 Do svetišča na podiju je vodil dvignjen dovoz. Vhodno stopnišče je ležalo v smeri ceste na jugozahodni strani svetišča. Sem so sodile monumentalne plastike.21 Območje osrednjega svetega prostora (cella) je Winkler pustil neraziskano, med celo in obodnim zidom ni od­kril najdb in tudi pozicija cele znotraj svetišča zaenkrat ni povsem pojasnjena. Odlomek korintskega kapitela, odkrit med raziskavami svetišča, in steber, ki stoji danes v cerkvi sv. Pankracija, kažeta, da so za stebrišče upora­ 19 Severovzhodni del načrta. road.18 The other road is marked on Winkler’s plan of the forum area (Fig. 5). The sanctuary was covered with brick roofing (Fig. 5).19 The southern side of the podium wall was over 16 m long. Architectural elements of a colonnade, build­ing foundations and a number of altars were discovered on the forum northwest of the cella (Fig. 6).20 A short, elevated road and the entrance stairway lay in the south­western side of the temple. According to the discovery of bronze fragments, it is presumed that the monumental bronze statues were erected here.21 Winkler left the cella unresearched, and no finds were discovered between the 18 In 2004-2005, archaeological research of the round­about (Fig. 3) did not confirm a continuation of this route toward the settlement. 20 Djura Jelenko 2004, 11. 19 The northeastern part of the ground plan. 21 Bronasti odlomki večjih figur; Djura Jelenko 2004, 17, 20 Djura Jelenko 2004, 11. t. 4: 35. 21 Djura Jelenko 2004, 17, pl. 4: 35. Sl. 7: Stari trg pri Slovenj Gradcu. Rimska naselbina. Raziskave leta 1996. Stranici pomožnega objekta, zgrajeni iz manjših prodnikov (2. st.). Pogled proti jugu. Fig. 7: Stari trg pri Slovenj Gradcu. Excavations in 1996. Building foundations built from pebbles. 2nd century. View to the south. (Foto / Photo: D. Snoj) a b bili vzhodnoalpski marmor (sl. 6: C).22 Odkriti votivni oltarji so bili posvečeni Jupitru, Jupitru Dolihenu in Marsu (sl. 6: A,F).23 Z območja templja izvira še nekaj odlomkov manj­ših marmornih votivnih figur s pripadajočimi profilira­nimi podstavki (sl. 6: E,D), pa tudi najdba votivne plošče, posvečene Jupitru Depulsorju (sl. 6: G).24 Votivna plošča in posvetilni oltarji so datirani v 2. in 3. st.25 Winkler je leta 1911 vzhodno od sodobne ceste odkril ostanke najmanj štirih objektov (sl. 4), od katerih je bila bivanjska stavba z dvoriščem locirana v osrednjem delu. Podobno zasnovani tlorisi stavb so bili odkriti v letih 1993 in 199626 (sl. 7), 1999 (sl. 9)27 in 2010–201128 (sl. 4). Na slednjih dveh sta bili odkriti dve oz. tri faze poselitve. Po najdbah sodeč spadajo odkriti objekti v čas od 1. do 3. st. Hiša, odkrita leta 1996, je bila eno­prostorna, s treh strani obdana z dvoriščem (ohranjena zahodna stranica dvorišča v dolžini 11,5 m), kakršne so značilne za naselbine v južnem Noriku.29 Odkriti temelji stavb v Kolacioni so kamniti, leseni (sl. 9; 19),30 ali grajeni v stojkasti tehniki (sl. 20).31 Za gradnjo temeljev so uporabili lomljence (sl. 7; 9), vogali so ponekod ojačani z večjimi apnenčastimi kamni (sl. 18; 19), nekateri temelji stavb so grajeni iz manjših oblic.32 Tla objektov so tlakovana z manjšimi kamni (sl. 9). 22 Djura Jelenko, Visočnik 2006, 357, 359, kat. št. 17, 20. 23 Djura Jelenko, Visočnik 2006, 383–386, kat. št. 26–29. 24 Djura Jelenko, Visočnik 2006, 385, kat. št. 29. 25 Djura Jelenko 2004; Djura Jelenko, Visočnik 2006. 26 Britovšek 2007. 27 Djura Jelenko 2001. 28 Djura Jelenko, Kajzer Cafnik 2011. Groh, Sedlmayer 2006, 113–115; Britovšek 2007, 75, pril. 3; Djura Jelenko, Groh 2006, 406.30 Djura Jelenko, Kumprej Gorjanc 2009, sl. 28. 31 Djura Jelenko, Kumprej Gorjanc 2009, sl. 29. 32 Britovšek 2007, 76, sl. 18. cella and the peripheral wall. Even the position of the cella inside the temple is not fully understood. A fragment of a Corinthian column, discovered during the temple ex­cavations, and a column from the church of St Pancras, indicate that Eastern Alpine marble was used for the col­onnade (Fig. 6: C).22 The discovered altars were dedicated to Jupiter, Jupiter Dolichenus, and Mars (Fig. 6: A,F).23 Several marble fragments of votive monuments (Fig. 6: E,D) originate from the temple area as well as a relief dedicated to Jupiter Depulsor (Fig. 6: G).24 The relief and altars are dated to the 2nd and 3rd centuries.25 In 1911, Winkler discovered the remains of at least four buildings, the central of which was a residential struc­ture with a courtyard (Fig. 4). Similarly designed building layouts were discovered between 1993 and 199626 (Fig. 7), in 1999 (Fig. 9)27, and 2010/2011 (Fig. 4).28 In the latter two, two or three settlement phases were discovered. On the basis of the finds, these buildings date from the 1st to the 3rd century. A single cell house, surrounded by a courtyard from three sides, was discovered in 1996. Western side of the yard survived only, which was 11.5 m long. Such houses were typical for settlements in southern Noricum.29 Building foundations in Colatio were made of stone, horizontal wooden beams (Figs. 9; 19)30 or verti­cal wooden posts (Fig. 20).31 They used stone rubble for 22 Djura Jelenko, Visočnik 2006, 357, 359, Cat. No. 17, 20. 23 Djura Jelenko, Visočnik 2006, 383–386, Cat. No. 26– 29. 24 Djura Jelenko, Visočnik 2006, 385, Cat. No. 29. 25 Djura Jelenko 2004; Djura Jelenko, Visočnik 2006. 26 Britovšek 2007. 27 Djura Jelenko 2001. 28 Djura Jelenko, Kajzer Cafnik 2011. 29 Groh, Sedlmayer 2006, 113–115; Britovšek 2007, 75, App. 3; Djura Jelenko, Groh 2006, 406. 30 Djura Jelenko, Kumprej Gorjanc 2009, fig. 28. 31 Djura Jelenko, Kumprej Gorjanc 2009, fig. 29. Petrološka sestava gradbenega materiala za zidove je bila določena ob izkopavanjih leta 1996. Prevladuje­jo apnenci, dolomiti, filitoidi, skrilavci in peščenjaki. Apnenec so domnevno pridobivali v neposredni bližini naselbine. Iz krednega apnenca je namreč sestavljendel območja Gradišča in sosednjega Šanclnovega vrha. Najbližji dolomit je na sedlu Grajske vasi, na območju opuščenega kamnoloma.33 Hipokavst je bil v skromnih sledovih odkrit pri raziskavah leta 1999,34 o ostankih centralne kurjave piše tudi Winkler.35 Ognjišče je bilo odkrito leta 1996 v dvoriščnem delu rimskodobnega objekta. Na podlagi najdb je datirano v prvo polovico 2. st.36 Na območju, ki je bilo interpretirano kot prostor zunaj objekta, sta bili leta 2011 odkriti dve kurišči oz. podlagi za kurišči. Na podlagi trenutnih rezultatov lahko rečemo, da najverjetneje pripadata tretji, najmlajši fazi oz. 3. st. (sl. 10). Med najdbami velja omeniti sigilatno posodje, žigosane opeke, oljenke ter grafite na keramiki.37 V V nanosih pohorskih grap med Slovenj Gradcem in Bukovsko vasjo pa najdemo skupaj filitoide, amfibolski skri­lavec in grödenski peščenjak ter dokumentirani vulkanski kamenini. Poročilo Iva Štrucla z dne 15. 9. 1996. Hrani KPM. 34 Djura Jelenko 2001, 190. 35 Djura Jelenko 2004, 24. 36 Britovšek 2007, 74–76, pril. 3. 37 Djura Jelenko, Kumprej Gorjanc 2009, 48, t. 13: 74,76,78, the foundations (Figs. 7; 9) and the corners could have been reinforced with large limestones (Figs. 18; 19). Some building foundations were built using pebbles.32 The floors could have been paved with small stones (Fig. 9). The petrological analysis was made on the wall stones during the excavations of 1996. Limestone, dolo­mite, philitoids, slate, and sandstone were predominant. Limestone was presumably acquired in the immediate vicinity of the settlement. Specifically, chalk has beenshown to compose parts of the Gradišče and Šanclnov vrh areas. The nearest dolomite can be found on the saddle of Grajska vas, in the abandoned quarry area.33 Modest traces of a hypocaust were discovered dur­ing the excavations of 1999,34 and Winkler also mentions central heating remains.35 A hearth from the first half of the 2nd century was discovered in the courtyard area of a Roman building in 1996.36 Two hearths, namely hearth bases, were discovered in 2011 in the area outside the buildings. They probably 32 Britovšek 2007, 76, fig. 18. 33 Filitoids, amphibole slate and Gröden sandstone can all be found in the deposits of Pohorje ravines between Slovenj Gradec and Bukovska vas as well as two volcanic rocks. Reportby Ivo Štrucl from September 15 1996. Archived by Koroški pokrajinski muzej. 34 Djura Jelenko 2001, 190. 35 Djura Jelenko 2004, 24. 36 Britovšek 2007, 74–76, App. 3. Kolacioni je bilo odkrito nekaj svinca nedoločenih oblik, precej pa tudi svinčenih plomb, s katerimi so popravljali ne samo sigilato, temveč tudi kuhinjsko posodje.38 Ohra-nil se je železen pečatni prstan z gemo (sl. 8).39 GROBIŠČE (Sl. 3; 11–18) Zgodnjerimsko južno grobišče je Winkler raziskal v letih 1911 in 1912, pri čemer je odkril številne arhi­tekturne dele bogatih grobnic (sl. 11). Razporejene so bile v smeri vzhod–zahod. Skromni ostanki grobišča so se ohranili še na južni strani današnje ceste Slovenj Gradec–Ravne.40 V vzhodnem in zahodnem delu nekro­pole41 so bile sredi obzidanih grobnih parcel postavljene razkošne grobnice, ki so pripadale podeželski eliti, v zahodnem delu grobišča pa so bili locirani manjši zidani grobovi (sl. 11). Zidane grobnice najdemo v južnem in jugovzhodnem delu province Norik. Najbolj izstopa okolica Viruna, pa tudi močno poseljena območja mest Flavije Solve/Wagna pri Lipnici, Celeje in Teurnije/Sv. Peter v Lesu.42 V vzhodnem delu grobišča sta se ohranili dve večji grobni parceli43 z obodnim zidom in ostanki na sredini stoječe grobnice.44 Grobna parcela grobnice I je merila 50 čevljev po širini in dolžini, podatek se je ohranil na mejniku grobne parcele (sl. 11: I).45 Poleg tega je Winkler odkril še zidane grobove, od katerih so se ohranili spodnji deli zidanih kamer (sl. 11: IV,V); leta 2004 in 2006 so bili odkriti še trije utrjeni platoji, pogojno interpretirani kot podstavki za spome­nike (sl. 11: III; 16). Na severozahodni strani grobišča se je ohranila še ena velika grobna parcela46 z grobovi, ki so bili vkopani zunaj grobne ograje (sl. 11: 1/77).47 Dela obzidane grobne parcele pa smo odkrili na dveh 79 (delavnice iz Rheinzaberna); Djura Jelenko 2004, 12 (ope­ke); – o. c., 24, t. 2: 10; 24: 184,185 (oljenke);Britovšek 2007, 39, kat. št. 125, sl. 9; Djura Jelenko 2004, t. 7: 61; 30 (grafiti). 38 Djura Jelenko 2004, sl. 36, t. 23: 175; 34: 249; Djura Je­lenko, Kumprej Gorjanc 2009, sl. 48, t. 3: 15; 12: 70; 15: 85.39 Djura Jelenko 1998, 131–133. 40 Strmčnik Gulič 1981; 1984; Djura Jelenko, Kumprej Gorjanc 2009. 41 Geofizikalno prospekcijo je v letu 2005 izvedlo pod-jetje Archeo Prospections z Dunaja. Meritve so pokazale, da se vzhodneje od odkritih Winklerjevih grobnic grobišče ne nadaljuje (Djura Jelenko, Groh 2006, Abb. 6). 42 Kremer 2001. 43 Grobna parcela grobnice I je merila 17 × 14,5 m (tako Winkler), grobnice II pa 18 × 18 m.44 Djura Jelenko 2004; Djura Jelenko, Groh 2006. 45 To pomeni pribl. 15 × 15 m; Djura Jelenko 2004, sl. 30; Djura Jelenko, Visočnik 2006, 382, kat. št. 25.46 V velikosti 11 × 9 m. 47 Strmčnik Gulič 1981. belong to the third (and the latest) phase, that is in the 3rd century (Fig. 10). Among the finds, terra sigillata pottery, stamped bricks, oil lamps, and graffiti on ceramics are all worth mentioning.37 Some lead of indeterminate shapes was discovered as well as lead clamps used to repair both terra sigillata and kitchen pottery.38 An iron finger ring with an engraved gem was found (Fig. 8).39 CEMETERY (Figs. 3; 11–18) The early Roman cemetery, explored by Winkler in 1911 and 1912, revealed the architecture of opulent tombs (Fig. 11). They were arranged in the east-west direction. Some modest cemetery remains were also preserved on the southern side of the current Slovenj Gradec-Ravne road.40 Luxurious tombs for the rural elite were built in the centres of walled grave plots.41 Small masonry graves were located in the western part of the cemetery (Fig. 11). The large masonry tombs can be found in the southern and southeastern parts of the Noricum province, in the areas of Virunum, Flavia Slova, Celeia, and Teurnia.42 Two large funerary plots43 with a boundary wall and the remains of a central tomb survived in the eastern part of the cemetery (Fig. 11: I,II).44 The inscription on the boundary stone revealed that the funerary plot of Tomb I was 50 feet wide and long (Fig. 11: I).45 Winkler discovered the foundations of another two masonry tombs (Fig. 11: IV,V). Three masonry platforms discovered in 2004 and 2006 might be interpreted as foundations of grave monuments (Fig. 11: III; 16). A large funerary plot was located at the northwestern edge of the cemetery 46 as well as graves outside of the plot (Fig. 11: 1/77).47 The remains of walled plots were discovered 37 Djura Jelenko, Kumprej Gorjanc 2009, 48, pl. 13: 74,76, 78,79 (workshops from Rheinzabern); Djura Jelenko 2004, 12 (bricks); right there, 24, pl. 2: 10; 24: 184, 185 (oil lamps); Bri­tovšek 2007, 39, Cat. No. 125, fig. 9; Djura Jelenko 2004, pl. 7: 61; 30 (graffiti). 38 Djura Jelenko 2004, fig. 36, pls. 23: 175; 34: 249; Djura Jelenko, Kumprej Gorjanc 2009, fig. 48; pl. 3: 1515; 12: 70; 15: 85. 39 Djura Jelenko 1998, 131–133. 40 Strmčnik Gulič 1981; 1984; Djura Jelenko, Kumprej Gorjanc 2009. 41 A geophysical survey was carried out in 2005 by Ar-cheo Prospections from Vienna. The cemetery does not ex­tend beyond tombs discovered by Winkler (Djura Jelenko, Groh 2006, fig. 6). 42 Kremer 2001. 43 The burial plot of Tomb I measured 17 × 14.5 m (ac­cording to Winkler), the plot of Tomb II measured 18 × 18 m. 44 Djura Jelenko 2004; Djura Jelenko, Groh 2006. 45 This means approx. 15 × 15 m; Djura Jelenko 2004, fig. 30; Djura Jelenko, Visočnik 2006, 382, Cat. No. 25. 46 Measuring 11 × 9 m. 47 Strmčnik Gulič 1981. ločenih mestih še med raziskavami krožišča leta 2004 (sl. 11: III; 16; 17). Večino odkritih kamnitih spomenikov predstav­ljajo nagrobne plošče z napisi, najden pa je bil tudi zgornji levi zaklink stele z upodobitvijo delfina.48 Na več mestih se omenjajo deli kamnitih ograj – loricae; še en dekorativni element grobne arhitekture je bil odkrit med raziskavami 2004 (sl. 11: III).49 Odlomek preklade z napisom so našli leta 1977 v drugotni legi, kot del konstrukcije groba.50 Vsi odkriti grobovi razen dveh, raziskanih leta 2004 (sl. 11: III/rdeča pika; 16), so žgani. Družbeno razslojenost nam poleg elitnih rodbinskih pokopov v obzidanih grobnih parcelah izpričujejo preproste grobne jame z žganino in brez vsakršnih pridatkov. at two separate locations during the excavations for a roundabout in 2004 (Fig. 11: III; 16; 17). The inscribed tombstones were found as well as the upper left corner of a stele depicting a dolphin.48 There are several mentions of loricae discoveries; another decora­tive element of funerary architecture was found during the excavations in 2004 (Fig. 11: III).49 A fragment of an inscribed lintel was discovered in 1977 in a secondary position, as part of the grave construction.50 All graves found in 2004, with the exception of two (Fig. 11: III-red dot; 16), were cremations. Social stratification is indicated by elite family burials in walled funerary plots as well as simple grave pits with nothing but cremated remains. 48 Djura Jelenko, Visočnik 2006. 48 Djura Jelenko, Visočnik 2006. 49 Djura Jelenko, Kumprej Gorjanc 2009, sl. 44. 49 Djura Jelenko, Kumprej Gorjanc 2009, fig. 44. 50 Djura Jelenko, Visočnik 2006, 380, kat. št. 21. 50 Djura Jelenko, Visočnik 2006, 380, Cat. No. 21. Sl. 13: Stari trg pri Slovenj Gradcu, grobnica I. Odlomek nagrobnega spomenika (AIJ 8). Rekonstrukcija po: Kremer 2001, sl. 148. Fig. 13: Stari trg pri Slovenj Gradcu. Tomb I. Tombstone frag­ment (AIJ 8). Reconstruction after: Kremer 2001, Fig. 148. (Foto / Photo: N. Lackner) Že leta 1911 je grobnico I delno raziskal Winkler, splet okoliščin pa je povzročil, da smo šele v letih 2004 in 2006 rekonstruirali njen celotni inventar.51 Grobnica I je izjemna tako po najdbah, položaju, kot veličastni ar­hitekturi (sl. 11: I; 12). Na sredini 250 m2 velike grobne parcele je stala grobna hiša, krita z opečnato streho, v notranjosti razdeljena na dva prostora. Glede na velikost tlorisa centralne grobne hiše (5,2 × 5,5 m) in temeljev širine 90 cm domnevamo, da je bil omogočen dostop 51 Djura Jelenko 2004; Djura Jelenko, Groh 2006; Groh, Djura Jelenko 2013. Tomb I was already partly explored by Winkler in 1911. Due to different circumstances, it was only possible to reconstruct its entire inventory in 2004 and 2006.51 Tomb I is significant for its finds, position and remark­able architecture (Figs. 11: I; 12). A large funerary house roofed with brick tiles and divided into two rooms stood in the centre of the 250 m2 funerary plot. It is assumed that interior access was possible due to the total size of its foundation (5.2 × 5.5 m) and a wall thickness of 90 cm. 51 Djura Jelenko 2004; Djura Jelenko, Groh 2006; Groh, Djura Jelenko 2013. v njeno notranjost. V grobnici I so v tretji četrtini 1. st. pokopali pripadnika mestnega vrhnjega sloja sosednjega municipija Celeje, najverjetneje enega od obeh županov, skupaj z dvema ženskama. Omenjeni nagrobnik z relie­fom kurulskega stola (sl. 13) pripada tej grobnici. Med pridatki izstopa razkošna mrtvaška postelja (kliné; sl. 15). Noriške mrtvaške postelje iz koščenih izrezljanih delov se na podlagi primerjav iz Italije pojmujejo kot uvoženi kosi. Dobre primerjave najdemo v severni Italiji, kjer so datirane že v prvo polovico in v sredino 1. st. Ostanki koščenih postelj so v severozahodnih provincah znani samo z 28 najdišč, od katerih sta bili samo dve najdeni v Noriku, obe v južnem delu Norika.52 Dve pasni gar-nituri, pinceta, šivanka in zapestnica pripadajo dvema ženskima osebama. Luksuzne najdbe iz grobnice I še dodatno podčr­tujejo pomen province, bogate z rudnimi bogastvi, v kateri se je tudi na podeželju oblikoval avtohton višji sloj. Primerjave za takšno obliko grobnic zasledimo na grobiščih rimskih zaselkov v Globasnici/Globasnitz, Hallstatt-Lahnu in Gleisdorfu.53 A male member of Celeia’s town elite (presumably a duumvir) was buried in Tomb I along with two women in the third quarter of the 1st century. The tombstone with the relief of a sella curulis belongs to this tomb (Fig. 13). A lavish deathbed (kline; Fig. 15) stands out among the grave goods. On the basis of comparison, Noricum deathbeds made of carved bone are considered to have been imported from Italy. North Italian analogies are dated to the first half and the mid-1st century. Bone deathbeds are only known from 28 sites in northwestern provinces, only two of which are located in Noricum — both in its southern part.52 The remains of two belt fittings, pincers, a needle, and a bracelet belonged to two women. The luxurious finds from Tomb I underline the im­portance of the autochthonous upper class settled in the countryside of the province, rich in mineral resources. Similar elite tombs can be found in the cemeteries of small Roman settlements in Globasnitz, Hallstatt-Lahen, and Gleisdorf.53 52 Djura Jelenko, Groh 2006. 52 Djura Jelenko, Groh 2006. 53 Kremer 2001, 358. 53 Kremer 2001, 358. Sl. 17: Stari trg pri Slovenj Gradcu, rimsko grobišče in poznorimska naselbina. Raziskave leta 2004 (lokacija na sl. 11: rdeča pika ob III). Poznorimska skeletna grobova sta poškodovala zgodnjeantični žgani zidan grob (kamnite plošče) znotraj obzidane grobne parcele. V ozadju starejša jama z jarkom. Pogled proti jugu. Fig. 17: Stari trg pri Slovenj Gradcu, Roman cemetery and late Roman settlement. Excavations in 2004 (location Fig. 11: III/red dot). Late Antique inhumation graves damaged the early Roman cremation grave (stone slabs) inside a walled funerary plot. An earlier pit with a ditch is visible in the background. View to the south. (Foto / Photo: P. Juvan) Dva ženska skeletna grobova sodita v poznorimski čas (sl. 11: III/rdeča pika; 17). Prva oseba je bila stara med 30–40 let in je obolela za paradontozo, druga je bila desničarka, starejša od 50 let, verjetno celo 60 ali več in se ni ukvarjala s težjimi fizičnimi deli.54 STAVBE NA OBMOČJU GROBIŠČA Že Winkler je odkril poznorimske stavbne ostanke na vzhodni in zahodni strani grobnice II (sl. 11). Med raziskavami 2004–2005 in 2006–2007 so se pokazali ostanki kamnitih in lesenih stavb (sl. 11: III,VIII/06; 19; 20), tlakovane površine, apnene jame, peč (sl. 18) in 54 Djura Jelenko, Kumprej Gorjanc 2009, 23. Sl. 19: Stari trg pri Slovenj Gradcu, naselbina. Raziskave leta 2007 (lokacija na sl. 11: VIII/6). Tloris zidanega objekta iz prve polovice 4. st. Fig. 19: Stari trg pri Slovenj Gradcu, settlement. Excavations in 2007 (location Fig. 11: VIII/6). Stone building from the first half of the 4th century. (Izvedba / Elaborated by: L. Jelenko) Sl. 20: Stari trg pri Slovenj Gradcu, naselbina. Raziskave leta 2005 (lokacija na sl. 11: III). a – Objekt II grajen v stojkasti tehniki z jamo v notranjosti. b – V jami so se ohranili ostanki lesa. Pogled proti zahodu (a) in severu (b). Fig. 20: Stari trg pri Slovenj Gradcu. Excavations in 2005 (loca­tion Fig. 11: III). a – Building II was constructed with vertical wooden posts. A pit was found in the interior. b – A pit with wooden remains. Views to the west (a) and north (b). (Foto / Photo: P. Juvan) The two inhumation graves date to the late Roman period (Fig. 11: III-red dot; 17). They belong to two women. The first one was aged between 30 and 40 years and suffered from periodontosis while the second one was right-handed, older than 50 (likely even 60) and did not partake in heavy physical labour.54 BUILDINGS IN THE CEMETERY AREA Winkler already discovered late Roman building remains on the eastern and western sides of Tomb II (Fig. 11). The remains of stone and wooden buildings (Fig. 11: III,VIII/06; 19; 20), paved surfaces, lime pits, a furnace (Fig. 18), and a large quantity of slag and lead were discovered during the excavations of 2004/2005 and 2006/2007.55 Nearly 80% of the coins discovered during the excavations in 2006/2007 (Fig. 11: VIII/06) date to the 3rd and 4th centuries.56 54 Djura Jelenko, Kumprej Gorjanc 2009, 23. 55 Djura Jelenko, Kumprej Gorjanc 2009. 56 16 coins from the 3rd century, 60 coins from the 4th century. večja količina žlindre ter svinca.55 Analiza novčnih najdb je pokazala, da slabih 80 % odkritih novcev iz raziskav 2006–2007 (sl. 11: VIII/06) sodi v 3. in 4. st.56 ŽIVALSKE KOSTI Borut Toškan in Janez Dirjec sta analizirala žival­ske ostanke iz rimskodobne Kolacione.57 Del najdenih kosti in zob predstavlja ostanek pogrebnih pojedin in grobnih popotnic.58 Za ekonomijo Kolacione in njenega zaledja je bila najpomembnejša domača žival govedo kot primarni vir mesa, maščob, mleka, kot delovna žival in za lokalni tovorni transport. Pri drobnici prevladujejo ostanki živali, ki so bile stare nad 3 leta. To kaže na velik pomen volne v lokalnem kmečkem gospodarstvu. Med zgodnjerimskim gradivom prevladuje domače govedo, v poznorimskem gradivu pa je delež omenjene vrste pov­sem enak deležu drobnice in prašiča.59 To bi kazalo na spremembo v prehrambnih navadah lokalnega prebival­stva, ki jih je prinesla romanizacija. Ugotovljeni razkorak avtorja pripisujeta sami naravi obeh vzorcev: obrednim jedem kot sestavnim delom pogrebnih pojedin na eni strani in vsakodnevnim obrokom poznorimskega pre­bivalstva na drugi. OKOLICA KOLACIONE Iz 1. in 2. st. poznamo iz bližnje okolice Kolacione ostanke vile rustike v Zgornjih, domnevno pa tudi v Spodnjih Dovžah (sl. 1: 6,7).60 Koncentracija najdenih rimskih kamnitih spomenikov kaže na močnejšo pose-litev še v Šentvidu pri Završah nad Gornjim Doličem in vzdolž rimske ceste, ki se je po Mislinjski Dobravi spustila proti Kolacioni in nadaljevala pot proti severu – proti avstrijski Koroški.61 Poznorimske in poznoantične najdbe so znane s Puščave62 in Turjaka nad Mislinjo.63 55 Djura Jelenko, Kumprej Gorjanc 2009. 56 16 novcev iz 3. st., 60 novcev iz 4. st. 57 ZRC SAZU – Inštitut za arheologijo; 2004–2005/Kro­ žišče, 2006–2007/Korobenz.58 Djura Jelenko, Kumprej Gorjanc 2009, 32. 59 Djura Jelenko, Kumprej Gorjanc 2009, sl. 43. 60 Djura Jelenko 2010, 66. 61 Djura Jelenko, Visočnik 2006. 62 Pleterski, Belak 2002. 63 Detektorske najdbe novembra 2016 ter topografija so pokazale, da lahko na Turjaku upravičeno pričakujemo po­ znoantično postojanko. ANIMAL BONES Borut Toškan and Janez Dirjec analyzed animal remains from Colatio dating to the Roman period.57 A certain amount of bone and teeth formed a part of grave goods, or they were the remains of burial feast.58 The most important domestic animals were cattle: as a primary source of meat, fats, milk, as working animals, and for local freight transport. Remains of animals older than 3 years are prevalent among small ruminants. This points to significant importance of wool in the local farm economy. Domestic cattle are predominant in the early Roman period, while the proportions between the latter and goat, sheep and pig are entirely even in the late Ro­man material.59 This indicates a change in dietary customs of the local population brought forth by Romanization. The authors attribute this distinction to the nature of both samples: ritual dishes as an integral part of funeral feasts on one side, and everyday meals of the late Roman population on the other. COLATIO’S SURROUNDINGS Villa rustica remains in Zgornje Dovže, purportedly even Spodnje Dovže, are known from the 1st and 2nd centuries (Fig. 1: 6–7).60 The concentrations of Roman stone monuments indicate settlement centres in Šentvid near Završe and along the Roman road, which descended toward Colatio in Mislinjska Dobrava and continued its route toward Austrian Carinthia.61 Late Roman and late antique finds are known from Puščava62 and Turjak.63 57 ZRC SAZU – Inštitut za arheologijo; 2004–2005/Kro­ žišče, 2006–2007/Korobenz.58 Djura Jelenko, Kumprej Gorjanc 2009, 32. 59 Djura Jelenko, Kumprej Gorjanc 2009, fig. 43. 60 Djura Jelenko 2010, 66. 61 Djura Jelenko, Visočnik 2006. 62 Pleterski, Belak 2002. 63 In November 2016, detector finds and topography showed that we can legitimately expect a late antique station at Turjak. STATUS NASELBINE IN POSAMEZNIKOV Manjše naselbine, kakršna je bila Kolaciona, so lahko imele različne statuse: postaja (statio), trg (forum), vas (vicus ali pagus).64 Arheološke najdbe iz Kolacione ne omogočajo zanesljive opredelitve statusa naselbine. Nedvomno je odigrala pomembno prometno in eko­nomsko vlogo, bila je versko in upravno središče krajine z minimalno avtonomijo. Na podlagi upodobitve na Tabuli Peuntigeriani je jasno le, da je opravljala (tudi) funkcijo poštne postaje.65 Prikaz uradniškega stola (sella curulis) je na grobu služil za osebno poveličevanje, kot simbol oblasti. Upo­dabljali so ga samo nosilci višjih funkcij, kar je župan iz Celeje, ki je bil pokopan v grobnici I v Kolacioni, nedvomno bil.66 Ob kurulskem stolu se je na nagrob­nem spomeniku67 ohranila podoba liktorja (sl. 13), ki je spremljal visoke uradnike. Iz province Norik poznamo upodobitve kurulskih stolov iz Bad Waltersdorfa, Grad-ca (Eggenberg), Celeje in Petovione.68 V grobnici I iz Kolacione so bili poleg nagrobnika odkriti tudi izjemni kosi koščenega pohištva – ostanki mrtvaške postelje, ki samo potrjujejo visok položaj pokojnika (sl. 15). Na nagrobni plošči za Lucija Apuleja lahko beremo, da je imenovani centurion (?) 10. dvojne, pobožne in zveste legije, eden od obeh županov (Klavdije Celeje), (postavil grob) Rufiji Krispini in Luciju Menavdoniju (sl. 14). Plošča69 je bila najdena v 60. letih 19. stoletja v grobnici II (sl. 11: II).70 Datirana je v 2. st., ko je legija službovala v Panoniji.71 Lucij Apulej je po opravljeni vojaščini prevzel funkcijo župana v Celeji. Iz maloštevilnih ohranjenih napisov izvemo, da je bilo prebivalstvo staroselsko, keltizirano, ni pa zanemar­ljiv niti delež latinskih imen, torej rimskih prišlekov – trgovcev, vojakov.72 64 Groh za Kolaciono navaja status vikusa (Djura Jelenko, Groh 2006).65 Tab. Peut. IV, 2. 66 Djura Jelenko 2004, 28; Djura Jelenko 2012, 43. 67 AIJ 8; RISt 379; ILLPRON 1910. 68 Walde 2005, 55. 69 AIJ 9; RISt 380. 70 Djura Jelenko 2004, 37, sl. 15. 71 Djura Jelenko, Visočnik 2006, 380, kat. št. 22. 72 Djura Jelenko, Visočnik, 2006. STATUS OF THE SETTLEMENT AND INDIVIDUALS Small settlements such as Colatio could hold dif­ferent status: statio, forum, vicus, or pagus.64 No material evidence was found in Colatio that could confirm its status more precisely. It undoubtedly had an important transport and economic role; it was a religious and administrative centre with minimal autonomy. On the basis of Tabula Peutingeriana, it is clear that it (also) functioned as a postal station.65 The depiction of a sella curulis on Tomb I served as a means of personal glorification, as a symbol of authority. It was only used for holders of higher functions, which was the case for the duumvir of Celeia, who was probably buried in Tomb I in Colatio.66 A lictor, which accompa­nied higher-ranking individuals, was depicted next to the curule seat (Fig. 13). 67 In Noricum, the curule seats are known from Bad Waltersdorf, Graz/Eggenberg, Celeia, and Poetovio.68 Exceptional pieces of bone furniture and deathbed remains from Tomb I confirm the high rank of the deceased (Fig. 15). The grave inscription of Lucius Appuleius is further significant evidence. Lucius Appuleius who was possibly a centurion of the Legio X Gemina and a duumvir (of Claudia Celeia) had erected the tomb to Rufia Crispina and Lucius Menaudonius (Fig. 14). The tombstone69 was found in the 1860s in Tomb II (Fig. 11: II).70 It dates to the 2nd century when the legion served in Pannonia.71 After completing his military service, Appuleius assumed the post of mayor in Celeia. The few surviving inscriptions indicate the autoch­thonous Celtic population with a fraction of Latin names that probably represent Roman immigrants: merchants and soldiers.72 Translation: Lucija Jelenko 64 Groh writes of a vicus status in relation to Colatio (Djura Jelenko, Groh 2006).65 Tab. Peut. IV, 2. 66 Djura Jelenko 2004, 28; Djura Jelenko 2012, 43. 67 AIJ 8; RISt 379; ILLPRRON 1910. 68 Walde 2005, 55. 69 AIJ 9; RISt 380. 70 Djura Jelenko 2004, 37, fig. 15. 71 Djura Jelenko, Visočnik 2006, 380, Cat. No. 22. 72 Djura Jelenko, Visočnik, 2006. Okrajšave / Abbreviations AIJ = V. Hoffiller, B. Saria, Antike Inschriften aus Jugoslavien, Heft I: Noricum und Pannonia Superior. – Zagreb 1938. CIL = Corpus inscriptiorum Latinarum. CSIR = Corpus signorum imperii romani – Corpus der Skulp­turen der römischen Welt. ILLPRON = M. Hainzmann, P. Schubert, Inscriptionum la-pidariarum Latinarum provinciae Norici usque ad annum MCMLXXXIV repertarum indices, Berolini 1986. lupa = UBI ERAT LUPA, F. und O. Harl, http://lupa.at/ (Bild­datenbank zu antiken Steindenkmälern). RISt = E. Weber, Die Römerzeitlichen Inschriften der Steier-mark, Graz 1969. BRITOVŠEK, T. 2007, Rimska naselbina Kolaciona. Rezultati arheoloških izkopavanj v letih 1993 in 1996. – V / In: K. Keber (ur. / ed.), Iz preteklosti Koroške. Svet samorastnikov, Obzorja Koroške 1, 11–49, Ljubljana. DJURA JELENKO, S. 1998, Novosti iz Colatia. – Koroški zbornik 2, 117–134, Ravne na Koroškem. DJURA JELENKO, S. 1999, Arheologija Koroške krajine. Stalna razstava. Colatio – 90 let (Archaeology of the Carinthian region / Archäologie im slowenischen Kärnten). – Slovenj Gradec. DJURA JELENKO, S. 2001, Še nekaj drobcev iz življenja v prvih stoletjih na naših tleh ali kot bi nam ti skrivnostni (ne)znanci dobro podkurili pod nogami. – V / In: Koroški zbornik 3, 180–199, Ravne na Koroškem. DJURA JELENKO, S. 2004, Dr. Hans Winkler in njegov prispe­vek k arheologiji Mislinjske doline (Dr Hans Winkler und sein Beitrag zur Archaologie des Mislinjatals). – Slovenj Gradec. DJURA JELENKO, S. 2010, Arheološka podoba Mislinje z okolico. – V / In: Občina Mislinja. Zbornik 2010, 65–75, Mislinja. DJURA JELENKO, S. 2012, Dis Manibus Sacrum. Božanskim manom (Den göttlichen Mannas / To the divine Manes). – Slovenj Gradec. DJURA JELENKO, S., S. GROH, 2006, Ein frühkaiserzeitlicher Grabbau in der Südnekropole des norischen vicus von Colatio, Slowenien. – Archäologisches Korrespondenzblatt 36/3, 405–422. DJURA JELENKO, S., M. KAJZER CAFNIK 2011, Stari trg pri Slovenj Gradcu – arheološko najdišče Colatio. – Varstvo spomenikov. Poročila 48/2011 (2013), 249–250. DJURA JELENKO, S., M. KUMPREJ GORJANC 2009, Svet živih – svet mrtvih. Kolaciona – 100 let. – Slovenj Gradec. DJURA JELENKO, S., J. VISOČNIK 2006, Rimski kamniti spo­meniki slovenske Koroške / The Roman stone monuments of Slovenian Carinthia. – Arheološki vestnik 57, 345–415. DJURA JELENKO et al. 2016 = DJURA JELENKO, S., D. BO­ŽIČ, A. ŠEMROV, B. RAJŠTER 2016, Sokličeva zbirka. »Tu mam pa ilirskega poglavarja«. Arheologija in numizmatika. Katalog stalne razstave / The Soklič collection. »Here's the Illyrian chief«. Archaeology and numismatics. Catalogue of the permanent exhibition. – Slovenj Gradec. EGGER, R. 1914, Ausgrabungen in Norikum 1912/13. – Jahreshefte des Österreichischen Archäologischen Institutes in Wien 17, Beiblatt, 5–86. GROH, S. , S. DJURA JELENKO, 2013, Mrtvaške postelje z izrezljanimi kostmi iz Norika. – Studia universitatis here-ditati 1/1–2, 25–38. GROH, S., H. SEDLMAYER 2006, Forschungen im vicus ost von Mautern-Favianis. Die Grabungen der Jahre 1997–1999. – Der römische limes in Österreich 44. KOLŠEK, V. 1997, Rimska nekropola v Šempetru. Vodnik / Römische Nekropole im Šempeter. Führer. – Celje. KREMER, G. 2001, Antike Grabbauten in Noricum. Katalog und Auswertung von Werkstücken als Beitrag zur Rekon­struktion und Typologie. – Österreichisches Archäologi­sches Institut. Sonderschriften 36. PAHIČ, S. 1968, Najstarejša zgodovina Koroške krajine. – V / In: 720 let Ravne na Koroškem, 6–55, Ravne. PAHIČ, S. 1975, Stari trg. – V / In: Arheološka najdišča Slove­nije, 277, Ljubljana. PLETERSKI, A., M. BELAK, 2002, Grobovi s Puščave nad Starim trgom pri Slovenj Gradcu (Die Gräber von Puščava oberhalb von Stari trg bei Slovenj Gradec). – Arheološki vestnik 53, 233–300. STRMČNIK-GULIČ, M. 1979, Prazgodovinske gomile na Legnu pri Slovenj Gradcu (Vorgeschichtliche Grabhügel auf dem Legen bei Slovenj Gradec). – Arheološki vestnik 30, 101–150. STRMČNIK-GULIČ, M. 1981, Antično grobišče v Starem trgu pri Slovenj Gradcu (Roman cemetery at Stari trg near Slovenj Gradec). – Arheološki vestnik 32, 348–389. STRMČNIK-GULIČ, M. 1984, Najnovejši podatki iz Starega trga pri Slovenj Gradcu (Die neuesten Daten aus Stari trg bei Slovenj Gradec). – Arheološki vestnik 35, 185–224. STRMČNIK GULIČ, M. 1995, Doneski k arheološki podobi slovenjgraškega okoliša. – V / In: Slovenj Gradec in Mislinj-ska dolina 1, 31–77, Slovenj Gradec. TERŽAN B. 1990, Starejša železna doba na Slovenskem Šta­jerskem / The Early Iron Age in Slovenian Styria. – Katalogi in monografije 25. VOMER-GOJKOVIČ, M. 1996, Grobišče pri Dijaškem do-mu v Rabelčji vasi na Ptuju (Das Gräberfeld neben dem Studentenheim in Rabelčja vas bei Ptuj). – V / In: Ptujski zbornik VI/1, 229–312. WALDE, E. 2005, Im Herrlichen Glanze Roms. Die Bilderwelt der Römersteine in Österreich. – Innsbruck. Saša Djura Jelenko Koroški pokrajinski muzej Glavni trg 24 SI-2380 Slovenj Gradec sasa.djura.jelenko@kpm.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 325–338 ZAGRAD Saša DJURA JELENKO Izvleček Hribovit zaselek Zagrad se razprostira južno od Prevalj v Spodnji Mežiški dolini. Arheološko najdišče Zagrad leži ob državni cesti (via publica) med Kolaciono in Jueno, verjetno kot mejna postaja med administrativnima teritorijema Celeje in Viruna. Ob povodnjih reke Meže in gradbenih delih v bližini Brančurnikovega mostu je bilo vse od druge polovice 19. stoletja odkritih okoli 50 kamnitih delov rimskih grobnic in arhitektonskega okrasa stavb (domnevno svetišča). Največ odlomkov pripada edikulam; izdelane so bile iz vzhodnoalpskega marmorja. Verjetno gre za žgane grobove. Izjema je Brančurnikov sarkofag (skeletni pokop), izdelan iz marmorja, ki izvira iz kamnoloma marmorja Kraig. Nenapisne reliefe iz Zagrada lahko datiramo od sredine 2. do sredine 3. st. Po kamnitih spomenikih sodeč, je morala v bližini stati ali obcestna naselbina ali pa razkošna podeželska vila. Ključne besede: Norik, Zagrad, rimska doba, naselbina, svetišče, grobišče, edikula, sarkofag Abstract The hilly settlement of Zagrad lies south of Prevalje in the Lower Meža valley, as part of the Carinthian region. The archaeological site of Zagrad is situated next to the public road (via publica) between Colatio and Iuenna, presumably as a former border station between the administrative territories of Celeia and Virunum. Due to the overflowing of the Meža River and construction work in the vicinity of Brančurnik’s bridge, more than 50 stone parts of Roman tombs and architectural decoration (presumably belonging to a sanctuary) have been discovered since the second half of the 19th century. Most fragments belong to aediculae made of Eastern Alpine marble. The majority are probably cremation graves with the exception of Brančurnik’s sarcophagus, which is made of the Kraig marble and represents evidence of an inhuma­tion burial. The reliefs from Zagrad are dated from the mid-2nd to the mid-3rd century. Judging by the stone monuments, there was either a settlement or a luxurious rural villa in the vicinity. Keywords: Noricum, Zagrad, Roman period, settlement, sanctuary, cemetery, aedicula, sarcophagus https://doi.org/10.3986/9789610502586_17 Sl. 1: Zagrad, arheološka najdišča. Fig. 1: Zagrad, archeological sites. 1 Dobja vas; 2 Farna vas; 3 Prevalje; 4 Brančurnikov dol 1; 5 Brančurnikov dol 2; 6 Dobji dvor; 7 Dobja vas – Tomaževo; 8 Brančurnikov most; 9 Meža 2015. (Vir za podlago / Map source: http://gis.arso.gov.si/evode/profile.aspx?id=atlas_voda_Lidar@Arso) GEOGRAFSKA LEGA Pretežno hribovit zaselek Zagrad leži južno od Pre­valj1 v Spodnji Mežiški dolini, na ozemlju Koroške (sl. 1; 2). Reka Meža je med Prevaljami in Ravnami oblikovala do 1 km široko dolino, po kateri je že od nekdaj potekala razmeroma enostavna prometna povezava preko Mis-linjske in Mežiške doline v osrednji del vzhodnoalpskega prostora proti Celovški kotlini. Na avstrijskem vojaškem zemljevidu iz druge polo-vice 18. stoletja (sl. 3) so Prevalje označene kot Zgornje in Spodnje Jezero,2 v mapi franciscejskega katastra3 pa so v okolici gostilne Brančurnik (sl. 1: 8), kjer so bili od druge polovice 19. stoletja dalje odkriti posamezni rimski kamniti spomeniki, zanimivi toponimi u Toplize, Studenzen, per Studenze, per Zeste. Za cerkvijo sv. Barba­ 1 Nadmorska višina Prevalj znaša 411 m. 2 Od tod tudi cerkev sv. Device Marije na Jezeru. Arhiv Republike Slovenije, SI AS 178 – Franciscejski kataster za Koroško, k. o. Dobja vas (K294), Farna vas (K329) in Zagrad (K388). GEOGRAPHICAL LOCATION The mostly hilly settlement Zagrad lies south of Prevalje1 in the Lower Meža valley, as part of the Carin-thian region (Fig. 1; 2). The Meža River carved an up to 1 km wide valley between Prevalje and Ravne, which has always provided a relatively easy traffic route through the Mislinja and Meža valleys onto the central part of the Eastern Alpine area leading to the Klagenfurt basin. On an Austrian military map from the second half of the 18th century (Fig. 3), Prevalje is marked as Zgornje and Spodnje Jezero;2 it is also interesting to note that on the Franciscan cadastre map,3 in the vicinity of the Brančurnik Inn (Fig. 1: 8) where individual Roman stone monuments have been discovered since the second half of the 19th cen­ 1 The elevation of Prevalje is 411 m above sea level. 2 Also located here is the church of sv. Devica Marija na Jezeru (St Mary the Virgin at Lake). 3 SI AS 178 – Franciscan cadastre for Carinthia, cadas­tral communities Dobja vas (K294), Farna vas (K329), and Zagrad (K388). Sl. 2: Zagrad. Gostilna Brančurnik, cerkev sv. Barbare s pokopališčem in nekdanja tovarna Paloma z Brančurnikovim mostom (ob desnem robu). Posnetek iz zraka proti jugozahodu. (Foto: P. Juvan) Fig. 2: Zagrad. Brančurnik Inn, the church of St Barbara and the Brančurnik’s bridge. Aerial photo towards southwest. (Photo: P. Juvan) Sl. 3: Zagrad. Izsek iz vojaškega zemljevida iz let 1763–1787 (Rajšp 2000, sekcija 139). Fig. 3: Zagrad. Military map dated between 1763–1787 (Rajšp 2000, section 139). re leži še kmetija p. d. pri Temelju oz. Temeljev vrh (sl. 4). Zagrad pomensko sorodno izhaja iz toponima Gradišče, Grad. Leta 1975 so bila z Zagradom povezana tri praz­godovinska in eno rimsko najdišče (Truhlar 1975, 106). Brančurnikov sarkofag, domneven grad pri Temelju in jezero so se ohranili v mitskih zgodbah.4 4 “Vsako pripoved o nastanku Prevalj začenjamo z eno od ljudskih pripovedk, starih storij o jezeru. Nekoč je bilo jezero in pri Temelnu v Zagradu je stal grad. Tri grajske hčerke, ki tury, toponyms such as u Toplize, Studenzen, per Studenze, per Zeste4 are used. There is also a farm behind the church of St Barbara, informally known as: at Temelj’s or Temeljev vrh (Foundation hill; Fig. 4). The name “Zagrad” is associated with and derived from the toponym Gradišče, Grad. One Roman and three prehistoric sites were known in Zagrad in 1975 (Truhlar 1975, 106). Brančurnik’s sarcophagus, the 4 i.e. Spa; Water Spring; At Water Spring; Near the Road. Sl. 4: Zagrad. Izsek iz Franciscejskega katastra iz leta 1827. Fig. 4: Zagrad. Franciscan cadastre 1827. (Vir / Source: https://mapire.eu) Rimsko grobišče Zagrad leži na tromeji katastrskih občin Farna vas, Dobja vas in Zagrad (sl. 1: med 8 in 9). so se vozile po jezeru, so v nenadnem viharju utonile. Dva na smrt obsojena ujetnika sta si odkupila življenji s tem, da sta razbila Votlo peč. In ko je jezero odteklo tam pod Votlo pečjo, je dal graščak na mestih, kjer so našli hčerki, sezidati cerkvi – v spomin na Marijo cerkev sv. Device Marije na Jezeru, v spomin na Barbaro cerkev sv. Barbare in Rozaliji votlino (kapelico) pod cerkvijo sv. Barbare”. Tri graščakove hčerke, zapisala: Franc Kotnik, Franc Sušnik (Horjak 2012, 68). alleged castle at Temelj’s and the lake were maintained in folktales.5 Furthermore, the Roman cemetery Zagrad lies at the juncture of three cadastral municipalities: Farna vas, Dobja vas, and Zagrad (Fig. 1: between 8 and 9). 5 “We start every origin story of Prevalje with one of the folk legends, old tales about the lake. Once there was a lake and at Temeln in Zagrad stood a castle. Three castle daughters who rowed across the lake drowned in a sudden storm. Two prisoners sentenced to death bought their lives by breaking the rock Votla peč. And once the lake flowed out from be­neath Votla peč, the lord of the castle erected two churches at the spots where his daughters were found—in memory of Mary, the church of St Mary the Virgin at Jezero, in memory of Barbara, the church of St Barbara, and to Rosalia, a cham­ber (chapel) below the church of St Barbara.” [Tri graščakove hčerke (Three Castle Daughters),Written by Franc Kotnik and Franc Sušnik]. (Horjak 2012, 68). ZGODOVINA RAZISKAV Rimske marmorne spomenike iz Zagrada poznamo od leta 1860.5 Ob raznih zemeljskih delih so na Peruzzi­jevi žagi ob Plešivčnikovem (kasneje Brančurnikovem) mostu odkrili prvih dvajset večjih obdelanih kamnov (sl. 1: 8).6 M. F. Jabornegg navaja ostanke okroglega ajdovskega templja v bližini Meže in polkrožne odlomke stebrov.7 V strugi Meže, blizu mostu pri Brančurniku, so 1870 odkrili sarkofag, še enega pa so v začetku 20. stolet­ja pustili ležati v vodi.8 Leta 1894 je lastnik žage postal F. Lahovnik, ki je na tem mestu kmalu zgradil tovarno bele lesovine. Gradbena dela ob gradnji tovarniškega kanala ob Meži so razkrila še več obdelanih antičnih kvadrov z reliefi in brez njih. Dvignili so jih iz Meže ali pa so bili opaženi v njeni neposredni bližini.9 Kvader z nimfo in kip centuriona sta v začetku 20. st. razkrili povodnji reke Meže. Vseh kamnov naj bi bilo okoli petdeset. Nekatere (od devet do enajst) so vzidali v Lahovnikovo brusilnico za les, reliefne plošče, kvader in nekaj konstrukcijskih plošč danes hranijo na Ravnah10 in Prevaljah, za večino se je izgubila vsakršna sled. Jaro Šašel jih je leta 1953 naštel še trideset. Kip centuriona in reliefno ploščo z upodobitvijo Atisa so odpeljali v Koroški deželni muzej v Celovcu11 in le sarkofag je ostal v bližini mesta odkritja.12 Z izjemo sarkofaga in napisne plošče, vzidane v tovarni, ki pa ni bila nikoli prepisana, so vsi ostali spomeniki brez napisov.13 Obdelane kamne z grobišča Zagrad so sekundarno uporabili kot klopi, mize, vogalne kamne ob hišah in gostilnah na Prevaljah in v okolici. Na območju nekdanje tovarne Paloma, tovarne lepenke Prevalje,14 je Koroški pokrajinski muzej (v na­daljevanju KPM) leta 2006 izvedel arheološke raziskave (Brančurnikov dol 1; sl. 1: 4). Ugotovljeno je bilo, da so območje tovarne v preteklosti zaradi neposredne bližine poplavne Meže 2–3 m visoko nasuli z odpadnim mate-rialom, arheološki depozit v ozkem pasu med reko in vznožjem Barbarinega griča pa je zaradi večdesetletnega gospodarskega izkoriščanja močno poškodovan. M. Kumprej je v svoji diplomski nalogi (2008) raz­iskala Jueno (Globasnica/Globasnitz) v luči nagrobnih 5 Vermischte Nachrichten, Klagenfurter Zeitung 178 (4. 8. 1860), str. 719. 6 Med njimi tudi dele zidnih vencev (napušče). 7 Jabornegg 1870, 131. 8 Djura Jelenko, Visočnik 2006, 354, 375. 9 Šašel 1953; Djura Jelenko, Visočnik 2006, 349. 10 Do leta 2004 so bili spomeniki hranjeni v Bitenčevem prehodu Koroške osrednje knjižnice dr. Franca Sušnika na Ravnah na Koroškem. Po obnovi knjižnice je bil lapidarij prestavljen v kapelo gradu Ravne. 11 Kip oficirja hrani Koroški deželni muzej. Leta 2008 so ga prestavili v Globasnico, od leta 2011 je razstavljen v lapi­dariju sub divo ob arheološkem romarskem muzeju. 12 Djura Jelenko, Visočnik 2006; Kumprej 2008. 13 Djura Jelenko, Visočnik 2006, 376. 14 Identična lokacija kot pri zgoraj omenjenih tovarnah. RESEARCH HISTORY The Roman marble monuments from Zagrad have been known since 1860.6 During various fieldworks at Peruzzi’s sawmill next to Plešivčnik’s (later Brančurnik’s) bridge, the first twenty large pieces of worked stone were discovered (Fig. 1: 8).7 Jabornegg mentions the remains of a circular pagan temple near the Meža River as well as semicircular fragments of pillars.8 In the Meža riverbed near Brančurnik’s bridge, a sarcophagus was discovered in 1870, and another one was left in the water at the beginning of the 20th century.9 In 1894, Lahovnik became the owner of the sawmill and con­tinued to build a softwood factory on that spot shortly after that. Construction work for the factory channel along the Meža River revealed an even greater number of worked antique slabs, some of which had reliefs. They were either extracted from the Meža River or observed in its immediate vicinity.10 A cube depicting a nymph and a centurion statue were revealed by two floods of the Meža River at the beginning of the 20th century. The total number of stones is estimated at around fifty. A number of them (nine to eleven) were built into La­hovnik’s sawmill, and while some, like the relief slabs, the cube, and some construction slabs, are now kept in Ravne11 and Prevalje; however, most were lost without a trace. In 1953, Jaro Šašel noticed thirty of them. The centurion statue12 and the relief slab of Attis were taken to the Regional Museum of Carinthia in Klagenfurt; only the sarcophagus remained near the discovery site.13 With the exception of the sarcophagus and the inscription slab built into the factory, which has never been transcribed, all remaining monuments carried no inscriptions.14 Worked stones were used secondarily as various benches, tables, and cornerstones near houses and inns in Prevalje and its vicinity. In 2006, in the area of the former Paloma factory and Prevalje cardboard factory,15 the Koroški pokrajinski muzej carried out archaeological research (Brančurnikov dol 1; Fig. 1: 4). It was discovered that due to common 6 Vermischte Nachrichten, Klagenfurter Zeitung 178 (4. 8. 1860), 719. 7 Including parts of the geison. 8 Jabornegg 1870, 131. 9 Djura Jelenko, Visočnik 2006, 396, 408. 10 Šašel 1953; Djura Jelenko, Visočnik 2006, 393–394. 11 Until 2004, the monuments were stored in the Bitenc passage of the Dr Franc Sušnik Central Carinthian Library. After the renovation of the library, the lapidarium was trans­ferred to the chapel of Ravne Castle. 12 The statue of the officer is kept by the Regional Mu­seum of Carinthia in Klagenfurt. In 2008, it was moved to Globasnitz and has been exhibited in the lapidarium sub divo next to the Archaeological Pilgrim Museum since 2011. 13 Djura Jelenko, Visočnik 2006; Kumprej 2008. 14 Djura Jelenko, Visočnik 2006, 408. 15 At the same location as the above-mentioned factories. spomenikov in pri tem ugotovila, da kamni iz Zagrada kažejo večjo podobnost s spomeniki iz Juene in Viruna kot pa z nagrobniki iz celejskega agra.15 PRAZGODOVINA Ploščata dletasta sekira izvira iz Dobje vasi (sl. 1: 1), preluknjana kladivasta sekira iz Podkraja, prazgodovin­ske črepinje naj bi bile najdene za Kristanovo domačijo v Farni vasi 6 (sl. 1: 2).16 Na Prevaljah naj bi bila najdena tudi tulasta bronasta sekira z ušescem in še dva kovinska (?) predmeta, ki pa sta neznano kje (sl. 1: 3).17 Lahko je šlo za depojsko ali grobno najdbo. Datirana je v Ha B.18 Odlomki prazgodovinske lončenine19 so bili odkriti med arheološkimi raziskavami lokacije Brančurnikov dol 1 (sl. 1: 4), nekaj atipičnih kosov pa tudi leta 2008 med izkopavanji lokacije Brančurnikov dol 2 (sl. 1: 5) južno od regionalne ceste Ravne–Prevalje oz. 300 m od gostilne Brančurnik.20 ARHEOLOŠKI OSTANKI RIMSKE DOBE (Sl. 5–12) Kljub velikemu številu obdelanih marmornih kamnov z grobišča Zagrad, ki naj bi bili najdeni bolj ali manj na istem kraju, še vedno ne poznamo natančne lokacije odkritja, še manj je znanega o pripadajoči vili oz. naselbini. Delno bi to lahko pripisali povodnjim reke Meže.21 Arheološke raziskave na obravnavanem obmo­ 15 Istega leta je Občina Prevalje začela projekt Vandrav-ski trg, kjer so osrednji prevaljški trg vsebinsko zapolnili z odlitki rimskih nagrobnih spomenikov iz Zagrada (Kumprej Gorjanc 2013). 16 Pahič 1975, 275. 17 Najdbe so odkupili iz zbirke Boehmker. Hrani jo Av-strijski arheološki inštitut na Dunaju. 18 Šinkovec 1995, 67, t. 17: 99. 19 8 kosov. 20 Poročilo o arheološkem orientacijskem sondiranju na lo-kaciji Prevalje (rimsko grobišče EŠD 7630) – Zagrad (z dne 11. 9. 2006) in Poročilo o arheološkem pregledu in sondiranjih na lokaciji Prevalje, Brančurnikov dol 2 (rimska cesta EŠD 7627) zaradi nameravanega odkupa parcel št. 135/1, 129/4, 129/7 in 130/6 k. o. Dobja vas s strani podjetja Inpos, d. o. o. (z dne 1. 9. 2008). Neobjavljeno, arhiv KPM. “Leta 1898 je bil decembra izredno hud mraz. Meža je zmrznila vse od Dravograda do Črne. Petega, šestega ja­nuarja naslednje leto pa je nenadoma prišla odjuga. Narasla voda je odnesla jez in še obrambni zid. V izgubo je šel ves material. Še isto leto so začeli graditi širši jez. Pri kopanju temeljev so med peskom in glenom v Meži odkrili zanimive stvari iz rimske dobe. Kipov, soh in rakev in drugega je bilo menda za cel vagon. Kipov je zdaj nekaj na Ravnah, nekaj v Mariboru, nekaj pa v Celovcu v muzejih” (Kordež 2003, 39; Kumprej 2008, 13). floods of the Meža River, its immediate vicinity was stacked 2-3 m high with waste material; archaeological deposits in the narrow belt between the river and the area beneath Barbara’s hill have been heavily damaged. In her undergraduate thesis (2008), M. Kumprej explored Iuenna (Globasnitz) with a central focus on tombstones and concluded that the stones from Zagrad showed greater similarity with statues from Iuenna and Virunum than those from the ager of Celeia.16 PREHISTORY The flat chisel axe originates from Dobja vas (Fig. 1: 1), another stone axe from Podkraj, and prehistoric shards were supposedly found behind Kristan’s home­stead at Farna vas 6 (Fig. 1: 2).17 The bronze axe was allegedly discovered at Prevalje together with two metal (?) objects, the present location of which remains un­known (Fig. 1: 3).18 It could have been a hoard or grave find dated to Ha B.19 Prehistoric pottery fragments20 were discovered during archaeological research of the Brančurnik dol 1 area (Fig. 1: 4), some atypical fragments were also found in 2008 during the excavations at Brančurnikov dol 2 (Fig. 1: 5), south of the Ravne–Prevalje regional road.21 ARCHAEOLOGICAL REMAINS OF THE ROMAN PERIOD (Figs. 5–12) Despite the high number of worked marble stones from the Zagrad cemetery which were supposedly found at roughly the same area, the exact discovery spot is still unknown, even less so the corresponding villa or settlement. This could be partly due to the Meža River floods.22 16 The copies of the Zagrad stones are exhibited in the central public square in Prevalje (Kumprej Gorjanc 2013). 17 Pahič 1975, 275. 18 They bought the finds from the Boehmker collection. It is kept by the Austrian Archeological Institute in Vienna. 19 Šinkovec 1995, 67, pl. 17:99. 20 8 pieces. 21 Unpublished reports, Archive of Koroški pokrajinski muzej. 22 “In 1898, December was extremely cold. The Meža River froze all the way from Dravograd to Črna. On the fifth or sixth of January next year, southern winds caused a sud­den thaw. Waters rose and took both the dam and the defen­sive wall. All material was lost. That same year, the construc­tion of a wider dam began. While digging for the foundation, interesting objects from the Roman period were discovered among the sand and sediment. Apparently, the amount of statues, coffins, and other objects could fill an entire wagon. The statues are now dispersed: some in Ravne and Maribor, Sl. 5: Zagrad. Torzo kipa oficirja. Fig. 5: Zagrad. Torso of the statue of an officer. (Foto / Photo: B. Bončina) čju, ki jih je KPM izvedel med letoma 2006 in 2008,22 niso razkrile sledov rimske ceste. Glede na pripovedi domačinov je verjetneje, da je potekala severno od ceste Ravne na Koroškem–Prevalje (sl. 1). Notico o templju je zabeležil Jabornegg; stari ljudje naj bi pripovedovali, da so se na mestu odkritja velike­ga števila “težkih” najdb na enem mestu videli okrogli sledovi temeljev “poganskega templja” in da njihova originalna lega ni dosti premaknjena.23 V prid svetišča govori odkritje obdelanega kamna, ki smo si ga septem-bra 2016 ogledali pri Erihu Sirku na Prevaljah (sl. 7).24 Arhitrav z dvema fascijama bi lahko pripadal nekemu svetišču, manj verjetno portiku. Marmorna plinta rimskega stebra kot spodnja pro-filacija podija oz. podstavka, na katerem stoji steber, bi prav tako lahko pripadala svetišču (sl. 8).25 Odkrili smo 22 Brančurnikov dol 1 (2006), Brančurnikov dol (2006, Hofer), Brančurnikov dol 2 (2008, Inpos). Neobjavljeno, ar­hiv KPM. 23 Šašel 1953, 16. 24 Vel. 129,5 × 35 × 34 cm. Kamen hrani pred svojo hišo (Ugasle peči 11) in ga je pred časom pripeljal s prevaljskega igrišča, kjer je vrsto let služil kot klop. Na kamen nas je opo­zorila M. Kumprej Gorjanc, za kar se ji lepo zahvaljujemo. 25 Glede na znan premer stebra, ki se ni ohranil, predvi­devamo višino stebra 2,85 m (za steber s korintskim kapite­lom; nekaj manj za steber z jonskim kapitelom). Za pomoč pri interpretaciji arhitrava in plinte se zahvaljujem Bojanu Djuriću. Archaeological research, carried out by the Koroški pokrajinski muzej between 2006 and 2008,23 has not re­vealed traces of a Roman road. According to local tales, it is more likely that it ran north of the Ravne–Prevalje road (Fig. 1). A note about the temple was recorded by Ja­bornegg: elderly local people reportedly mentioned that at the discovery site of a large quantity of “heavy” finds circular foundation traces of a pagan temple were seen at one spot and that their original location has not changed much.24 This is supported by a worked stone find that we examined at Erih Sirk in Prevalje in Sep­tember 2016 (Fig. 7).25 An architrave with two fasciae could have belonged to a temple, less likely to a portico. A marble plinth of a Roman column—the bottom of a profiled podium or pedestal on top of which the column was positioned—may have also belonged to a some in Klagenfurt museums” (Kordež 2003, 39; Kumprej 2008, 13). 23 Brančurnikov dol 1 (2006), Brančurnikov dol (2006, Hofer), Brančurnikov dol 2 (2008, Inpos). Reports (unpub­lished). Archive of Koroški pokrajinski muzej. 24 Šašel 1953, 16. 25 Size: 129.5 × 35 × 34 cm. The stone is kept outside the house (at Ugasle peči 11) and was transferred there some time ago from its previous location at a playground in Pre­valje, where it served as a bench for many years. The stone was brought to our attention by Maja Kumprej Gorjanc, for which we kindly thank her. Sl. 6: Zagrad. Deli nagrobnikov. Fig. 6: Zagrad. Fragments of tombstones. (Foto / Photo: T. Jeseničnik; archive of Kärntnerisches Landesmuseum, Klagenfurt) jo pri zasebnem zbiralcu Jožefu Kozlarju iz Poljane, ki hrani tudi nekaj drugih kamnov iz Zagrada.26 Kvader z nimfo bi lahko bil del kapele ali svetišča, posvečenega nimfam (sl. 9). Na znanih reliefih z upodob­ljenimi nimfami iz Flavije Solve (Leibnitz / Lipnica),27 Aquae Iasae (Varaždinske Toplice),28 Bretzfeld-Unter­heimbacha29 so predstavljene nimfe zlasti v nimfejih.30 Pogosto so bile čaščene v naravi, še posebej pri izvirih, jamah in vodnjakih.31 Spomnimo se na toponime iz franciscejskega katastra Prevalj z imeni u Toplize, Stu-denzen, per Studenze (sl. 4). Večina odkritih spomenikov pripada nagrobnim kapelicam – edikulam. Do danes je znanih približno 15 gradbenih elementov, ki pripadajo edikulam,32 štirje 26 Kumprej Gorjanc 2013, kat. št. 9,10,18. 27 Diez 1980, sl. 2. 28 lupa 5390. 29 lupa 7436. 30 Diez 1980, 107; Kastelic 1998, 269; Kumprej 2008, 101. 31 Käppel 2000, 1071; Kumprej 2008, 57. Djura Jelenko, Visočnik 2006, 349, kat. št. 5 (sl. 4), 353–354, kat. št. 13 (sl. 11); Kumprej Gorjanc 2013, 75. Štirje kamni v gostilni Kocka na Poljani (J. Kozlar), ostalo na Pre­valjah: do sedem plošč (nekdaj pri Lahovniku, sedaj A. Suš­nik); profiliran kvader (A. Grošelj; Kumprej Gorjanc 2013, kat. št. 12); profilirana plošča (A. Pezdirc Vehovar; Kumprej 2008, kat. št. 34); vogalni odbijač (bivša tovarna Paloma; Dju- temple (Fig. 8).26 It was discovered in a private collection of Jožef Kozlar from Poljana, who also stores some other slabs from Zagrad.27 A stone cube with a relief of a nymph could have been a part of a chapel or shrine dedicated to nymphs (Fig. 9). The reliefs of nymphs from Flavia Solva (Leibnitz),28 Aquae Iasae (Varaždinske Toplice)29 and Bretzfeld-Unterheimbach30 mainly originate from nym­phaea.31 The nymphs were often worshipped in nature, especially near springs, caves, and wells.32 Let us recall toponyms from the Franciscan cadastre of Prevalje with names such as u Toplize, Studenzen, per Studenze (Fig. 4). Most discovered monuments belong to funerary chapels — aediculae. To date, there are approximately 26 Given the preserved diameter of the column, the as­sumed height is 2.85 m (for the column with the Corinthian capital; a little less for the one with the Ionic capital). I would like to thank Bojan Djurić for his help with the interpretation of the architrave and plinth. 27 Kumprej Gorjanc 2013, Cat. No. 9, 10, 18. 28 Diez 1980, fig. 2. 29 lupa 5390. 30 lupa 7436. 31 Diez 1980, 107; Kastelic 1998, 269; Kumprej 2008, 101. 32 Käppel 2000, 1071; Kumprej 2008, 57. Sl. 7: Zagrad. Arhitrav (neobjavljeno; hrani Erih Sirk, Prevalje). Fig. 7: Zagrad. Architrave (unpublished; stored by Erih Sirk, Prevalje). kamni, ki pripadajo podnožju edikule (sl. 6a–c; 9; razli-čice A.1 po Kremer 2001), plošča z dvema figurama (sl. 6d), ki sodi k notranjosti stebrnega dela različice A.2; v nadstropje edikule spada tudi odlomek z upodobitvijo ramena osebe.33 Kip oficirja (sl. 5) pripada edikuli bal­dahinskega tipa. Pokopi v edikulah so značilni za premožnejši sloj prebivalstva, pri nas poznamo takšne grobnice iz Šem­petra, Celeje in Petovione.34 Na nekaterih rimskih spomenikih s Koroške smo opravili analizo marmorja. Ugotovili smo, da je večina kamnitih izdelkov narejena iz marmorja, ki izvira iz po­horskih marmolomov Bojtina, Lunežnik in Črešnova,35 le Brančurnikov sarkofag (sl. 10)36 je izdelan iz rdečka­stega marmorja, ki izvira iz marmoloma Kraig severno od Celovca. Sarkofag je po obliki ter okrasu pletenine in akanta, ki sta sicer tipična za edikule, edinstven v širšem prostoru Viruna in Flavije Solve.37 Na širšem območju Prevalj so bili odkriti posa­mični rimski novci, ki domnevno izvirajo iz Leš (vas ali istoimenski potok), na terasi pri Dobjem dvoru pa so nad današnjim poslopjem Mercatorja našli rimski vrček (sl. 1: 6).38 Pri Oblakovih v Dobji vasi so nad gramozno jamo na Tomaževem odkrili ostaline rimskega objekta (sl. 1: 7).39 Obstaja možnost, da tudi nagrobna plošča za ra Jelenko, Visočnik 2006, 354, kat. št. 13); novoodkrita kam­ na iz Meže in Leš (neobjavljeno). 33 Djura Jelenko, Visočnik 2006, 352, kat. št. 9. 34 Kremer 2001. 35 Iz marmorja iz Črešnove je bila izdelana nagrobna plo-šča za Kvinkta iz Poljane (Djura Jelenko, Visočnik 2006, 374, kat. št. 11; Djura Jelenko 2012, 29). 36 CIL III 6522. 37 Djura Jelenko, Visočnik 2006, 375, kat. št. 13; Kumprej Gorjanc 2013, 77, 97. 38 Pahič 1975; Djura Jelenko 2009, sl. 40. 39 Pahič 1967, 28. Sl. 8: Zagrad. Marmorna plinta (neobjavljeno; hrani Jožef Kozlar, Poljana). Fig. 8: Zagrad. Marble plinth (unpublished; stored by Jožef Kozlar, Poljana). 15 known building elements from aediculae;33 four stones belong to the base of aedicula variant A.1 after Kremer (Fig. 6a–c; 9), a slab with two figures from the interior pillar section of variant A.2 (Fig. 6d); a fraction depicting a shoulder also belongs to the upper part of an aedicula.34 A statue of an officer was part of an aedicula of the baldachin type (Fig. 5). Burials in aediculae were typical for the upperclass, and they are known from Šempeter, Celeia, and Poetovio.35 Marble from a number of Roman statues from Carinthia was analysed. The majority of stones are made of marble from Pohorje quarries such as Bojtina, Lunežnik and Črešnova;36 only Brančurnik’s sarcopha­gus (Fig. 10)37 is made of reddish marble which origi­nates from the Kraig marble quarry north of Klagenfurt. The peculiar shape and decoration of interwoven pattern and acanthus, otherwise typical for aediculae, render 33 Djura Jelenko, Visočnik 2006, 393–394, Cat. No. 5 (fig. 4); 396, Cat. No. 13 (fig. 11); Kumprej Gorjanc 2013, 75. Four stones in the Kocka Inn at Poljana (Kozlar), the rest in Pre­valje: up to seven slabs (formerly at Lahovnik’s, now owned by Sušnik); profiled cube (Grošelj; Kumprej Gorjanc 2013, Cat. No. 12); profiled slab (Pezdirc Vehovar; Kumprej 2008, Cat. No. 34); corner buffer (former Paloma factory; Djura Jelenko, Visočnik 2006, 396, Cat. No. 13); newly found stones from the Meža River and Leše (unpublished). 34 Djura Jelenko, Visočnik 2006, 395, Cat. No. 9. 35 Kremer 2001. 36 Tombstone for Quinctus found in Poljana is made of marble from Črešnova (Djura Jelenko, Visočnik 2006, 407, Cat. No. 11; Djura Jelenko 2012, 29). 37 CIL III 6522. Sl. 9: Zagrad. Kvader z upodobitvijo nimfe. Del podnožja edikule različice A.1 (po Kremer 2001) ali del nimfeja. Fig. 9: Zagrad. Cube with a relief of a nymph. Part of the aedicula base variant A.1 after Kremer (2001) or part of a nymphaeum. (Foto / Photo: T. Jeseničnik) the sarcophagus unique in the wider area of Virunum and Flavia Solva.38 In the wider Prevalje region, individual finds of Roman coins originate from Leše (the village or stream of the same name). A small Roman jug (Fig. 1: 6) was found at a terrace near Dobji dvor, above the current Mercator building.39 At the Oblak family’s property in Dobja vas, Roman building remains (Fig. 1: 7) were discovered above a gravel pit at Tomaževo.40 There is even a possibility that the tombstone for Quinctus from Poljana and the fragment of a funerary slab from Mežica originate from Zagrad.41 Stone statues from Zagrad are still being discov­ered. Carinthian Regional Museum documented the ar­chaeological finds during the construction works along the Meža River in 2014 and 2015. On January 17th 2015, during the deepening of the riverbed approximately 250 metres from Brančurnik’s bridge, workers discovered a Roman stone and lifted it onto the bank (Fig. 1: 9; 11). It may have been deposited there by the Meža River down­ 38 Djura Jelenko, Visočnik 2006, 375, 408, Cat. No. 13; Kumprej Gorjanc 2013, 77, 97. 39 Pahič 1975; Djura Jelenko 2009, fig. 40. 40 Pahič 1967, 28. 41 Djura Jelenko, Visočnik 2006, 407. Kvinkta iz Poljane in odlomek nagrobne plošče iz Mežice izvirata iz Zagrada.40 Kamnite spomenike iz Zagrada še vedno odkriva-mo. V letih 2014 in 2015 je Koroški pokrajinski muzej izvedel arheološko dokumentiranje na območju vzdolž reke Meže od obrata Sistemske tehnike do Brančurni­kovega mostu (ob gradbenih delih v zvezi s projektom izboljšanje poplavne varnosti v Dobji vasi). Pri poglab­ljanju struge pribl. 250 m od Brančurnikovega mostu so 17. 1. 2015 delavci odkrili rimski kamen in ga dvignili na brežino (sl. 1: 9; 11). Nizvodno od grobišča v Zagradu ga je ob povišanju voda morda prikotalila reka Meža, morda pa gre za najdbo in situ, saj se je še v drugi po­lovici 18. stoletja nekoliko gorvodno od najdišča Meža razcepila na dva kraka in ustvarila manjši otoček (sl. 4). Marmorni kamen je obdelan, močno zlizan od vode, nekoč je predstavljal del grobne kapelice.41 Masivno kamnito ploščo smo dokumentirali ok­tobra 2016 pri Mesnerjevih na Lešah (sl. 12).42 Ohranili sta se dve luknji za povezovanje, prepoznavna so ležišča za stebričke oziroma noge (?). Morda je pripadala zgorn­jemu delu edikule. Glavnino nenapisnih reliefov iz Zagrada lahko datiramo od sredine 2. do sredine 3. st. To je tudi čas največjega razcveta nagrobne umetnosti v Noriku.43 Gle­de na ohranjene kamnite spomenike lahko domnevamo, da je v Zagradu ležalo rimsko grobišče z grobnicami v obliki edikul in žganimi grobovi. Edina izjema je Bran­čurnikov sarkofag. Čeprav vemo, da so bili kamni odkriti na razdalji pribl. 150 m med Brančurnikovim mostom in začetkom stream of the Zagrad cemetery during a stronger flow, or it could have been found in situ, given that as late as the second half of the 18th century, somewhat upstream from the Meža site, the river split into two branches and created a smaller islet (Fig. 4). The marble stone is worked, badly eroded by water, and once represented a part of a funerary chapel.42 A massive slab was documented in October 2016 at the Mesner family in Leše (Fig. 12).43 Two holes for attaching survived, locks for small pillars or feet (?) are also visible. It may have belonged to the upper part of an aedicula. The bulk of uninscribed reliefs from Zagrad can be dated from the mid-2nd to the mid-3rd century. This century was also a period of the most significant development of funerary art in Noricum.44 According to the surviving stone statues, it can be assumed that a Roman cemetery with aediculae tombs and cremation graves was situated in Zagrad. Brančurnik’s sarcophagus is the only exception. Although the stones were discovered a distance of approx. 150 m (between Brančurnik’s bridge and the woodworking plant), the exact location of the necropolis remains unknown. Test trenching at this spot revealed modern filling and various construction activities that have badly damaged any archaeological traces. Because graves were usually located in the immediate vicinity of villas,45 we can assume that traces of the settlement can be found in the vicinity of Brančurnik’s bridge. 40 Djura Jelenko, Visočnik 2006, 374. 41 Začasno so ga shranili pri družini Klobučar, danes je shranjen v lapidariju kapele na gradu Ravne. 42 Vel. 148 × 72 × 27 cm. Za pomoč pri interpretaciji se zahvaljujem Bojanu Djuriću.43 Pochmarski 1991, 139; Walde 2005, 20; Kumprej 2008, 101. 42 It was temporarily kept by the Klobučar family and is now stored in the chapel lapidarium of Ravne Castle. 43 Size: 148 × 72 × 27 cm. I would like to thank Bojan Djurić for his help with the interpretation.44 Pochmarski 1991, 139; Walde 2005, 20; Kumprej 2008, 101. 45 Horn 2002, 295. lesarskega obrata, natančna lokacija nekropole še vedno ni znana. Sondažne raziskave na tem mestu so razkrile novodobna zasipavanja ter številne gradbene dejavnosti, ki so močno poškodovale arheološke sledove. Grobovi so bili navadno locirani v neposredni bližini vil,44 torej smemo ostanke naselbine iskati v neposredni bližini Brančurnikovega mostu. PODATKI O DRUŽBI Nekateri ohranjeni spomeniki kažejo družbeni status pokojnikov. Edinstven je kip, ki predstavlja centu­riona (sl. 5).45 Ob levem boku nosi meč v nožnici, katere zaključek je značilen za konec 2. in za 3. st.46 Po mnenju A. Schoberja orožje opredeljuje pokojnika kot aktivnega vojaka.47 Portretiranec je upodobljen brez centurionske palice (vitis), ki naj bi opredeljevala njegov čin. Zaradi podobnih resastih pramenov ter oklepa s poudarjenimi mišicami je primerljiv z vojakom iz Lorcha v bližini Ennsa (Lauriacum), ki je datiran v severski čas.48 Nagrobnik pisarja (sl. 6a) poudarja višji sloj.49 Nogo drži na posodi za akte (capso), oblečen je v kratko tuniko, ki spominja na srajco. To oblačilo je značilno predvsem za upodobitve služabnikov darovalcev in pisarjev v No-riku. Jantsch meni, da se takšna tunika nosi v povezavi s tipično domačimi kosi oblačil.50 Služabniki pisarji se lahko pojavljajo na stranicah ar, kot je značilno za Flavijo Solvo,51 pa tudi na reliefu edikule različice A.1, tip 2, kot pri primeru iz Zagrada.52 Relief žalujočega Atisa in Erota se pojavi v sestav­ljeni kompoziciji (sl. 6b). Erot v levici drži okroglo po­krito cilindrično posodo. Waldejeva meni, da tak Erot nadomešča mesto služabnic, ki v rokah držijo skrinjico z nakitom za gospodarico.53 Njegova sporočilnost je krogotok časa in s tem večno vračanje v novo življenje in cvetenje po fazi dozorenja in smrti.54 Erot je upodobljen tudi na odlomku reliefne plošče, na kateri sta se ohranili zgolj lebdeči stopali (sl. 6c). Oba Erota sta stala v podnožju edikule različice A.1. Tudi moški na reliefni plošči z dvema figurama (sl. 6d) je upodobljen v gosposki pozi, oblečen v togo z gubo v obliki črke U (variante b). Na njegovi desni strani sedi 44 Horn 2002, 295. 45 Odkrit je bil jeseni leta 1903 ob visokih vodah reke Meže (Djura Jelenko, Visočnik 2006, 352; Kumprej Gorjanc 2013, 78). 46 Kumprej Gorjanc 2013, 80. 47 Schober 1923, 177. 48 CSIR Österreich III/2, Nr. 23; Kumprej 2008, 81, 90. 49 V literaturi je prvič omenjen leta 1922 (Djura Jelenko, Visočnik 2006, 350, kat. št. 7). 50 Jantsch 1934, 70; Kumprej 2008, 83. 51 Pochmarski, Hainzmann 2004, 21. 52 Kumprej 2008, 99. 53 Walde 2005, 120; Kumprej 2008, 50. 54 Walde 2005, 143; Kumprej 2008, 97. SOCIETY Some statues indicate the social status of the de­ceased. The statue of a centurion (Fig. 5) is unique.46 At his left side, he has a sword in a scabbard, the tip of which is characteristic of the end of the 2nd and the 3rd century.47 According to Schober, weapons define the deceased as an active soldier.48 The portrayed is depicted without a staff of office (vitis), which would define his rank. Due to similar coarse locks and a muscle cuirass, he is comparable to the soldier from Lorch near Enns (Lauriacum), which is dated to the Severan dynasty period.49 The tombstone with a motif of a scribe (Fig. 6a) also originates from a monumental tomb.50 His foot is positioned on a cist (capsa), he is wearing a short tunic resembling a shirt. This attire was particularly charac­teristic of depictions of servants and scribes in Noricum. Jantsch believes that such tunics are typical domestic clothing.51 Servant scribes may appear on sides of al­tars, as is typical for Flavia Solva,52 as well as on reliefs of aedicula variant A. 1, type 2, as in the Zagrad case.53 Grieving Attis and Eros appear on the same relief (Fig. 6b). Eros is holding a round, cylindrical pot with a lid in his left hand. In Walde’s opinion, such a depiction of Eros replaces the servants’ position, who usually ap­pear holding their master’s jewellery box.54 Its message is the cyclical nature of time; one’s eternal return to new lives and the prospect of flourishing after periods of maturation and death.55 Only two floating feet of Eros survived on a frag­mented slab (Fig. 6c). Both Erotes were standing at the base of aedicula variant A.1. The relief slab with two figures (Fig. 6d) represents a man in a noble posture, wearing a toga with a U-shaped sinus (variant b) and the grieving woman sitting at his right side. The slab once stood in the interior with columns of aedicula variant A.2.56 Similar relief slabs with whole figures are not exactly common among the Noricum material; a similarly depicted pair of deceased individuals can be found in St. Veit an der Glan.57 46 It was discovered in fall 1903 during high waters of the Meža River (Djura Jelenko, Visočnik 2006, 395; Kumprej Gorjanc 2013, 78). 47 Kumprej Gorjanc 2013, 80. 48 Schober 1923, 177. 49 CSIR Österreich III/2, No. 23; Kumprej 2008, 81, 90. 50 It was first mentioned in literature in 1922 (Djura Jelenko, Visočnik 2006, 394, Cat. No. 7). 51 Jantsch 1934, 70; Kumprej 2008, 83. 52 Pochmarski, Hainzmann 2004, 21. 53 Kumprej 2008, 99. 54 Walde 2005, 120; Kumprej 2008, 50. 55 Walde 2005, 143; Kumprej 2008, 97. 56 Kremer 2001, 321; Djura Jelenko, Visočnik 2006, 394, Cat. No. 6. 57 CSIR Österrech II/2, 44, Nr. 163, Taf. 39; Kumprej 2008, 24. žalujoča ženska. Plošča je nekoč stala v notranjosti ste­brnega dela edikule različice A.2.55 Primerljive reliefne plošče s celimi figurami niso ravno pogoste v noriškem materialu, podobno upodobljen par pokojnikov najde-mo v Gradesu (St. Veit an der Glan).56 NASELJE Na Tabuli Peuntigeriani položaj antičnega Zagrada ni posebej označen. Leži tik ob trasi ceste, ki je povezovala južnonoriški mesti – Celejo in Virunum, na sredini med Kolaciono (Stari trg pri Slovenj Gradcu) in Jueno, od vsake odda­ljena 11 rimskih milj (16,3 km), kar ustreza razdalji, na kateri so bile lahko vzpostavljene obcestne postaje (mansio, mutatio).57 Zagrad je zanimiv predvsem kot mejna postaja med administrativnima ozemljema Celeje in Viruna.58 Razkošni kamniti spomeniki – deli edikul in sarkofag – bi lahko pripadali grobišču podeželske elite iz bližnje vile oziroma posestva ali pa grobišču neke manjše obcestne naselbine. Rimska podeželska grobišča naj bi od naselbin ležala v oddaljenosti med 100 in 400 m59 Ostanke podobnih družinskih grobnic poznamonpr. iz naselbin Šempeter ali Santikum (Villach/Beljak), postavljene so bile znotraj obzidanih grobnih parcel. Takšne družinske grobnice so značilne tudi za bližnje antično grobišče v Kolacioni.60 Rimski spomeniki iz Zagrada po svoji razkošno­sti kažejo bogato podeželsko aristokracijo, ki je med zgodnjeantoninskim in srednjeseverskim obdobjem ali bivala v vili rustiki, ki je morda prevzela funkcijo mansia, ali pa v manjši naselbini. Več argumentov govori v prid pomembnejšega naselja in ne samo vile: število kakovostnih nagrobnih spomenikov, arhitektonski členi stavb, templja, ki niso nagrobni spomeniki, pa tudi sama lega v neposredni bližini ceste na mejnem območju med dvema mestnima teritorijema (Celeia in Virunum) in natanko v sredini med Kolaciono in Jueno. Domnevna rimska postaja v Zagradu je bila morda povezana tudi z oddihom v “toplicah”. Ledinska imena kažejo izvire tople vode. 55 Kremer 2001, 321; Djura Jelenko, Visočnik 2006, 349, kat. št. 6. CSIR Österrech II/2, 44, Nr. 163, Taf. 39; Kumprej 2008, 24. 57 Tab. Peut. IV, 2. 58 Kumprej 2008, 16. 59 Kumprej Gorjanc 2008, 95. 60 Djura Jelenko 2004, 33; 2012, 43. SETTLEMENT On Tabula Peutingeriana, the location of Roman settlement at Zagrad is not specifically marked. It lies right off the route of via publica that con­nected two towns of southern Noricum—Celeia and Virunum, positioned in the middle of Colatio (Stari trg pri Slovenj Gradcu) and Iuenna, namely 11 Roman miles (16.3 km) from each town, which could correspond to the distance at which road stations (mansio, mutatio) were built.58 Zagrad is particularly interesting as a border station between the administrative territories of Celeia and Virunum.59 The stone statues—parts of aedicu-lae—and the sarcophagus could belong to a cemetery of the rural elite from a nearby villa or property, or to a cemetery of a small settlement near the public road. Rural Roman cemeteries were presumably positioned between 100 and 400 m from the settlements.60 Remains of similar family tombs are known, e.g., from Šempeter or Santicum (Villach) settlements, they were positioned inside of walled funerary plots. Such family tombs are also typical of the nearby Roman cemetery in Colatio.61 Roman monuments from Zagrad, in their luxury, point to a rich rural aristocracy, which resided either in the villa rustica that assumed the position of mansio or in a smaller settlement between the early Antonine and mid-Severan period. A number of arguments point to a relatively significant settlement rather than mere villa: the quantity of quality funerary monuments, the architectural elements of the various buildings (possibly also a temple), the location in the immediate vicinity of the road in a border area of two municipalities (Celeia and Virunum) at the exact midpoint between Colatio and Iuenna. As the toponyms point to hot water springs, the location of the Roman thermal baths is also possible. Translation: Lucija Jelenko 58 Tab. Peut. IV 2. 59 Kumprej 2008, 16. 60 Kumprej Gorjanc 2008, 95. 61 Djura Jelenko 2004, 33; 2012, 43. Okrajšave / Abbreviations CIL = Corpus inscriptiorum Latinarum. CSIR = Corpus signorum imperii romani – Corpus der Skulp­turen der römischen Welt. lupa = UBI ERAT LUPA, F. und O. Harl, http://lupa.at/ (Bild­datenbank zu antiken Steindenkmälern). DIEZ, E. 1980, Quellnymphen. – V / In: F. Krinzinger, B. Otto, E. Walde-Psenner (ur. / eds.), Forschungen und Funde, Festschrift Bernhard Neutsch, 103–108, Innsbruck. DJURA JELENKO, S. 2004, Dr. Hans Winkler in njegov prispe­vek k arheologiji Mislinjske doline (Dr Hans Winkler und sein Beitrag zur Archäologie des Mislinjatals). – Slovenj Gradec. DJURA JELENKO, S. 2009, Arheologija Koroške krajine. Stalna razstava. Kolaciona. MCMIX-MMIX. 1909-2009. – Slovenj Gradec. DJURA JELENKO, S. 2012, Dis Manibus Sacrum. Božanskim manom (Den göttlichen Mannas / To the divine Manes). – Slovenj Gradec. DJURA JELENKO, S., J. VISOČNIK 2006, Rimski kamniti spo­meniki slovenske Koroške / The Roman stone monuments of Slovenian Carinthia. – Arheološki vestnik 57, 345–415. HORJAK, M. 2012, Mitska podoba Uršlje gore in okolice. – Diplomsko delo / BA Thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). HORN, H. G. 2002, Die Römer in Nordrhein-Westfalen. – Stuttgart. JABORNEGG-ALTENFELS, M. F. 1870, Kärntens römische Alterhümer. – Klagenfurt. JANTSCH, F. 1934, Norische Trachtendarstellungen in Kärn-ten. – V / In: Carinthia I, 2, 65–74. KASTELIC, J. 1998, Simbolika mitov na rimskih nagrobnih spo­menikih. Šempeter v Savinjski dolini (Sepulchral symbolism of the mythological imagery on Roman tomb monuments. Šempeter in the Valley of Savinja). – Ljubljana. KÄPPEL, L. 2000, Nymphen. – V / In: H. Cancik, H. Schnei­der (ur. / eds), Der neue Pauly. Enzyklopädie der Antike 8, 1071–1072, Stuttgart, Weimar. KORDEŽ, I. 2003, Doživetja in spomini. – Slovenske večernice 153, Celje. KREMER, G. 2001, Antike Grabbauten in Noricum. – Sonder­schriften des Österreichischen Archäologischen Instituts 36. KUMPREJ, M. 2008, Iuenna – nagrobni spomeniki in njihovo mesto v noriški funerarni umetnosti. – Diplomsko delo / BA Thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). KUMPREJ GORJANC, M. 2013, Kip centuriona iz rimskega grobišča Zagrad pri Prevaljah. – Studia Universitatis Here-ditati 1/1–2, 69–85. PAHIČ, S. 1975, Prevalje. –V / In: Arheološka najdišča Slove­nije, 275, Ljubljana. PAHIČ, S. 1967, Naši kraji pod Rimljani. – Koroški Fužinar 17/3 22–29. POCHMARSKI, E. 1991, Zur Datierung von Solvenser Por­träts. – In / V: Akten des 1. Internationalen Kolloquiums über Probleme des provinzialrömischen Kunstschaffens,Graz 27.–30.4.1989, Mitteilungen der Archäologischen Gesell­schaft Steiermark 5, 99–109. POCHMARSKI, E., M. HEINZMANN 2004, Steine erzählen. Römische Steindenkmäler auf Schloss Seggau bei Leibnitz. – Graz. RAJŠP, V. (ur. / ed.) 2000, Slovenija na vojaškem zemljevidu 1763–1787 / Josephinische Landesaufnahme 1763-1787 für das Gebeit des Republik Slowenien. – Ljubljana. SCHOBER, A. 1923, Die römischen Grabsteine von Noricum und Pannonien. – Sonderschriften des Österreichischen Archäologischen Institutes 10, Wien. ŠAŠEL, J. 1953, Zagrad pri Prevaljah – rimsko grobišče. – Kronika 1, 15–23. ŠINKOVEC, I. 1995, Katalog posameznih kovinskih najdb bakrene in bronaste dobe / Catalogue of individual metal finds from the Eneolithic and Bronze Ages. – V / In: B. Teržan (ur. / ed.), Depojske in posamezne kovinske najdbe bakrene in bronaste dobe na Slovenskem / Hoards and in­dividual metal finds from the Eneolithic and Bronze Ages in Slovenia 1, Katalogi in monografije 29, 29–127. TRUHLAR, F. 1975, Krajevna imena Gradišče, Gomila, Gro­blje, Žale. – V / In: Arheološka najdišča Slovenije, 106–112, Ljubljana. WALDE, E. 2005, Im Herrlichen Glanze Roms. Die Bilderwelt der Römersteine in Österreich. – Innsbruck. Saša Djura Jelenko Koroški pokrajinski muzej Glavni trg 24 2380-SI Slovenj Gradec sasa.djura.jelenko@kpm.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 339–348 SLOVENSKA BISTRICA Mira STRMČNIK GULIČ Izvleček Rimska naselbina v Slovenski Bistrici leži ob cesti Celeja–Petoviona. Zgodnjim lesenim stavbam, katerih tlorisi niso jasni in so datirane v 1.–2. st., sledijo v 3.–4. st. velike zidane stavbe, morda povezane s transportom in skladiščenjem. Odkriti so bili sledovi kovaške delavnice. Ob naselbini je ležalo grobišče z žganimi pokopi. Ključne besede: Zgornja Panonija, Slovenska Bistrica, rimska doba, naselbina, grobišče, kovaška delavnica Abstract The Roman settlement in Slovenska Bistrica is located on the road connecting Celeia and Poetovio. Excavations re­vealed it had wooden buildings in the 1st-2nd centuries, the ground plans of which could not be fully investigated. These were followed in the 3rd–4th centuries by large masonry buildings, presumably associated with transport and storage of goods. Remains of a smithy were found in the settlement, in its vicinity also a cemetery with cremation burials. Keywords: Upper Pannonia, Slovenska Bistrica, Roman period, settlement, cemetery, smithy https://doi.org/10.3986/9789610502586_18 ZGODOVINA RAZISKAV Stanko Pahič je leta 1982 podal prvi sistematični arheološki pregled na bistriškem območju in predstavil vse dotlej znano gradivo. Prispevek bogatijo in vsebin­sko dopolnjujejo pregledne tabele gradiva, kronološke opredelitve, načrti najdišč in arheološki zemljevid pokrajine ob jugovzhodnem Pohorju.1 Prav posebej se je posvetil vrednotenju številnih rimskih marmornih spomenikov kot enega najpomembnejših arheoloških virov teh krajev. Njihovo pogostnost lahko deloma pri­pisujemo bližini pohorskih kamnolomov, o čemer nam pričajo še danes razpoznavni in zapuščeni kamnolomi vokolici Šmartnega na Pohorju.2 S poglavitnimi podatki in navajanjem napisov je Pahič poskušal razvozlati izvor in družbeni položaj pokojnikov. Pri tem je poudaril, da se je ob pristnih italskih imenih obdržalo precej imen keltskega porekla in latiniziranih oblik.3 Na širšem bistriškem območju so se arheološke raziskave razmahnile v sedemdesetih letih prejšnjega stoletja s projektom raziskovanja odseka rimske ceste Celeja–Petoviona, ki je prečkala bistriški svet od jugo­zahoda proti severovzhodu. Ugotovitve so prinesle ključ­na spoznanja o poteku in izgradnji cest ter prispevale k razumevanju obcestnih postaj oz. naselbin.4 Ob cesti so odkrili zaporedje naselij, ki so nastala v povezavi z državno cestno službo. Čeprav nobeno izmed najdišč ni bilo zadovoljivo raziskano, pa lega in najdbe naselbin­skega značaja lahko govorijo za preprežne postaje pri Čadramu (ledina Belo mesto), v Spodnjem Grušovju (mansio Ragando?) in pri Spodnji Novi vasi.5 NASELBINA Na ožjem območju današnje Slovenske Bistrice je bila odkrita in delno raziskana trasa rimske ceste, dva sklopa rimskih stavb, ki sta drug od drugega oddaljena okoli 500 m, in del grobišča (sl. 1). Stavba A (sl. 1: 1; 2) Stavba A je stala na zahodnem delu rimske na­selbine (sl. 1: 1), na nižji vzpetini, ob vzhodnem robu naravne terase. Jugovzhodni rob vzpetine je prečkala rimska cesta, ki je bila prislonjena ob pobočje. Ostanki stavbe A so bili prvič zaznani leta 1957.6 Traso rimske ceste v bližini so ugotovili leta 1971. Z večjim arheološkim posegom leta 1974 so odkrili sko­ 1 Pahič 1982. 2 Pahič 1982, 63–73; Djurić, Müller 2009. 3 Pahič 1982, s poglavitno literaturo in epigrafskimi viri. 4 Pahič 1969; Pahič 1978. 5 Pahič 1969. 6 Pahič 1960. HISTORY OF RESEARCH The first systematic archaeological survey of the Slovenska Bistrica area was published in 1982. In this publication, Stanko Pahič presented the small finds known up to that time, their chronological attribution, plans of individual sites and an archaeological map of the area along the south-eastern part of the Pohorje Hills.1 Of the movable artefacts, he paid particular attention to the Roman marble monuments as one of the most important archaeological sources from the area. Their high number is believed in part to be the result of the proximity of marble quarries in the Pohorje Hills, traces of which arestill visible around Šmartno na Pohorju.2 Pahič studied the monuments and their inscriptions to assess the origin and social status of the deceased, observing that Italic names appeared alongside numerous names of Celtic origin and Latinised forms.3 The Bistrica area was more intensively investigated in the 1970s, as part of the project of researching the Roman road from Celeia to Poetovio that crossed the area from southwest to northeast. The investigation results significantly advanced the knowledge of the course and construction of the Roman road, as well as of the roadside stations or settlements.4 A succession of settlements was detected lining the road, associated with the state road services. None have been investigated in detail, but the location and recovered habitation finds do allow us to posit relay stations at Čadram (the Belo mesto site), Spodnje Grušovje (mansio Ragando?) and Spodnja Nova vas.5 SETTLEMENT Investigations in the area of the present-day town of Slovenska Bistrica unearthed sections of the Roman road, two groups of Roman buildings located at a distance of some 500 m from one another, as well as part of the associated cemetery (Fig. 1). Building A (Fig. 1: 1; 2) It stood in the west part of the Roman settlement, on a low elevation at the eastern edge of a natural terrace (Fig. 1: 1). The Roman road passed not far from the building, along the south-eastern edge of this terrace. 1 Pahič 1982. 2 Pahič 1982, 63–73; Djurić, Müller 2009. 3 Pahič 1982, with main bibliography and epigraphic sources. 4 Pahič 1969; Pahič 1978. 5 Pahič 1969. Sl. 1: Slovenska Bistrica. Raziskana območja rimske naselbine (1–2), grobišče (3) in potek ceste. Fig. 1: Slovenska Bistrica. Investigated areas of the Roman settlement (1–2), cemetery and road. (Vir / Source: http://giskd6s.situla.org/evrd/) raj celotni tloris stavbe.7 Izkopi so zajeli dobro tretjino površin v stavbi in okrog nje, vendar na različnih koncih in v vseh prostorih. S sondažnimi izkopi ob jugozahodni fasadi so poskušali ugotoviti razmerje med stavbo in cesto. Območje stavbe in njene okolice je bilo sondirano tudi leta 2004.8 Temelji stavbe A so ležali 0,30–0,50 m pod povr­šino. Vkopani so bili v ilovico, na katero so nasuli plast gramoza in nanj položili kamne zidu. Ohranjena višina temeljev je bila 0,30–0,40 m in širina 0,80–0,90 m. Na temeljih se je ohranila tanka plast ilovice, na katero so postavili kamnit zid, vezan z malto. Z navzven razširje­nimi vogali so poskrbeli še za dodatno trdnost stavbe. Temelji notranjih predelnih zidov, v širini 0,50–0,65 m, so bili postavljeni neposredno na ilovico. Stavba je pravokotnega tlorisa, velika okoli 20 × 31 m. Vzdolž daljših stranic so bili razporejeni po trije oziroma štirje prostori (A–D, H–F), v sredini je ležalo odprto dvorišče (I). Z gramozom posuta površina vodi od ceste proti južnemu vogalu stavbe in nakazuje lego vhoda (sl. 2). Po Pahičevem mnenju je bil 20 × 4 m velik prostor G hlev za vprežno živino, drugi prostori okrog dvorišča pa bi lahko bili skladiščni in bivalni. Tla v prostoru A 7 Jerman 1976; 1977; Pahič 1976, 246, sl. 8; 1978, 138– 140. 8 Strmčnik Gulič, Kajzer Cafnik 2004. Remains of Building A were first recorded in 1957,6 while the road was documented in 1971. A more extensive archaeological investigation in 1974 unearthed over one third of the building interior and the area around it, but in different places and in all the rooms, revealing the layout almost in its entirety.7 The trial trenching along the south­west façade was aimed at establishing the relationship between the building and the adjacent road. The building and its surroundings were again trial trenched in 2004.8 The foundations of Building A were found 0.30–0.50 m under the surface. The foundation trench was dug into loamy soil, a layer of gravel was then deposited on the bottom of the trench and the foundations constructed on top. The foundations survived to the height of 0.30–0.40 m and width of 0.80–0.90 m. A thin layer of loam survived on top of the foundations, onto which a mortar-bound stone wall was built. The construction was strengthened by widening the walls in the corners. The 0.50–0.65 m wide foundations of the interior partition walls were laid directly onto the loamy soil. The building was rectangular in plan and measured roughly 20 × 31 m. Three and four rooms (A–D, H–F) were arranged along its respective longer sides, with an 6 Pahič 1960. 7 Jerman 1976; 1977; Pahič 1976, 246, Fig. 8; 1978, 138– 140. 8 Strmčnik Gulič, Kajzer Cafnik 2004. Sl. 2: Slovenska Bistrica. Stavba A in rimska cesta s plastnicami antične površine (po Pahič 1977, 245, sl. 8). Fig. 2: Slovenska Bistrica. Building A and road with the contour lines of the Roman-period ground (from Pahič 1977, 245, Fig. 8). so bila pomešana z drobci žganine, kamenja in ostanki ilovnatega ometa, z oblicami pa so bili tlakovani prostori B, H in G. V prostoru F najdena žganina na steptanih ilovnatih tleh je verjetno ostanek lesenega poda. Tu je bilo v ruševinski plasti odkrite precej razbite opeke, zlasti tegul. V stavbi je bilo skromno število keramičnih in drugih najdb, pa še te le v severni in vzhodni polovici poslopja. Gre za domačo kuhinjsko lončenino, ki jo v glavnem datiramo v 4. st. Datacija je podkrepljena z novcema cesarjev Konstantina in Valensa.9 13 m jugovzhodno od stavbe A poteka cestna trasa in ob njej plitev obcestni jarek.10 Od glavne ceste se je 9 Pahič 1978, 138–140, t. 2: 1–8. 10 Jerman 1977. open courtyard along the centre (I). Leading from the road towards the south corner of the building was a gravel path that indicates the location of the entrance to the building (Fig. 2). According to Pahič, the 20 × 4 m large Room G was a stable for draught animals, while other rooms around the courtyard were intended for either storage or habitation. The floor in Room A was a mixture of earth, bits of burnt remains, stones and patches of loam wall plaster. Rooms B, H and G had a cobbled floor. Room F had a floor of beaten loam with burnt remains that were presumably the remains of wooden floor boards; the debris layer in this room also revealed a fair amount of broken bricks, particularly tegulae. proti severu odcepila še ena cesta in vodila vzhodno od stavbe A. Severovzhodno od stavbe A so bili odkriti tudi sledovi kovaške dejavnosti.11 Stavbe B–E (sl. 1: 2; 3–5) Območje leži ob rimski cesti, na najnižjem pros-toru Bistriškega polja, pod rečno teraso blizu potoka Bistrice, ki je vseskozi poplavljal in nanašal debele plasti naplavin.12 V več raziskovalnih etapah je bilo odkritih pet stavb (sl. 1: 2). Ob gradbenih delih so našli tudi posamezne zelo poškodovane temelje in kamnite tlake, ki pričajo o obstoju še drugih stavb.13 Po predhodnih sondažnih izkopih so bili v letih 1973-1974 odkriti kamniti temelji stavbe B (sl. 3; 4), kvadratnega tlorisa v velikosti 19,8 × 19,8 m.14 Stala je 6,5 m južno od rimske ceste. Vhoda se ni dalo zanes­ljivo prepoznati. Razčlenjena je bila na notranje dvo­rišče (prostor L), ki ga je obkrožalo 11 različno velikih prostorov (A–K). Drobne najdbe kažejo začetek v prvi polovici 2. st. in življenje vsaj do konca 3. st.15 Južno od rimske ceste sta bili evidentirani še dve skupini temeljev in drobnih najdb, opredeljenih kot stavbi C in Č (sl. 3).16 Pri občasnih arheoloških nadzorih so prišle na dan posamezne raztresene najdbe, tako da je bilo na tem območju po vsej verjetnosti še več stavb, ki pa so jih novodobni posegi poškodovali ali celo uničili. Severno od rimske ceste sta bili leta 1984 odkriti stavbi D in E (sl. 3). 17 Na območju stavbe D je bilo najprej leseno poslopje, ki ga nakazujejo ostanki jam. Zidana stavba D je bila ve­lika 26,5 ×13,50 m (sl. 3). Kamniti temelji so bili grajeni iz večjih rečnih prodnikov in lomljencev z vmesnim drobnim peskom, brez sledov malte. Večinoma so bili ohranjeni še v dveh nizih do 35 cm visoko. Vhod je bil na vzhodni strani.18 Stavba je bila po dolgem predeljena z dvema vrstama kamnitih temeljev za lesene stebre, ki so podpirali strešno konstrukcijo. V severovzhodni vrsti so bili ohranjeni trije temelji, v jugozahodni pa samo eden. Ostali temelji so se videli samo še kot temne lise. Razdalja med stebri v vrstah je bila po 3,2 m. Vrsti sta bili oddaljeni okoli 4,5 m od sten. V stavbi je bilo skupaj odkritih 42 različno globokih jam (globine od 0,20 do 1,70 m) z ostanki ožganih sten, ožganega lesa in kame­ 11 Strmčnik Gulič, Kajzer Cafnik 2004. 12 Pahič 1978, 131, karta 1. 13 Pahič 1978, 140–148; Mikl-Curk 1976, 23–27; Strmč­nik Gulič, Pahič 1985, 257. 14 Pahič 1978, 145, sl. 5. 15 Pahič 1978, 147. 16 Mikl-Curk 1976, 23–26. 17 Strmčnik Gulič, Pahič 1985; Strmčnik Gulič 1990. 18 Kljub močno razmočenemu terenu in visoki talni vodi je bilo mogoče odkriti tri zunanje stene objekta z vogali, četr-ti – severni –, pa je ležal dva metra pod nivojem in bil v celoti poplavljen. The building yielded rare pottery and other finds, and only in the north and east parts. The pottery belongs to coarseware, mainly attributable to the 4th century. This dating is corroborated by two coins of Constantine and Valens, respectively.9 Building A was located 13 m northwest of the main road that was lined with a shallow roadside ditch.10 A road forked off the main one east of Building A to lead northwards. Northeast of Building A, archaeologists found traces of smithing activities.11 Buildings B–E (Fig. 1: 2; 3–5) This group of Roman buildings also lies along the Roman road, in the lowest part of the plain of Bistriško polje and below the river terrace along the Bistrica stream that is prone to flooding and deposited thick alluvial layers (Fig. 1: 2).12 The area has been investi­gated in several campaigns and revealed five buildings. Recent construction activities in the area also revealed individual sections of heavily damaged foundations and stone floors/pavings that indicate the existence of additional buildings.13 The trial trenching in 1973-1974 unearthed the masonry foundations of Building B (Fig. 3; 4), of a square plan measuring 19.8 × 19.8 m.14 It stood 6.5 m south of the Roman road. Its entrance could not be positively identi­fied. It had a central courtyard (Room L) surrounded by eleven variously large rooms (A–K). The recovered small finds suggest the building was in use from the first half of the 2nd century to at least the late 3rd century.15 Two more groups of foundations and small finds were recorded south of the Roman road, interpreted as Buildings C and Č (Fig. 3).16 The watching briefs in this area revealed stray finds, suggesting the existence of other buildings damaged or even destroyed during modern earthwork or construction activities. In 1984, Buildings D and E were found north of the Roman road (Fig. 3). 17 Postholes indicate that a wooden building preceded the masonry Building D, the latter measuring 26.5 × 13.50 m (Fig. 3). Its foundations were constructed of large cobbles and rubble mixed with fine sand, without the use of mortar and mostly surviving in two courses up to 35 cm high. The entrance was in the east.18 The building was 9 Pahič 1978, 138–140, Pl. 2: 1–8. 10 Jerman 1977. 11 Strmčnik Gulič, Kajzer Cafnik 2004. 12 Pahič 1978, 131, Map 1. 13 Pahič 1978, 140–148; Mikl-Curk 1976, 23–27; Strmč­ nik Gulič, Pahič 1985, 257. 14 Pahič 1978, 145, Fig. 5. 15 Pahič 1978, 147. 16 Mikl-Curk 1976, 23–26. 17 Strmčnik Gulič, Pahič 1985; Strmčnik Gulič 1990. 18 In spite of the rain-sodden conditions with high Sl. 3: Slovenska Bistrica. Stavbe B, D in E ter lega stavb C in Č (dopolnjeno po Strmčnik Gulič, Pahič 1985, 254). Fig. 3: Slovenska Bistrica. Buildings B, D and E, as well as the location of Buildings C and Č (supplemented after Strmčnik Gulič, Pahič 1985, 254). njem, v nekaterih so bili tudi odlomki lončenih posod. Poslopje je bilo po vsej verjetnosti uničeno v požaru, nadgradnja se je porušila navzven, kjer je ob zidovju ležala do 30 cm debela ruševinska plast z žganino, kosi oglja in keramiko. Tudi zunaj objekta so bili ostanki lesenih stebrov in kolov, ki govorijo za lesene prizidke ali morda za samostojne lesene gradnje. Na zunanji strani severovzhodnega zidu je bilo odkrito ovalno kurišče. Keramične najdbe so bile redke (največ jih je bilo v žganini zunaj severovzhodnega zidu), zaradi česar sklepamo na gospodarsko funkcijo objekta. Prevladujejo ostanki grobe hišne lončenine, katere začetek sodi v 2. st., glavnina v 3. st., nadaljuje pa se še v 4. st. Najden je bil tudi novec Konstancija Gala (352–354). Tudi na območju stavbe E je najprej stalo leseno poslopje, od katerega so se ohranile jame za stebre ter 17 m dolg in do 0,35 m globok jarek (sl. 3; 5). Zidana stavba E je bila velika 35,20 × 19,65 m. Imela je osrednje dvorišče, ki je v širini 8 m potekalo po celotni dolžini stavbe in ob katerem so bili razporejeni različno veliki prostori. V naslednji fazi so vsaj del prostorov opustili. V notranjosti so postavili dve vrsti kamnitih temeljev za stebre, v medsebojni razdalji 3,5 m. Na jugozahodni strani se je ohranila linija devetih temeljev stebrov, od katerih sta dva stala na predelnih stenah starejših pros-torov. Na severovzhodni strani stavbe so bili ohranjeni longitudinally divided with two rows of masonry plinths for wooden columns or posts that supported the roof construction. Three plinths survived of the northeast row and only one of the southwest row, while all others were detectable as dark patches. The columns were positioned at a distance of 3.2 m from one another and the rows stood roughly 4.5 m from the walls. The building also revealed a total of 42 pits of different depths (0.20 to 1.70 m) that contained the remains of burned walls, burnt wood and stones, some also pottery sherds. The building was most probably destroyed in a fire with the superstructure col­lapsing outward, as suggested by an up to 30 cm thick debris layer with burnt remains, pieces of charcoal and pottery alongside the exterior walls. Traces of wooden posts were also found outside the building, indicating the existence of wooden buildings or structures, either as extensions or as independent units. An oval fireplace was unearthed on the exterior side of the northeast wall. Pottery finds were rare (most recovered among the burnt remains outside the northeast wall), which indicates that the building was not used as living quarters. Most sherds are coarseware; they begin in the 2nd century, the groundwater levels, archaeologists were able to identify three exterior walls of the building and three corners, while the fourth one in the north lay two metres below the ground and was completely submerged. Sl. 4: Slovenska Bistrica. Stavba B (po Pahič 1978, 145, sl. 5). Fig. 4: Slovenska Bistrica. Building B (from Pahič 1978, 145, Fig. 5). Sl. 5: Slovenska Bistrica. Stavba E, pogled proti severozahodu, izkopavanja 1983 Fig. 5: Slovenska Bistrica. Building E, view towards nord west. Excavations in 1983. (Foto / Photo: arhiv / archives ZVKDS OE Maribor) trije temelji stebrov. Ob jugovzhodnem pročelju je ležal obsežen predprostor, od katerega se je ohranil le seve­rovzhodni zid. V primerjavi s stavbo D je bilo v stavbi E več opeke in keramike. Lesena faza stavbe je datirana v 1.–2. st., zidana stavba s prezidavami pa v 3.–4. st. GROBIŠČE Na položnem grebenu na Brinju, severozahodno od naselbine, je bilo med arheološkim nadzorom odkritih 26 žganih grobov, a grobišče ni bilo v celoti raziskano (sl. 1: 3).19 Grobovi so skromni, pri nekaterih je bila v grobno jamo položena glinena žara, pri drugih pa je bil pepel prosto nasut v grobno jamo. Večinoma so bili brez pridatkov, razen nekaj kosov keramike; tudi ostankov nagrobnikov ni bilo. Apnena peč, datirana v antiko, je bila zgrajena na prostoru grobišča in ga je deloma uničila. Bila je nepra­vilno pravokotno oblikovana z merami 6 × 2,50 m in 0,50 m globoka, na dnu so bili ožgani kamni, na robovih debela plast ožgane gline in oglja. V peči so lahko žgali bližnje marmorne kamne, bodisi nagrobne spomenike bodisi nanose potokov Bistrice in Devine. RIMSKA CESTA Na bistriškem območju je bila rimska cesta pod-robno raziskovana (sl. 1; 6). Profili so pokazali prerez cestišča s 6,6 m širokim gramoznim nasutjem in ob-cestnimi jarki ob straneh. Ponekod leži cestišče tik pod površino, drugod so ga prekrile meter ali več debele naplavine v dolinah pohorskih potokov. Na Bistriškem polju je bila podlaga ceste večinoma narejena iz različ­no velikih prodnikov, površina pa še dodatno utrjena z drobnejšim prodom. Gramoz je bil lahko nasut na nekdanja tla, izjemoma na podlago iz gline. Ohranil se je v debelini okrog 15 cm, redko do 40 cm, na nekaterih odsekih tudi več.20 Ponekod so bile ugotovljene obnove cestišča in kolesnice ter različno globoki obcestni jarki na eni ali obeh straneh. Na sistematično raziskanem odseku (pri gradnji trgovine Spar) v dolžini skoraj 60 m je največja ohranjena širina vključno s spranim peskom s cestišča v tlorisu znašala 12,64 m, najmanjša pa 6,04 m. V najbolje ohranjenem profilu je širina ceste znašala 7,1 m, debelina nasutja pa 0,31 m.21 Novejši podatki o izgradnji cestišča so znani tudi iz jugozahodnega dela Slovenske Bistrice, kjer je bila prav tako zgrajena iz zbitega gramoza, prodnikov, zbi­tega drobnega peska in ilovice. Nasutje je bilo debelo 19 Strmčnik Gulič et al. 2008; Pahič 1978, 159 ss. 20 Pahič 1983, 267–272. 21 Kajzer Cafnik, Predan 2005. majority is attributable to the 3rd and some to the 4th century. Also found was a coin of Constantius Gallus (352–354). A wooden building also preceded Building E, surviving as postholes, as well as a 17 m long and up to 0.35 m deep ditch (Figs. 3; 5). Masonry Building E measured 35.20 × 19.65 m. It had a central courtyard that ran along the whole length of the building in the width of 8 m and was flanked on both sides by variously large rooms. In the second phase, at least some of the rooms were abandoned. Two rows of masonry plinths for columns or posts were set up in the interior, at a dis­tance of 3.5 m from one another. Nine plinths survived of the southwest row, two of which stood on top of the partition walls of earlier rooms. Three plinths survived of the northeast row. A large anteroom was located at the southeast façade, of which only the northeast wall survived. In contrast with Building D, Building E yielded more bricks and pottery. Its wooden phase is dated to the 1st–2nd centuries, the masonry phase with renovations to the 3rd–4th centuries. CEMETERY The watching brief on the gentle slope called Brinje, to the northwest of the settlement, revealed 26 cremation burials that form part of the cemetery associated with the Roman settlement (Fig. 1: 3).19 Burials were modest; some burial pits held a ceramic urn, others had the cremated remains simply strewn across the bottom. Most only contained the odd piece of pottery. There were no remains of tombstones or other burial markers. A limekiln from the Roman period was also found in the area of the cemetery that destroyed part of the burials. It was sub-rectangular in plan, measuring 6 m in length, 2.50 m in width and 0.50 m in depth. It held burned stones on the bottom and a thick layer of burned clay and charcoal along the edges. The kiln may have been used to burn pieces of marble found in the vicinity, either funerary monuments or pieces carried down by the Bistrica and Devina streams. ROMAN ROAD The Roman road running through the Bistrica area has been investigated in detail (Figs. 1; 6). Its cross sec­tions have shown a 6.6 m wide road with a gravel surface flanked on both sides by roadside ditches. In some places, the road was found at a very shallow depth, elsewhere it was covered by a metre or more thick alluvial deposits of the streams running from the Pohorje Hills. In Bistriško polje, the bedding for the road was mainly composed of 19 Strmčnik Gulič et al. 2008; Pahič 1978, 159 ff. do 20 cm in rahlo napeto. Ohranjena širina cestišča je znašala 6,85 m, odtočnih jarkov niso odkrili.22 Na severovzhodni strani antične naselbine se je cesta iz ravninskega Bistriškega polja povzpela na Brinje, kjer so jo na zamočvirjenih tleh utrdili s hras­tovimi debli, nato pa je po vzhodnih pobočjih dosegla dolino potoka Devine in potekala dalje ob južnem robu Velenika.23 Druga cestna trasa se je od glavne ceste odcepila in potekala vzhodno od stavbe A proti severu.24 ZAKLJUČEK V južnem delu današnje Slovenske Bistrice so bili na obeh straneh rimske ceste odkriti ostanki vsaj šestih stavb. Tlorisi stavb A, B, D in E se dajo delno rekon­struirati, stavbi C in Č pa sta skoraj povsem uničeni. V bližini stavbe A so bili odkriti sledovi kovaške dejavnosti in ostanki še ene tlakovane ceste, ki je od glavne rimske ceste vodila proti severu, proti Pohorju. Sklepamo, da je bila naselbina z velikimi stavbami izrazito povezana s prometom po cesti, pri čemer moramo pomisliti tudi na transport, skladiščenje in grobo obdelavo marmorja iz pohorskih kamnolomov. Morda je pri prevozu mar-morja s Pohorja imela določeno vlogo stranska cesta skozi naselbino.25 22 Žerjal, Černe 2013. 23 Strmčnik Gulič et al. 2008. 24 Strmčnik Gulič, Kajzer Cafnik 2004. Strmčnik Gulič, Kajzer Cafnik 2004. Kolegici Miheli Kajzer Cafnik se iskreno zahvaljujem za tehnično pomoč pri nastajanju članka. variously large cobbles covered with the surface layer of finer gravel, elsewhere gravel was laid directly onto the former ground or in rare cases onto a bedding of loam. The gravelly road surface in most cases survived in the thickness of roughly 15 cm, sometimes up to 40 and only exceptionally more.20 Reparations to the road surface were established in some places, as well as rut marks and variously deep roadside ditches on one or both sides. An almost 60 m long section was systematically investigated in advance of constructing a Spar store, revealing the maximum surviving width of 12.64 m, including the gravel washed off the road, and the minimum width of 6.04 m; the best surviving cross section showed a road 7.1 m wide and 0.31 m thick.21 More recent data on the road construction come from the south-western part of Slovenska Bistrica. The road there was also constructed of beaten gravel, cobbles, beaten fine sand and loam. It was up to 20 cm thick, with the slightly elliptical surface surviving in the width of 6.85 m; no roadside ditches were detected.22 To the northeast of the Roman settlement, the road ascended from the flatland of Bistriško polje to the el­evated terrain at Brinje, where the marshy ground dictated that the road be supported by oak logs. It then proceeded to the valley of the Devina stream and continued along the southern foot of the hill of Velenik.23 A road forked off the main road east of Building A to lead northwards.24 CONCLUSION The remains of at least six buildings have been unearthed in the southern part of present-day Slovenska Bistrica, on both sides of the Roman road. Buildings A, B, D and E can be partly reconstructed in ground plan, while Buildings C and Č were almost completely destroyed. Traces of smithing activities were found in the vicinity of Building A, as were the remains of a minor gravel road that forked off the main road to lead north towards the Pohorje Hills. The Roman settlement with large buildings was presumably closely linked with the traffic along the road, also with the transport, storage and rough process­ing of the marble quarried in the Pohorje Hills; the minor road that came to the settlement from the north may have been associated with the transport of marble.25 Translation: Andreja Maver 20 Pahič 1983, 267–272. 21 Kajzer Cafnik, Predan 2005. 22 Žerjal, Černe 2013. 23 Strmčnik Gulič et al. 2008. 24 Strmčnik Gulič, Kajzer Cafnik 2004. 25 Strmčnik Gulič, Kajzer Cafnik 2004. I would like to thank my colleague, Mihela Kajzer Cafnik, for her technical support in the contribution. DJURIĆ, B., H. W. MÜLLER 2009, White Marbles in Noricum and Pannonia: an outline of the Roman Quarries and their Products. – V / In: Ph. Jockey (ur. /ed. ), Marbres et autres roches de la Méditerranée antique: études interdisciplinaires, Actes du VIIIe Colloque international de l’Association for the study of marble and other stones used in antiquity (ASMOSIA), 111–127, Paris. JERMAN, M. 1976, Zavarovalno izkopavanje v Slovenski Bistrici. – Časopis za zgodovino in narodopisje 12/1, 6–15. JERMAN, M. 1977, Slovenska Bistrica. – Varstvo spomenikov 21, 262–264. KAJZER CAFNIK, M., P. PREDAN 2005, Slovenska Bistrica. – Varstvo spomenikov. Poročila 42 (2006), 136–137. MIKL-CURK, I. 1976, Nekaj novih zapažanj o rimskih naseljih v Spodnjem Grušovju in Slovenski Bistrici. – Časopis za zgodovino in narodopisje 12/1, 16–30. PAHIČ, S. 1960, Slov. Bistrica. – Varstvo spomenikov 7, 293. PAHIČ, S. 1969, Doslej neraziskan odsek rimske ceste Celeia – Poetovio (Ein bisher unerforschter Abschnitt der Römer­strasse Celeia – Poetovio). – Razprave SAZU 6, 309–361. PAHIČ, S. 1976, Poskusna izkopavanja rimske ceste v Slo­venski Bistrici (Versuchsgrabungen an der Römerstrasse in Slovenska Bistrica). – Arheološki vestnik 27, 232–259. PAHIČ, S. 1978, Najdbe z rimske ceste Slovenska Bistrica – Pragersko (Funde aus dem Römerstrassenbereich Slovenska Bistrica – Pragersko). – Arheološki vestnik 29, 129–289. PAHIČ, S. 1982, Bistriški svet v davnini. – Zbornik občine Slovenska Bistrica 1, 39–90. PAHIČ, S. 1983, Die Erforschung der Römerstrassen im nordöstlichen Slowenien. – Arheološki vestnik 34, 247–287. STRMČNIK GULIČ, M., S. PAHIČ 1985, Slovenska Bistrica. – Varstvo spomenikov 27, 252–257. STRMČNIK GULIČ, M. 1990, Nova dognanja arheoloških raz­iskav v občini. – Zbornik občine Slovenska Bistrica 2, 11–23. STRMČNIK GULIČ, M., M. KAJZER CAFNIK 2004, Poročilo o opravljenem arheološkem sondiranju na območju predvi­dene podjetniško-obrtne cone v Slovenski Bistrici. – Poro-čilo / Report, arhiv ZVKDS OE Maribor (neobjavljeno / unpublished). STRMČNIK GULIČ, M., P. PREDAN, D. SALECL, M. KAJ­ZER CAFNIK 2008, Spodnja Nova vas – rimska cesta Celeia–Poetovio. – Varstvo spomenikov. Poročila 44 (2007), 252–253. ŽERJAL, T., M. ČERNE 2013, Slovenska Bistrica – rimska cesta Celeia–Poetovio. – Varstvo spomenikov. Poročila 48 (2011), 250. Mira Strmčnik Gulič Zvezna ulica 68 SI-2000 Maribor mira.strmcnik.gulic@gmail.com Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 349–362 ANČNIKOVO GRADIŠČE Zvezdana MODRIJAN Izvleček Ančnikovo gradišče je utrjeno poznorimsko naselje, ki leži na južnih obronkih Pohorja. Hrib je bil prvič poseljen že v prazgodovini. Posamične najdbe so iz zgodnjerimske dobe, vendar ni bilo odkritih naselbinskih ostankov iz tega časa. Glavnina poselitve sodi v drugo polovico 4. in začetek 5. st. V tem obdobju je bilo naselje obdano s kamnitim obzidjem. Vhod na zahodni strani je bil zavarovan z dvema notranjima stranskima kriloma. Ob obzidju so se vrstile hiše, ki so bile večinoma lesene ali lesene s kamnitimi temelji. Številne najdbe različnega orodja kažejo na dokaj samooskrbno naselbino, uvoženi predmeti (npr. glazirana keramika, keramika z glajenim okrasom) pa dokazujejo tudi stike s širšim prostorom imperija. Dele vojaške noše povezujemo s prisotnostjo manjše vojaške posadke. Naselbina je propadla pred sredino 5. st. Zaradi močnih plasti žganine sklepamo, da je bila uničena v požaru. Ključne besede: Zgornja Panonija, Ančnikovo gradišče, poznorimsko obdobje, utrjena naselbina, orodje, vojaška noša, keramika Abstract Ančnikovo gradišče is a late Roman fortified settlement on the southern edges of the Pohorje Mountains. The hill was first settled in prehistory. Some isolated finds belong to the early Roman period, but no settlement remains from that time have been discovered. The bulk of the settlement dates to the second half of the 4th and the beginning of the 5th century. During that time, the settlement was surrounded by a stone wall. The entrance was on the west side, protected by two interior wings on the sides. A row of houses, most of them wooden or wooden with stone foundations, stood along the wall. Frequent finds of various tools suggest a relatively self-sufficient settlement, while the presence of imported objects (e.g. glazed pottery, pottery with burnished decoration) indicates contacts with the wider Empire. Items of military attire are associ­ated with the presence of a small garrison. The settlement collapsed before the mid-5th century. Thick layers of burnt material indicate that it was destroyed in a fire. Keywords: Pannonia Superior, Ančnikovo gradišče, fortified settlement, Late Roman period, tools, military attire, pottery https://doi.org/10.3986/9789610502586_19 Sl. 1: Ančnikovo gradišče in rimska naselja v dolini. Fig. 2: Ančnikovo gradišče and the Roman settlements below. UVOD Ančnikovo gradišče spada sicer v skupino pozno­antičnih utrjenih naselij, vendar v njem odkrito gradivo kaže še dobro vpetost v organizacijske strukture rimske države. Ker je v nasprotju z večino drugih podobnih na­selij, ki življenje nadaljujejo tudi še v drugi polovici 5. in v 6. st., Ančnikovo gradišče propadlo pred sredino 5. st., omogoča analizo prehoda rimske poselitve z ravninskih naselij na višinske lokacije. LEGA (sl. 1) Ančnikovo gradišče leži na jugovzhodnem delu Pohorja, na strmem pomolu, ki se dviga nad sotesko Bistriški vintgar. Pomol se na zahodu in jugu prevesi v izravnano sedlo, nato se znova dvigne. Z zahodne strani je dostop najlažji in na severozahodni strani je umeščen tudi vhod v naselbino. Naselje ponuja dober razgled na dolino, po kateri je potekala rimska cesta Emona–Poe- INTRODUCTION While the Ančnikovo gradišče belongs to the group of late antique fortified settlements, the discovered material indicates that it was still well-embedded in the organizational structures of the Roman state. Unlike the majority of similar settlements, which persisted in the second half of the 5th century and in the 6th century, Ančnikovo gradišče collapsed before the mid­5th century, which allows us to analyse the transition of Roman settlements from lowlands to higher areas. POSITION (Fig. 1) Ančnikovo gradišče lies in the southeastern part of the Pohorje Mountains, on a promontory rising steeply above the Bistriški vintgar gorge. To the west and south, the promontory becomes a flat saddle and then rises again. The settlement is most easily acces­sible from the western side and on the north-western side also the entrance was located. The settlement tovio, pa tudi na nasproti ležeče poznoantične višinske naselbine v zaledju Celeje in Petovione. Poselitev Pohorja v rimskem obdobju je povezana predvsem z izkoriščanjem pohorskih kamnolomov. Sre­dišče kamnoseštva je bilo v Šmartnem na Pohorju.1 Za šmarski revir Primož Predan predvideva obstoj šestih kamnolomov, za katere pa zaradi kasnejše rabe ni mogel dokazati izrabe v antičnem času.2 V bližini je bilo v Kalšah odkrito grobišče iz 1. in 2. st. Izkopanih je bilo 22 žganih grobov.3 Delavnico za predelavo marmorja domneva Bojan Djurić pod Pohorjem, na blagi vzpetini Velenik. Delovala naj bi do sredine 3. st., medtem ko naj bi pohor­ske kamnolome v manjšem obsegu izkoriščali še v 4. st.4 Rimskodobna najdišča v dolini so povezana s po­tekom rimske ceste iz Emone proti Petovioni. Starejši potek ceste iz 1. st. je bil delno odkrit na relaciji Strani­ce–Slovenska Bistrica tik pod Pohorjem, a je cesta zaradi hudourniških potokov s Pohorja v večji meri uničena ali prekrita z debelimi sloji naplavin.5 V 2. st. je bila zgrajena cesta mimo Slovenskih Konjic in Spodnjega Grušovja, po Ložniški dolini do Slovenske Bistrice ter dalje mimo Velenika proti Ptuju.6 Ostanki podeželske vile so bili odkriti v Gladomesu.7 Pri Spodnjem Grušovju se domneva obcestna postaja mansio Ragando.8 Rimski zaselek z več zidanimi stavbami, datiranimi v čas od 2. do 4. st., je bil v Slovenski Bistrici.9 Rimska cesta, ki je na začetku pomenila spodbudo za razvoj, je kasneje postala glavna žila za vpad različnihljudstev v času propada rimskega imperija. Življenje v dolini je postalo prenevarno, zato so se prebivalci začeli zatekati na višje ležeče lokacije. Odločilna za nastanek in razvoj Ančnikovega gradišča je bila njegova nekoliko odmaknjena lega, s katere pa je bilo vseeno mogoče nadzorovati dogajanje v dolini, omogočala pa je tudi komuniciranje s številnimi sočasnimi naselji v okolici.10 Nekatera bolj izpostavljena (Brinjeva gora, Ptujski grad) tako kot Ančnikovo gradišče niso preživela sredine 5. st.,11 na bolj odmaknjenih (Tinje nad Loko pri Žu-smu, Gradec pri Prapretnem, Rifnik) pa se je življenje nadaljevalo še vse do konca 6. st.12 1 Djurić 1995, 73–86; 2005, 76. 2 Predan 2001, 103. 3 Strmčnik Gulič, Predan 2006. 4 Djurić 2001, 63–65. 5 Pahič 1983a, 55. 6 Izbrana literatura: Pahič 1967; 1968; 1969; 1974a; 1978. 7 Kajzer Cafnik, Predan 2013. 8 Pahič 1969, 329; Mikl-Curk 1974; Mikl-Curk 1976. 9 Pahič 1985, 252–257; Strmčnik Gulič 1990a; glej tudi članek Strmčnik Gulič 2020, v tej knjigi. 10 Brinjeva gora (Pahič 1980; Ciglenečki 1987, 67–68); Zbelovska gora (Pahič 1980; Ciglenečki 1987, 67–68); Do-načka gora (Ciglenečki1985); Rifnik (Bolta1981); Tinje nad Loko pri Žusmu (Ciglenečki 2000); Gradec pri Prapretnem (Ciglenečki 1981); Ptujski grad (Ciglenečki 1987, 67–68). 11 Ciglenečki 1993. 12 Tinje nad Loko pri Žusmu (Ciglenečki 2000); Gradec provides a good view of the plain below, where the Roman Emona - Poetovio road ran, and also of the late antique settlements on the hills on the opposite side of the valley, in the hinterland of the towns of Celeia and Poetovio. The Roman-period settlements in the Pohorje Mountains were predominantly connected with the exploitation of the Pohorje quarries. The centre ofquarrying was in the village of Šmartno na Pohorju.1 Primož Predan assumed the existence of six quarriesin the area of Šmartno, but was unable to prove their use in the Roman period due to later exploitation.2 In the nearby village of Kalše, a cemetery from the 1st and 2nd centuries was discovered and 22 crema­tion graves were excavated.3 Bojan Djurić assumes a marble workshop below the Pohorje, on the small hill of Velenik. The workshop seems to have functioned until the mid-3rd century, while on a smaller scale, the exploitation of the Pohorje quarries probably survived into the 4th century.4 The Roman period sites in the lowlands are related to the course of the Roman road from Emona to Poetovio. Part of the course of an earlier road from the 1st century was discovered between Stranice and Slovenska Bistrica, just below the Pohorje, but due to streams from the mountains, most of the road was either destroyed or covered by thick layers of alluvial deposits.5 In the 2nd century, a road was built past Slovenske Konjice and Spodnje Grušovje, along the valley of Ložniška dolina to Slovenska Bistrica and beyond, past Velenik and towards Ptuj.6 Remains of a villa rustica were discovered in Gladomes.7 The mansio Ragando road station was supposedly located near Spodnje Grušovje.8 A Roman settlement with several masonry buildings, dated between the 2nd and 4th centuries, was in Slovenska Bistrica.9 While the Roman road was first an incentive for the development, it later became the main artery for incursions of various peoples in the time of the decline of the Roman Empire. Life in the lowlands became too dangerous and the population started to retreat to higher locations. Crucial for the beginning and development of Ančnikovo gradišče was its somewhat remote position, which at the same time still enabled control over the events below and communication with many contemporary settlements in the vicinity.10 1 Djurić 1995, 73–86; 2005, 76. 2 Predan 2001, 103. 3 Strmčnik Gulič, Predan 2006. 4 Djurić 2001, 63–65. 5 Pahič 1983a, 55. 6 Selected literature: Pahič 1967; 1968; 1969; 1974a; 1978. 7 Kajzer Cafnik, Predan 2013. 8 Pahič 1969, 329; Mikl-Curk 1974, 156; Mikl-Curk 1976. 9 Pahič 1985, 252–257; Strmčnik Gulič 1990a; see also the contribution Strmčnik Gulič 2020, in this book. 10 Brinjeva gora: Pahič 1980, Ciglenečki 1987, 67–68; ZGODOVINA RAZISKAV Pred letom 1984 je bil kopasti vrh Ančnikovega gradišča poraščen z gozdom in podrastjo, na zahodnem delu gozdne kope pa so bili vidni do 2,5 m visoki ostanki zidov, zato je najdišče že zgodaj pritegnilo pozornost. Pr-vo poročilo o Ančnikovem gradišču v strokovni literaturi je iz leta 1915, ko ga je Walter Schmid pravilno časovno uvrstil v poznorimsko dobo.13 Nekaj kasneje je Balduin Saria menil, da predstavljajo zidovi razvaline srednjeve­škega gradu.14 Podobno je na začetku menil tudi Stanko Pahič, ki pa je spremenil mnenje, potem ko je leta1981 izvedel prvo sondiranje in odkril značilno poznorimsko gradivo. Tedaj je bil izdelan tudi prvi načrt naselbine.15 Leta 1984 je dal lastnik gozdno zemljišče poravnati, da bi ga spremenil v obdelovalno površino, pri tem pa je uničil zidove in kulturne plasti na zahodnem delu.16 Zato so se leta 1986 začele najprej zaščitne raziskave, ki so kasneje prerasle v sistematske in so s presledki potekale do leta 2015. Izkopavanja med 1986 in 2010 je izvajal ZVKD Maribor (vodja Mira Strmčnik Gulič), v letu 2015 pa ZRC SAZU in Skupina Stik (vodja Zvezdana Modrijan).17 Prve raziskave so se začele z oceno nastale škode in nadaljevale s sondiranjem severozahodnega dela utrjene kope, kjer je bil odkrit del obrambnega obzidja z južnim krilom vhoda.18 V devetdesetih letih so potekale raz­iskave južnega in vzhodnega dela obrambnega obzidja in nanj vezanih objektov.19 Z raziskavami med letom 2006 in 2010 so se spet osredotočili na severozahodni, vhodni del naselbine.20 Osrednji del najdišča je bil raziskan z nekaj sondami, ki pa niso pokazale ostankov poselitve. V letu 2015 je bil raziskan še del severnega obzidja v dolžini 35 m in objekti ob njem (sl. 3).21 Rezultati prve faze raziskav so predstavljeni tudi v poljudni publikaciji,22 gradivo iz izkopavanj med leto-ma 1986 in 2010 je bilo tema diplomske naloge Mateje Ravnik.23 pri Prapretnem (Ciglenečki 1981); Rifnik (Bolta 1981).13 Schmid 1915, 229–305. 14 Saria 1939, 32 (Jurišna vas). 15 Pahič 1983a, 223–227; Pahič 1983b. 16 Strmčnik Gulič 1985, 227. 17 Raziskave so potekale v okviru slovensko-hrvaškega projekta Interreg: Rojstvo Evrope. 18 Izbrana literatura: Strmčnik Gulič 1987; 1988; 1989; 1990a; 1990b. 19 Izbrana literatura: Strmčnik Gulič 1991; 1992; 1995; 1997. 20 Strmčnik Gulič, Kajzer Cafnik 2010; Strmčnik Gulič 2012. 21 Modrijan 2017. 22 Strmčnik Gulič, Ciglenečki 2003. 23 Ravnik 2006. Like Ančnikovo gradišče, some of the more exposed of these settlements (Brinjeva gora, Ptuj Castle) did not survive the middle of the 5th century,11 while in the more remote ones (Tinje above Loka pri Žusmu, Gradec near Prapretno, Rifnik) life continued until the end of the 6th century.12 RESEARCH HISTORY Before 1984, the domed top of Ančnikovo gradišče was covered by forest and undergrowth. Since up to 2.5 m high remains of walls were visible in the western part of the dome, the site attracted atten­tion early. The first report on Ančnikovo gradišče in archaeological literature is from 1915, when it was correctly dated to the late Roman period by Walter Schmid.13 Not much later, however, Balduin Saria interpreted the walls as the ruins of a medieval castle14. Stanko Pahič was first of the same opinion, but changed his mind after 1981, when he conducted the first test excavation and discovered typical late Roman material. That was also when the first plan of the settlement was made.15 In 1984 the owner had the forest levelled in order to make it cultivable land. By doing so, he destroyed the walls and cultural layers in the western part.16 Rescue excavations, which started in 1986, later became systematic, lasting intermittently until 2015. Between 1986 and 2010, the excavations were conducted by the ZVKD Maribor (under the direction of Mira Strmčnik Gulič), while in 2015 they were conducted by ZRC SAZU, Inštitut za arheologijo / Institute of archaeology and the Skupina Stik Institute (under the direction of Zvezdana Modrijan).17 The surveys began with damage assessment and continued with test trenching in the northwestern part of the fortified hilltop, where a part of the defence wall and the southern entrance wing were discovered.18 In the 1990s, surveys of the southern and eastern parts of the defence wall and the associated buildings were Zbelovska gora: Pahič 1980, Ciglenečki 1987, 67–68; Do-načka gora: Ciglenečki 1985; Rifnik: Bolta1981; Tinje above Loka pri Žusmu; Ciglenečki 2000; Gradec near Prapretno: Ciglenečki 1981; Ptuj Castle: Ciglenečki 1987, 67-68. 11 Ciglenečki 1993. 12 Tinje above Loka pri Žusmu: Ciglenečki 2000; Gradec near Prapretno: Ciglenečki 1981; Rifnik: Bolta 1981.13 Schmid 1915, 229–305. 14 Saria 1939, 32 (Jurišna vas). 15 Pahič 1983a, 223–227; Pahič 1983b. 16 Strmčnik Gulič 1985, 227. 17 The research was done within the framework of a Slo­ venian – Croatian Interreg project The Birth of Europe.18 Selected literature: Strmčnik Gulič 1987; 1988; 1989; 1990a; 1990b. Sl. 2: Ančnikovo gradišče. Rimska kipca. Fig. 2: Ančnikovo gradišče – Roman statuettes. (Foto / Photo: ZVKD Maribor) ARHEOLOŠKI OSTANKI PRAZGODOVINA Najstarejša poselitev Ančnikovega gradišča sega v prazgodovino, vendar za zdaj ni bilo odkritih naselbin­skih ostankov, ki bi jih lahko pripisali temu obdobju. V prid poselitvi tako govorijo le redke drobne najdbe. Najstarejše je kamnito orodje s ploskovno retušo, ki je bilo skupaj z nekaj koščki ožje nedoločljive prazgo­dovinske keramike odkrito na vrhu sterilne plasti na severnem delu naselbine. Opredeljeno je v eneolitik.24 Edini časovno opredeljiv keramični kos (jezičast ročaj) sodi v konec bronaste dobe (Ha B) oziroma okvirno v 8. st. pr. Kr.,25 najden pa je bil v premešani plasti, ki sodi v čas po propadu naselbine. ZGODNJERIMSKA DOBA (1.–3. st.) Tudi iz tega časa ni bilo odkritih ostankov poselitve. Obdobje je zastopano le z drobnimi najdbami, med kate­rimi izstopata bronasta kipca (Merkur in Apolon; sl. 2), nekaj novcev, del čaše iz modrega stekla ter nekaj kosov keramike. Vsi predmeti so bili najdeni v poznorimskih commenced.19 The investigations between 2006 and 2010 again focused on the northwestern part of the settlement, where the entrance was.20 The central part of the site was investigated with a few test trenches, which revealed no settlement remains. In 2015, a 35-m-long part of the northern wall and the adjoining buildings were investigated (Fig. 3).21 The results of the first phase of the research were pre­sented, among others, in a popular-science publication,22 while the material from the 1986–2010 excavations is the subject of the graduation thesis of Mateja Ravnik.23 ARCHAEOLOGICAL REMAINS PREHISTORY While the earliest habitation at Ančnikovo gradišče dates back to prehistory, no settlement remains from that period have been discovered so far and only a few small finds bear witness to the settlement. The earliest 19 Selected literature: Strmčnik Gulič 1991; 1992; 1995; 1997. 20 Strmčnik Gulič, Kajzer Cafnik 2010; Strmčnik Gulič 2012. 21 Modrijan 2017. 24 Opredelitev Matija Turk. 22 Strmčnik Gulič, Ciglenečki 2003. 25 Dular 1993, 104–105. 23 Ravnik 2006. Sl. 3: Ančnikovo gradišče, tloris. Fig. 3: Ančnikovo gradišče – plan view. plasteh ali plasteh iz časa po uničenju naselbine. Glede na odsotnost naselbinskih ostankov se zdi verjetnejše, da so bili ti predmeti na naselbino prineseni iz doline. Morda pa predstavljajo ostanek še nelociranega svetišča na hribu. POZNORIMSKA DOBA Arhitektura Poznorimska naselbina je bila obdana z obrambnim obzidjem (sl. 3). Obzidje je najširše (med 1,4 in 1,8 m) na najlažje dostopnem jugozahodnem delu, kjer je kopa utrdbe le nekaj metrov dvignjena nad sedlo. Na severni strani, vzhodno od vhoda, kjer je strmina največja, je obzidje široko med 0,8 in 1 m. Na zahodnem delu je uni-čeno, njegov zaključek na severovzhodni strani pa še ni raziskan. Sonde na jugovzhodnem delu niso več pokazale ostankov obzidja.26 Tudi lidarsko snemanje (zračno laser-sko skeniranje) ni povsem razjasnilo poteka obzidja na vzhodni strani (sl. 4).27 Na lidarskem modelu reliefa je na vzhodnem delu videti ozek jezik z več terasami, vendar ni popolnoma jasno, ali je ta del dejansko pripadal naselbini ali pa terase pomenijo izravnavo terena zunaj obzidja. 26 Strmčnik Gulič, Ciglenečki 2003, 10–21. Posnetek je bil interpretiran v okviru slovensko-hr­vaškega projekta Interreg: Enjoy heritage. artefact is facially retouched stone tool, discovered on top of the sterile layer in the northern part of the settlement, together with fragments of prehistoric pot­tery that cannot be more precisely identified. The tool dates to the Neolithic period.24 The only chronologi­cally identifiable fragment of pottery (a tongue-shaped handle) dates to the end of the Bronze Age (Ha B), roughly to the 8th century BC.25 It was discovered in a mixed layer from the time after the collapse of the settlement. EARLY ROMAN PERIOD (1st–3rd centuries) Again, no settlement remains have been discovered. The period is represented only by small finds, among which two bronze statuettes (Mercury and Apollo) stand out (Fig. 2). A few coins, a fragment of a blue glass beaker, and some pottery fragments were also found. All the artefacts were discovered either in late Roman layers or in layers from after the destruction of the settlement. With the absence of settlement remains, it seems more likely that these artefacts were brought to the hillfort from below. They may, however, represent the remains of a Roman sanctuary which is not yet located. 24 Dated by Matija Turk. 25 Dular 1993, 104–105. Sl. 4: Ančnikovo gradišče, lidarski posnetek. Fig. 4: Ančnikovo gradišče – Lidar derived digital surface model. (Vizualizacija / Visualisation: E. Lozić) Pot v naselbino je vodila s severozahodne strani, kjer je bil v obzidju vhod, varovan z dvema notranjima stranskima kriloma, potem pa je zavila proti središču naselbine. Večina stavb na južni strani naselja je razporejenih ob obzidju, dve pa sta od njega nekoliko odmaknjeni. Središče naselbine verjetno ni bilo poseljeno, saj tam sonde niso pokazale sledov stavb. Na severnem delu stavbe sledijo poteku naravne terase, ki nekoliko ublaži padec terena. Večina raziskanih stavb je bila lesenih. Izjema sta dve ob južnem zidu, ki sta imeli kamnite temelje. Natančno število lesenih objektov je težko ugotoviti, saj vseh jam za kole ni mogoče povezati v smiselne tlorise. Nekaj objektov je tako določenih le na podlagi zasekov za izravnavo površine, ognjišč in kulturnih plasti. Pri gradnji sta bili uporabljeni tako gradnja z nav-pičnimi koli (stojkami) kot tudi gradnja s horizontalno položenimi bruni. Taka gradnja je vidna na severni stra­ni najdišča, kjer je padec terena zelo velik. Stavbe so bile tam postavljene na delno zravnano geološko osnovo, na južni (višji) strani zavarovane s kamnito škarpo, na osta­lih stranicah pa zgrajene iz horizontalnih brun. Lesena tla so bila na južni strani naslonjena na rob, na severni, nižji strani pa dvignjena od tal. Kamnite škarpe so očitno varovale objekte pred erozijo s strmega pobočja. LATE ROMAN PERIOD Architecture The Late Roman settlement was surrounded by a defensive wall (Fig. 3). The wall is the thickest (between 1.4 and 1.8 m) in the accessible southwestern part, where the dome with the hillfort is just a few metres above the saddle. In the northern part, east of the entrance, where the slopes are the steepest, the wall is between 0.8 and 1 m wide. In the western part the wall is destroyed, while the northeastern end of the wall has not been investigated. In the southeastern part, test trenches revealed no remains of the wall.26 Even the airborne laser scanning (Lidar) did not entirely clarify the course of the wall on the eastern side (Fig. 4).27 There, the Lidar derived digital surface model shows a narrow tongue with several terraces, but it is unclear whether the area actually belonged to the settlement or the terraces are just the result of levelling the ground outside the wall. The way to the settlement led to the northwestern side, where there was an entrance in the wall, protected 26 Strmčnik Gulič, Ciglenečki 2003, 10–21. 27 Interpretation of the image was done within the fra­mework of the Slovene – Croatian Interreg project Enjoy heritage. Sl. 5: Ančnikovo gradišče. Del obzidja in črna kulturna plast, ostanek lesenega objekta ob obzidju. Fig. 5: Ančnikovo gradišče – a segment of the wall and a black cultural layer – remains of a wooden building next to the wall. (Foto / Photo: G. Babič) Ob severnem delu obzidja, na strmem pobočju vzhodno od vhoda, je bil postavljen tudi na obzidje naslonjen in od tal dvignjen objekt (sl. 5). Da gre za bi-valni objekt, ne le za preprosto leseno platformo za lažji dostop do obzidja, kaže predvsem izredno veliko število za bivalne objekte značilnih najdb (npr. groba keramika, brusi, žrmlje) v žganinsko-ruševinskih plasteh. Najdbe Najštevilčnejšo skupino najdb predstavlja kerami­ka. Prevladuje groba hišna lončenina lokalne izdelave (sl. 6: 1–3). Od uvoženega posodja je najštevilčnejše glazirano, med katerim so zastopane predvsem odprte oblike (sklede in melnice), pojavljajo pa se tudi vrči (sl. 6: 4–5). Tovrstno posodje je v podonavskem prostoru datirano v drugo polovico 4., pojavlja pa se tudi še na začetku 5. st.28 Zastopana je tudi glajena keramika (sl. 6: 6), ki je datirana v čas zadnje tretjine 4. do sredine 5. st.29 in prav tako kaže vplive panonskega prostora. Afriški importi so izredno redki. Med kovinskim gradivom prevladuje orodje. Med drugim so bili najdeni noži, srpi, rovnice, škarje za ovce, sekire, klešče, strgala, žage in šestilo (sl. 7: 1–3; 9). 28 Pregled pri Hárshegyi, Ottományi 2013, 489–499. 29 Pregled pri Hárshegyi, Ottományi 2013, 500–509. by two interior side wings, and then turned towards the centre of the settlement. While most of the buildings in the southern part of the settlement were arranged along the wall, two buildings stood slightly further from it. Test trenches revealed no remains of buildings in the centre of the settlement, which was probably uninhabited. In the northern part, buildings follow the course of a natu­ral terrace, which slightly mitigates the steepness of the slope. Most of the investigated buildings were made of wood. There are two exceptions near the southern wall, which had stone foundations. The number of wooden buildings is difficult to determine with precision since all postholes cannot be connected into a coherent ground plan. Some buildings are identified only on the basis of ground levelling cuts, fireplaces and cultural layers. Construction was done using either vertical posts (postholes) or horizontal logs. The latter can be seen in the northern part of the site, where the fall of the ground is very steep. The buildings in that part stood on the partially levelled geological base. On the southern (higher) side, they were protected by a stone escarpment, while the other three sides were made of horizontal logs. The wooden floor leaned on the edge on the southern side, while on the lower northern side it was raised from the ground. Stone 67 Sl. 6: Ančnikovo gradišče. Izbor keramike. M. = 1 : 3. Fig. 6: Ančnikovo gradišče – a selection of pottery. Scale = 1 : 3. Izrazita je tudi skupina najdb, povezanih z vojsko. Med elemente vojaške noše spadajo dve fibuli s čebu­lastimi glavicami in deli poznorimskih pasnih garnitur (sl. 8). Od orožja je bilo odkritih nekaj puščic,30 sulična ost31 in ježek za oviranje konjenice.32 30 Strmčnik Gulič, Ciglenečki 2003, 27; t. 1: 9–12. 31 Modrijan 2017, sl. 4. 32 Strmčnik Gulič, Ciglenečki 2003, 11; t. 1: 5. escarpments seem to have protected the buildings from erosion down the steep slope. At the northern segment of the wall, on the steep slope east of the entrance, a building was raised from the ground and leaning on the wall (Fig. 5). A very high number of residential artefacts (e.g. coarse pot­tery, whetstones, quern-stones) in burnt rubble layers indicates that this was a residential dwelling rather than merely a simple wooden platform for easier ac­cess to the wall. Sl. 7: Ančnikovo gradišče. Izbor orodja. M. = 1 : 2. Fig. 7: Ančnikovo gradišče – a selection of tools. Scale = 1 : 2. Kronologija in interpretacija Poselitev Ančnikovega gradišča lahko postavimo v čas po sredini 4. st., morda celo v njegovo zadnjo tretjino.Življenje na njem se je končalo verjetno že pred sredino 5. st. V tem dokaj kratkem časovnem obdobju naselje kaže dve fazi, ki si sledita brez večje časovne prekinitve. Na severnem delu je bilo obzidje delno porušeno in po­tem popravljeno, ob njem sta bili odkriti dve žganinski plasti, ki kažeta, da je objekt ob zidu zgorel, bil postavljen na novo in potem zgorel še enkrat. Najdbe iz obeh plasti so zelo enotne, kar kaže, da se je to dogajalo v dokaj kratkem časovnem razponu. Prav tako so sledovi dveh faz vidni na južnem delu obzidja.33 Po zadnjem požaru je bilo naselje za nekaj stoletij zapuščeno, sledovi vnovične poselitve so šele iz zgod­njega srednjega veka. Naselbina na Ančnikovem gradišču je nastala v času, ko je postopoma zamirala poselitev v dolini. Pre-hod iz nižinske v višinsko poselitev na obravnavanem območju očitno ni bil trenuten. Nekateri elementi kažejo vsaj občasno obljudenost nekaterih nižinskih naselij in podeželskih vil še po sredini 4. st. (Spodnje Grušovje, 33 Strmčnik Gulič, Ciglenečki 2003, 10–19. Sl. 9: Ančnikovo gradišče. Šestilo. Fig. 9: Ančnikovo gradišče. Compass. Finds The most numerous group of finds is pottery. Coarse kitchenware of local production prevails (Fig. 6: 1–3). When it comes to imported vessels, glazed ware is the most common, especially open forms (bowls and mortaria), but also jugs (Fig. 6: 4–5). In the Danube region such pottery is dated to the second half of the 4th century, but it still occurs as late as the beginning of the 5th century.28 Burnished pottery (Fig. 6: 6), dated between the last third of the 4th century and the middle of the 5th century,29 is also represented and shows the influence of the Pannonnian area. African imports are very rare. Tools are the most common metal finds. The ar­tefacts include knives, sickles, hoes, sheep-shearing scissors, axes, pincers, scrapers, saws, and a compass (Fig. 7: 1–3; 9). A significant number of finds are associated with the army. The elements of military attire include two crossbow fibulae and fragments of late Roman belt sets (Fig. 8). Weapons include a few arrows,30 a spearhead,31 and a small cheval de frise.32 Chronology and interpretation The occupation of Ančnikovo gradišče can be dated to the time after the middle of the 4th century, perhaps even the final third of the 4th century. Life in the settlement probably ended before the mid-5th century. In this relatively short time, two phases can be identified, which follow each other without a gap between them. In the northern part, the wall was partly demolished and then rebuilt. Two burnt layers indicate that the building next to the wall burned down, was rebuilt and then burned down again. The finds from both layers are very uniform, which indicates that all this happened in a relatively short time. The two phases can also be discerned in the southern part of the wall.33 After the last fire, the settlement was abandoned for a few centuries, and habitation traces do not reappear before the Early Middle Ages. Ančnikovo gradišče was inhabited in the period when lowland settlements were gradually abandoned. It seems that the transition from lowland to highland habitation in the area did not happen instantly. Certain elements indicate that some lowland settlements and villae rusticae were still inhabited, at least periodically, 28 Overview in Hárshegyi, Ottományi 2013, 489–499. 29 Overview in Hárshegyi, Ottományi 2013, 500–509. 30 Strmčnik Gulič, Ciglenečki 2003, 27, Pl. 1: 9–12. 31 Modrijan 2017, Fig. 4. 32 Strmčnik Gulič, Ciglenečki 2003, 11, Pl. 1: 5. 33 Strmčnik Gulič, Ciglenečki 2003, 10–19. Betnava, Radvanje).34 Po drugi strani nekateri elementi (npr. grobišče na Brinjevi gori) kažejo prvo poselitev višinskih lokacij že v 3. st. Zgodnje poselitve za Ančni­kovo gradišče tako za zdaj ne moremo potrditi, vendar je življenje na vsaj nekaterih višinske in nižinske lokacijah očitno nekaj časa potekalo vzporedno. Struktura najdb na Ančnikovem gradišču kaže dokaj samooskrbno naselbino. Deli vojaške noše in uvožene skupine keramike (glazirana, glajena) pa doka­zujejo, da je bila v zadnji tretjini 4. in prvi tretjini 5. st. tu nastanjena tudi manjša vojaška posadka.35 Ker je bila večina z vojsko povezanih predmetov odkrita v predelu ob vhodu v naselje, lahko na tem mestu domnevamo bivališče te posadke. Posadka je nadzirala promet po do-lini, morda pa njeno prisotnost lahko povezujemo tudi z nadzorom pohorskih rudišč (marmor, železova ruda). ZGODNJI SREDNJI VEK Po uničenju poznorimske naselbine je bil predel nekaj časa neposeljen. Vnovični sledovi bivanja sodijo v čas zgodnjega srednjega veka. Gre predvsem za značilno lončenino, poznano tudi iz sočasnih najdišč v dolini, ki je bila najdena v glavnem pomešana s poznorimskim gradivom. Naselbinske strukture iz tega obdobja niso bile zanesljivo ugotovljene.36 Tako se postavlja vprašanje, ali ti predmeti doka­zujejo stalno bivanje ali gre morda le za občasne obiske uničene poznorimske naselbine (morda zaradi paše?). Spodnje Grušovje (Mikl Curk 1976, 20); Betnava (Strmčnik Gulič 1997, 269–272; t. 1: 1); Radvanje (Strmčnik Gulič 1997, 269–272). 35 Strmčnik Gulič, Ciglenečki 2003, 11; Modrijan 2017, 168–169. 36 Ciglenečki, Strmčnik Gulič 2002, 72; sl. 13. even after the mid-4th century (Spodnje Grušovje, Betnava, Radvanje).34 On the other hand, some ele­ments (e.g. the cemetery at Brinjeva gora) suggest that the first highland settlements occur as early as the 3rd century. While such an early date of first settlement cannot be ascertained for Ančnikovo gradišče, some highland and lowland sites do seem to have existed simultaneously for a time. The structure of the finds from Ančnikovo gradišče indicates a relatively self-sufficient settlement. Items of military attire and imported types of pottery (glazed, polished) provide evidence that a small garrison was stationed there in the last third of the 4th century and the first third of the 5th century.35 Since the majority of the military-related items were discovered in the area around the entrance to the settlement, it can be surmised that the garrison house was situated there. While the crew supervised the traffic below, its pres­ence could perhaps also be associated with the control of the Pohorje marble and iron ore deposits. EARLY MIDDLE AGES After the destruction of the late Roman settlement, the site was left uninhabited for a while. Traces of habitation reappear in the Early Middle Ages. They predominantly include the characteristic pottery known from contemporary lowland sites. It was most often discovered mixed up with the late Roman material. Settlement structures from this period have not been identified with certainty.36 The question presents itself whether the Early Medieval artefacts indicate permanent habitation or just occasional visits to the demolished late Roman settlement (perhaps for grazing?). Translation: Meta Osredkar 34 Spodnje Grušovje: Mikl Curk 1976, 20; Betnava: Strmčnik Gulič 1997, 269–272, Pl. 1: 1; Radvanje: Strmčnik Gulič 1997, 269–272. 35 Strmčnik Gulič, Ciglenečki 2003, 11; Modrijan 2017, 168–169. 36 Ciglenečki, Strmčnik Gulič 2002, 72, Fig. 13. BOLTA, L. 1981, Rifnik pri Šentjurju. Poznoantična naselbina in grobišče (Rifnik - spätantike Siedlung und Gräberfeld). – Katalogi in monografije 19. CIGLENEČKI, S. 1981, Rezultati prvih raziskovanj na Gradcu pri Prapretnem (Ergebnisse der ersten Forschungen auf Gradec bei Prapretno). – Arheološki vestnik 32, 417–453. CIGLENEČKI, S. 1985, Donačka gora. – Varstvo spomenikov 27, 275–276. CIGLENEČKI, S. 1987, Höhenbefestigungen aus der Zeit vom 3. bis 6. Jh. im Ostalpenraum (Višinske utrdbe iz časa 3. do 6. st. v vzhodnoalpskem prosotoru. – Dela 1. razreda SAZU 31. CIGLENEČKI, S. 1993, Arheološki sledovi zatona antične Petovione (Archaeological Traces of the Fall of Roman Poetovio). – V / In: B. Lamut (ur. / ed.), Ptujski arheološki zbornik. Ob 100-letnici muzeja in Muzejskega društva, 505–520, Ptuj. CIGLENEČKI, S. 2000, Tinje nad Loko pri Žusmu. Pozno­antična in zgodnjesrednjeveška naselbina / Tinje oberhalb von Loka pri Žusmu. Spätantike und frühmittelalterliche Siedlung. – Opera Instituti Archaeologici Sloveniae 4. CIGLENEČKI, S., M. STRMČNIK GULIČ 2002, Sledovi zgodnje slovanske poselitve južno od Maribora. – V / In: M. Guštin (ur. / ed.), Zgodnji Slovani. Zgodnjesrednjeveška lončenina na obrobju vzhodnih Alp / Die frühen Slaven. Frühmittelalterliche Keramik am Rand der Ostalpen, 67–75, Ljubljana. DJURIĆ, B. 1995, Eastern Alpine Marble and Pannonian Trade. – V / In: B. Djurić, I. Lazar (ur. / eds.), Akten des IV. internationelen Kolloquiums über Probleme des provinzialrö­mischen Kunstschaffens / Akti IV. mednarodnega simpozija o problemih rimske provincialne umetnosti, Situla 36. DJURIĆ, B. 2001, The end of Roman quarrying on Pohorje. – V / In: M. Vomer-Gojkovič, N. Kolar (ur. / eds.), Ptuj v Rimskem cesarstvu. Mitraizem in njegova doba / Ptuj im Römischen Reich. Mithraskult und seine Zeit / Ptuj in the Roman Empire. Mithraism and its era. – Ptuj. DJURIĆ, B. 2005, Poetovio and the Danubian marble trade. – V / In: M. Mirković (ur. / ed.), Römische Städte und Festungen an der Donau. Akten der regionalen Konferenz, Beograd, 16-19. Oktober 2003, 75–82, Beograd. DULAR, J. 1993, Začetki železnodobne poselitve v osrednji Sloveniji. – Arheološki vestnik 44, 101–112. HÁRSHEGYI, P., K. OTTOMÁNYI 2013, Imported and local Pottery in late Roman Pannonia. – V / In: L. Lavan (ur. / ed.), Local Economies? Production and Exchange of Inland Regions in Late Antiquity, Late Antique Archaeology 10, 471–528. KAJZER CAFNIK, M., P. PREDAN 2013, 22. Gladomes – villa rustica. – Varstvo spomenikov. Poročila 48 (2011), 60–61. MIKL CURK I. 1974, Spodnje Grušovje pri Slovenskih Kon­jicah. – Varstvo spomenikov 17–19/1, 156. MIKL-CURK, I. 1976, Nekaj novih zapažanj o rimskih na­seljih v Spodnjem Grušovju in Slovenski Bistrici (Sur les habitations romaines de Spodnje Grušovje et Slovenska Bistrica). – Časopis za zgodovino in narodopisje 12, 16–30. MODRIJAN, Z. 2017, New military finds from Ančnikovo gradišče. – V / In: I. Dörfler, P. Gleirscher, S. Lädstatter, I. Pucker, Ad amussim. Festschrift zum 65. Geburtstag von Franz Glaser, Kärntner Museumsschriften 85, 159–174. PAHIČ, S. 1967, Slov. Konjice – Slov. Bistrica – rimska cesta. – Arheološki pregled 9, 91–92. PAHIČ, S. 1968, Slovenska Bistrica – Šikole – rimska cesta. – Arheološki pregled 10, 246–247. PAHIČ, S. 1969, Doslej neraziskan odsek rimeske ceste Celeia – Poetovio (Ein bisher unerforster Abschnitt der Römer­strasse Celeia - Poetovio). – Razprave SAZU 6, 309–361. PAHIČ, S. 1974a, Slovenska Bistrica – rimska cesta in naselje. – Arheološki pregled 16, 102–103. PAHIČ, S. 1974b, Velenik pri Spodnji Polskavi. – Varsvo spo­menikov 17–19/1, 160-162. PAHIČ S. 1978, Najdbe z rimske ceste Slovenska Bistrica–Pra­gersko (Funde aus dem Römerstrassenbereich Slovenska Bistrica–Pragersko). – Arheološki vestnik 29, 1978, 129–289. PAHIČ, S. 1980, Prvi podatki o grobi hišni lončenini z Brinjeve gore (Vorbericht über die Grobe Hauskeramik von Brinjeva gora). – Arheološki vestnik 31, 89–112. PAHIČ, S. 1983a, Bistriški svet v davnini. – V / In: Zbornik občine Slovenska Bistrica 1 (ob 750 letnici mesta), 39–91, Slovenska Bistrica. PAHIČ, S. 1983b, Jurišna vas. – Varstvo spomenikov 25, 223–227. PAHIČ, S. 1985, Slovenska Bistrica. – Varstvo spomenikov 27, 252–257. PREDAN, P. 2001, Črešnjevec, cerkev Sv. Martina. Problematika rimskih spolij (kamnolomi, transport, marmorata). – Diplomska naloga / BA thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). RAVNIK, M. 2006, Zaščitne raziskave Ančnikovega gradišča pri Jurišni vasi v letih od 1986 do 1994. – Diplomska naloga / BA thesis, Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). SARIA, B. 1939, Archäologische Karte von Jugoslawien: Blatt Rogatec. SCHMID, W. 1915, Die Ringwälle des Bachengebirges. – Mit­teilungen der Prähistorischen Kommission 2/3, 229–305. STRMČNIK GULIČ, M. 1985, Jurišna vas. – Varstvo spome­nikov 27, 227–228. STRMČNIK GULIČ, M. 1987, Jurišna vas, Ančnikovo gradiš-če. – Varstvo spomenikov 29, 275–277. STRMČNIK GULIČ, M. 1988, Jurišna vas, Ančnikovo gradiš-če. – Varstvo spomenikov 30, 258–263. STRMČNIK GULIČ, M. 1989, Jurišna vas, Ančnikovo gradiš-če. – Varstvo spomenikov 31, 250–251. STRMČNIK GULIČ, M. 1990a, Nova dognanja arheoloških raziskav v občini. – V / In: Zbornik občine Slovenska Bistrica 2, 11–24, Slovenska Bistrica. STRMČNIK GULIČ, M. 1990b, Jurišna vas, Ančnikovo gra­dišče. – Varstvo spomenikov 32, 177–178. STRMČNIK GULIČ, M. 1991, Jurišna vas – Ančnikovo gra­dišče. – Varstvo spomenikov 33, 218–220. STRMČNIK GULIČ, M. 1992, Jurišna vas – Ančnikovo gra­dišče. – Varstvo spomenikov 34, 226–230. STRMČNIK GULIČ, M. 1995, Jurišna vas. – Varstvo spome­nikov 35 (1993), 101–103. STRMČNIK GULIČ, M. 1997, Mariborsko - bistriško območje v poznorimski dobi (Der Raum Maribor und Bistrica in spätrömischer Zeit). – Arheološki vestnik 48, 269–288. STRMČNIK GULIČ, M., 2012, Ančnikovo gradišče – nekaj novosti o naselbinski podobi (Ančnikovo gradišče – New Insights into the Form of the Settlement). – V / In: B. Migotti, P. Mason, B. Nadbath, T. Mulh (ur. / eds.), Scripta in Honorem Bojan Djurić, Monografije CPA 1, 205–214. STRMČNIK GULIČ, M., 2020, Slovenska Bistrica. - V / In: J. Horvat, I. Lazar, A. Gaspari (ur. / eds.), Manjša rimska naselja na slovenskem ozemlju / Minor Roman settlement in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 339-348. STRMČNIK GULIČ, M., S. CIGLENEČKI 2003, Ančnikovo gradišče pri Jurišni vasi. Poznoantična in zgodnjesrednjeveš­ka naselbina na Pohorju / Late Antique and Early Medieval settlement on Pohorje. – Slovenska Bistrica. STRMČNIK GULIČ, M., M. KAJZER CAFNIK 2010, Jurišna vas – Ančnikovo gradišče. – Varstvo spomenikov. Poročila 46 (2009–2010), 106–107. STRMČNIK GULIČ, M., P. PREDAN 2006, Kalše - rimsko gro­bišče. - Varstvo spomenikov. Poročila 39–41(2000–2004), 68–69. Zvezdana Modrijan Znanstvenoraziskovalni center SAZU Inštitut za arheologijo Novi trg 2 SI-1000 Ljubljana zvezda@zrc-sazu.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 363–385 PRAETORIUM LATOBICORUM – TREBNJE Uroš BAVEC Izvleček Naselbina Praetorium Latobicorum je ležala ob itinerarski cesti na območju današnjega Trebnjega. Lokalno prebivalstvo so bili keltski Latobiki. Epigrafski spomeniki pričajo o beneficiarski postaji in svetišču dolihenskega Jupitra. V zgodnje­cesarsko obdobje sodita dve stavbi, od katerih je bila ena livarska delavnica. V pozni rimski dobi so bile ob tlakovani cesti nanizane dolge hiše z obrtniškimi delavnicami, ki so povezane s tekstilno obrtjo. Dobro je poznano zahodno grobišče. V bližini je bilo odkritih več vil in kamnolom. Ključne besede: Zgornja Panonija, Trebnje, Praetorium Latobicorum, rimska doba, cesta, naselbina, beneficiarska postaja, grobišče, dolga hiša, livarska delavnica, tekstilna obrt, svinčena etiketa Abstract Praetorium Latobicorum lay on the main eastbound Roman road from Emona, in the area of present­day Trebnje. Its local inhabitants were the Celtic Latobici. Epigraphic evidence reveals that a statio of the beneficiarii and a temple of Jupiter Dolichenus stood the settlement. Its Early Imperial remains include two buildings, one of which was a foundry. In the Late Roman period, strip­houses with workshops connected to textile lined the road. The settlement is associated with cemeteries, of which part of the west cemetery has been investigated, but also a countryside villa and a stone quarry in the vicinity. Keywords: Pannonia Superior, Trebnje, Praetorium Latobicorum, Roman period, road, settlement, statio of the beneficiarii, cemetery, strip­house, foundry, textile, lead tag https://doi.org/10.3986/9789610502586_20 Sl. 1: Rimskodobne naselbine in poti v okolici Trebnjega. Fig. 1: Roman­period settlements and communications around Trebnje. LEGA Naselbina Praetorium Latobicorum je ležala ob itinerarski cesti, ki je Emono povezovala s Siscijo.1 V Antoninskem itinerariju iz začetka 3. st. je zapisana kot Praetorium Latovicorum in je od Emone oddaljena 34 milj, od Nevioduna pa 31 milj.2 Na Tabuli Peutingeriani je označena kot Ad Protorium, ki je oddaljena 14 milj od postaje Acervo in 16 milj od postaje Crucium.3 Nastanek imena bi lahko bil povezan s prisotnostjo vojske.4 Po navedenih razdaljah, poteku ceste in številnih naselbinskih najdbah je mogoče locirati Praetorium Latobicorum na območje Trebnjega.5 Čez Trebnje, ki leži v dolini reke Temenice, je potekala najkrajša naravna povezava iz Ljubljanske kotline proti Brežiškim vratom in verjetno hkrati ena najzgodnejših cestnih povezav 1 Müllner 1879, 93–106; Premerstein, Rutar 1899, 17–37; Truhlar 1975, 101; Šašel 1975a, 75; Šašel Kos 1997, 26–33; Lovenjak 1998, 333–375; 2006, 39–40. 2 It. Ant. 259, 13. 3 Tab. Peut. IV, 2. 4 Fodorean 2017, 344. 5 Lovenjak 1998, 223–225; Slabe 1993, 32; Šašel 1975b, 231. LOCATION The Roman settlement called Praetorium Lato­bicorum was located on the main road connecting Emona and Siscia.1 In the Antonine Itinerary from the beginning of the 3rd century, it is noted as Praetorium Latovicorum, located 34 Roman miles from Emona and 31 miles from Neviodunum.2 It is marked as Ad Protorium in the Tabula Peutingeriana and located 14 miles from the post of Acervo and 16 miles from the post of Crucium.3 Its name suggests that its origins may be connected with a military presence.4 The distances in the two road maps, the location of the Roman roads and the numerous habitation traces show that Praetorium Latobicorum lies in the area of present­day Trebnje, in the valley of the River Temenica.5 It is here that led the shortest natural communication 1 Müllner 1879, 93–106; Premerstein, Rutar 1899, 17–37; Truhlar 1975, 101; Šašel 1975a, 75; Šašel Kos 1997, 26–33; Lovenjak 1998, 333–375; 2006, 176. 2 It. Ant. 259, 13. 3 Tab. Peut. IV, 2. 4 Fodorean 2017, 344. 5 Lovenjak 1998, 223–225; Slabe 1993, 32; Šašel 1975b, 231. Sl. 2: Trebnje, območja raziskav. Fig. 2: Trebnje, investigated areas. 1 Župna cerkev, 2 Pristava, 3 Benečija, 4 Trebanjski grad, 5 Trebnje – industrijska cona, 6 Dolenjska ulica, 7 DSO, 8 Levji dvorec, 9 Rimski dvori, 10 Rimska cesta, 11 Mercator center (Kartografski vir / Map source: http://gis.arso.gov.si/evode/profile.aspx?id=atlas_voda_Lidar@Arso) bodoče Panonije.6 Obenem se tu križajo lokalne poti, ki vodijo proti dolinam Save in Krke (sl. 1). ZGODOVINA RAZISKAV Simon Rutar je naselbino Praetorium Latobicorum lociral na območje vasi Štefan pri Trebnjem (sl. 1),7 vendar se danes zdi, da je opazoval sledove grobnih parcel grobišča na Pristavi pri Trebnjem. Tone Knez je na Pristavi leta 1959 ugotovil žgane rimske gobove (sl. 2: 2). Leta 1960 in 1967 je odkril še 18 grobov in grobiščepostavil na vzhodni rob domnevne naselbine v Štefanu.8 Trije grobovi so bili izkopani leta 1970 (parc. št. 232 k. o. 6 Za vio militaris Šašel 1977, 459, 466; Lovenjak 2006, 39–40. 7 Premerstein, Rutar 1899, 25; podobno Knez 1969, 131 in Gabrovec 1956, 30, 40. 8 Knez 1969, 120–132; v to število niso šteti leta 1933 odkrita grobna žara, ki jo navaja, in grobovi iz nekaterih sta­rejših poročil. from the Ljubljana Basin towards the Brežice Gate that was probably one of the earliest Roman roads in the province of Pannonia.6 It is also the junction of local roads leading towards the valleys of the Rivers Sava and Krka (Fig. 1). HISTORY OF RESEARCH Simon Rutar located Praetorium Latobicorum in the area of the village of Štefan near Trebnje (Fig. 1),7 probably based on the traces of burial enclosures he had observed at adjacent Pristava near Trebnje. In 1959, Tone Knez recorded Roman cremations at Pristava (Fig. 2: 2). He unearthed further eighteen burials in 1960 and 1967, and noted that the cemetery extended along the eastern edge of the settlement he also presumed 6 For the via militaris, see Šašel 1977, 459, 466; Lovenjak 2006, 39–40. 7 Premerstein, Rutar 1899, 25; similarly Knez 1969, 131 and Gabrovec 1956, 30, 40. Štefan),9 med letoma 1972 in 1979 pa je bilo izkopanih še 111 grobov.10 Leta 2002 so na skrajnem zahodnem robu nekropole na Pristavi (Rimska cesta 24) odkrili še dvanajst novih grobov, med katerimi je bilo kar 5 gro­bov skeletnih (med 132 objavljenimi sta bila skeletna le dva).11 Dodatne raziskave v letih 2007, 2010 ter 2014 in 2015 so potrdile domneve, da se rimsko grobišče širi tudi po njivah južno in severno od že raziskanih površin.12 Med rimskodobnimi naselbinskimi ostanki na območju stanovanjskega objekta Levji dvorec sta bila odkrita en žgani in en skeletni grob (sl. 2: 8).13 Po legi grobišča in na podlagi najdišč napisnih kam­nov sta Jaroslav Šašel in Marijan Slabe jedro naselbine postavila na lokacijo trebanjske župne cerkve (sl. 2: 1).14 Za župno cerkvijo, na parc. št. 236 k. o. Trebnje, so bili odkriti rimski gradbeni ostanki in kloaka, poleg tega pa še zgodnjesrednjeveško grobišče iz 10. st. (sl. 2: 1).15 Leta 1988 je bila s pomočjo geofizikalnih raziskav in sondiranj ugotovljena poselitev na območju Benečije, dobro ohranjeni zidani arhitekturni ostanki, ki jih od preostale naselbine loči močvirna ravnica in tok reke Temenice, so bili interpretirani kot ostanki vile rustike (sl. 2: 3).16 Leta 1992 in 1993 so na Trebanjskem gradu poteka­le sondažne raziskave, s katerimi so bili pod razvojnimi fazami gradu iz 15. st. dokazani rimskodobni in verjetno celo prazgodovinski naselbinski ostanki (sl. 2: 4).17 Zahodni rob rimske naselbine je bil izkopavan v letih 2006, 2007 in 2009. Na 6566 m2 skupno raziskane površine so bili odkriti ostanki različno ohranjenih stavb, ki jim je bilo ponekod mogoče določiti funkcijo (sl. 2: 8,9; 4).18 Leta 2014 je bilo v jedru Trebnjega (Rimska cesta) ob gradbenih delih odkritih še več dobro ohra­njenih zidov in naselbinskih plasti, zamejenih s širino gradbenega izkopa polovice današnjega cestišča (pribl. 2,5 m), le v majhnem delu (na dolžini pribl. 11 m) je bil izkop razširjen tudi na površino pločnika. Tu odkrite zgradbe so primerljive z ostanki na zahodnem robu naselja (sl. 2: 10; 4).19 Raziskave na skrajnem vzhodnem robu današnjega Trebnjega (v letih 2001, 2005, 2009) niso potrdile obstoja vzhodnega mestnega grobišča ali razkrile arhitekturnih ostankov, ki bi jih smeli pripisati rimski naselbini (prim. 9 Oman 1974, 139–142. 10 Slabe 1993, 5. 11 Josipovič 2002, 16; Bavec 2006a, 159. 12 Bavec, Murko, Predan 2007; Kovač 2010; Tiran 2015; Brečić, Jovanović 2018. 13 Bavec 2007, 225. 14 Šašel 1975b, 231; Slabe 1993, 32. 15 Knez 1966, 509–510, 512. 16 Breščak, Waters 1990; Breščak 1990. 17 Mason 1993, 2. 18 Bavec 2007, 224 ss; 2008, 258 ss; 2010, 383 ss. 19 Brečić, Jovanović 2018; Jovanović 2018b, tloris in os­ novne informacije mi je prijazno odstopil izvajalec raziskav S. Olić (Arhos d. o. o.). at Štefan.8 Three graves came to light in 1970 (LotNo. 232, Štefan cadastral municipality),9 further 111 between 1972 and 1979.10 In 2002, twelve new burials were found at the westernmost edge of the cemetery at Pristava (Rimska cesta 24) that include five inhuma­tions (only two of the 132 graves published thus far are inhumations).11 Investigations in 2007, 2010, 2014 and 2015 have confirmed the suppositions that the Roman cemetery extended across the fields south and north of the already investigated areas.12 Also, the Roman­period remains excavated in advance of the construction of the Levji dvorec residential complex include a cremation and an inhumation burial (Fig. 2: 8).13 The location of the cemetery and the findspots of two inscriptions led Jaroslav Šašel and Marijan Slabe to presume that the core of the Roman settlement lay at the parish church in Trebnje (Fig. 2: 1).14 Roman building and sewer remains, as well as an early medieval cemetery from the 10th century were found behind the parish church, on Lot No. 236 in the Trebnje cadastral municipality (Fig. 2: 1).15 The geophysical investigations and trial trench­ing in 1988 have also established habitation remains at Benečija, while the well preserved masonry remains separated from the rest of the settlement by a marshy plain and the River Temenica have been interpreted as the remains of a Roman countryside villa (Fig. 2: 3).16 In 1992 and 1993, trial trenching at Trebnje Castle revealed habitation remains from the Roman period and possibly even prehistory under the castle constructions from the 15th century (Fig. 2: 4).17 The western edge of the Roman settlement was excavated in 2006, 2007 and 2009. The 6566 m2 large area of investigations revealed the remains of buildings surviving to different degrees, some of which could be determined as to their function (Figs. 2: 8,9; 4).18 More well preserved walls and habitation layers were exca­vated in 2014 in the centre of Trebnje (Rimska cesta), in advance of construction works in the width of one half of the modern road through Trebnje (ca. 2.5 m). Excavations only widened onto the pavement in a short, roughly 11 m long section, which revealed a building 8 Knez 1969, 120–132; the number does not include the urn found in 1933 and the graves from some of the earlier reports. 9 Oman 1974, 139–142. 10 Slabe 1993, 5. 11 Josipovič 2002, 16; Bavec 2006a, 159. 12 Bavec, Murko, Predan 2007; Kovač 2010; Tiran 2015; Brečić, Jovanović 2018. 13 Bavec 2007, 225. 14 Šašel 1975, 231; Slabe 1993, 32. 15 Knez 1966, 509–510, 512. 16 Breščak, Waters 1990; Breščak 1990. 17 Mason 1993, 2. 18 Bavec 2007, 224 ff; id. 2008, 258 ff; id. 2010, 383 ff. sl. 2: 5,7).20 Hkrati nam ostanki vogala rimske stavbe in ruševin na Dolenjski ulici kažejo, do kod bi utegnila segati sklenjena poselitev na vzhodu (sl. 2: 6).21 Nove indice o jugovzhodnem robu rimske naselbine prinašajo še nedokončane raziskave nad trebanjsko železniško postajo.22 PRAZGODOVINA Na lokaciji Rimski dvori so bili ugotovljeni štirje vkopi za vertikalne sohe in odlomki značilne grafitne keramike (sl. 2: 9; 5: A). Menimo, da gre za ostanke latenskodobnega objekta (objekt A), ki je bil odstranjen ob gradnji rimskodobnega cestišča in mlajših stavb.23 RIMSKA DOBA CESTA Na širšem območju Trebnjega so bili odkriti trije miljniki, ki pa nimajo ohranjenih napisov. Prvi naj bi bil odkrit v Temenici pri Kamnem potoku, drugi se je nahajal pri cerkvi sv. Petra pri naselju Jezero (sl. 1),24 tretji pa je bil najden na vzhodnem robu Trebnjega (sl. 2: 11; 3).25 Ni znano, ali ima toponim rimska cesta, s katerim se srečamo tako v Trebnjem kot tudi na Pristavi, svoj izvor v ustni tradiciji. Morda kaže nekdaj vidne ostanke prastare cestne povezave, ki je ponekod dokazljiva z arheološkimi podatki.26 Pod današnjim cestiščem, ki gre skozi Pristavo, se v dolžini okoli 51 m omenja bela proga kamenja. Ta proga je bila pripisana rimski cesti.27 Tudi “georadarska raziskava” je pod nasutji obstoječe vaške ceste nakazala obstoj od 0,8 m do 0,9 m debele plasti močno utrjene­ga drobljenca, ki je bila kasneje s strojno izkopanimi testnimi jarki še dodatno potrjena.28 Podobno nasutje beležimo na večji globini tudi na lokaciji Rimski dvori (sl. 4-6). Tu je potekala vzporedno z današnjo glavno cesto, ki vodi v Trebnje iz smeri Štefana. Gre za zelo dobro utrjena nasutja, ki jih tvorita spodnja plast grušča 20 Bavec 2006b; Udovč 2010; Rutar et al. 2012. 21 Čaval, Breščak 2003, pril. 2–4. 22 Jovanović 2018a. 23 Bavec 2010, 384 ss; objekt je označen za prazgodovin­skega bolj na podlagi stratigrafske situacije kot skromnih ma­terialnih ostankov. 24 Lovenjak 2006, 41–46; 1998, 342-343; miljnik pri sv. Petru je izgubljen. 25 Breščak 2008, 286–287. 26 Oman 1974, 140 ss; Bavec 2006a, 159; 2010, 384, sl. 1,2. 27 Pojavlja se od hišne št. 8 proti zahodu: Oman 1974, 140 ss. 28 Josipovič 2002; Bavec 2006a, 159. similar to those at the western edge of the settlement (Figs. 2: 10; 4).19 The 2001, 2005 and 2009 investigations in the east­ernmost part of modern Trebnje unearthed no burial or habitation remains (cf. Fig. 2: 5,7).20 The corner of a Roman building and building debris in the street of Dolenjska ulica indicate the eastern edge of the densely occupied area (Fig. 2: 6).21 New findings are also coming to light in the area above the railway station at Trebnje, at the south­eastern edge of the Roman settlement.22 PREHISTORY Four postholes and sherds of graphite ware have been excavated at the Rimski dvori site (Figs. 2: 9; 5: A). They are believed to be the remains of a building from the La Tene period (Building A) removed during the construction of the Roman road and later buildings.23 ROMAN PERIOD ROAD Three milestones were recovered in the wider area of Trebnje, though without surviving inscriptions. The first was reportedly discovered in the River Temenica near its confluence with the Kamni potok, the second stood at the church of St Peter near the village of Jezero (Fig. 1),24 the third one came to light at the eastern edge of Trebnje (Figs. 2: 11; 3).25 There is a street named Rimska cesta (‘Roman road’ in translation) in both Trebnje and Pristava. It is a toponym that may refer to the once visible remains of an ancient road. Traces of a road have actually been archaeologically recorded in several locations.26 A roughly 51 m long ‘white strip of stones’ has been reported under the present­day road through Pristava and interpreted as a Roman road.27 Georadar investiga­tions also indicated the existence of a 0.8 to 0.9 m thick layer of compact rubble under the layers of the road 19 Brečić, Jovanović 2018; Jovanović 2018b. S. Olić (Ar­hos d.o.o.) kindly provided the plan and basic information. 20 Bavec 2006 b; Udovč 2010; Rutar et al. 2012. 21 Čaval, Breščak 2003, Apps. 2–4. 22 Jovanović 2018a. 23 Bavec 2010, 384 ff; the building is interpreted as pre­historic based on stratigraphic evidence rather than the scant material remains. 24 Lovenjak 2006, 41–46; 1998, 342-343; the milestone from the church is now lost. 25 Breščak 2008, 286–287. 26 Oman 1974, 140 ff; Bavec 2006a, 159; 2010, 384, Figs. 1, 2. 27 It has been recorded west of the house at Pristava 8: Oman 1974, 140 ff. in apnenčastih lomljencev ter zgornje, do 0,8 m debelo peščeno nasutje. Cestišče je bilo na močvirni podlagi temeljeno z lesenimi (neohranjenimi) piloti. Cesto je spremljal jarek, obložen z večjimi apnenčastimi lomljen­ci, ohranjen v dolžini pribl. 15 m (sl. 5; 7). 29 Morda smemo s cestiščem ali obcestnim tlakom povezati tudi okoli 7,5 m širok pas tlakovane površine na lokaciji Levji dvorec, ki ga zaznamuje presenetljivo število rimskih novčnih najdb od 1. do 5. st. Drenažni jarki tod niso bili ugotovljeni (sl. 16; 17). NASELBINA Naselbina se širi na levem bregu reke Temenice. Njen zahodni rob določa grobišče na Pristavi, vzhodni rob ni bil odkrit. Najbolje so poznani naselbinski ostanki na zahodnem delu današnjega Trebnjega. 29 Bavec 2010, 384 ss, sl. 1; 2. Sl. 3: Trebnje. Miljnik z vzhodnega dela Trebnjega – danes Mercator center. Fig. 3: Trebnje. Milestone from the eastern part of Trebnje – today Mercator Centre. (Foto / Photo: arhiv / archives ZVKDS OE Novo mesto) through Pristava, which was later confirmed during trial trenching.28 A similar deposit, though at a greater depth, has been recorded at the Rimski dvori site (Figs. 4-6). The road here ran parallel with the modern main road leading to Trebnje from the west. It had a compact bedding of gravel and limestone rubble covered by an up to 0.8 m thick surface layer of sand. The road was strengthened with wooden piles in marshy areas (the actual piles have not survived). Also unearthed was a roadside ditch lined with large pieces of limestone and survived in the length of ca. 15 m (Figs. 5; 7). 29 The 7.5 m wide strip of paved surface at the Levji dvorec site, which revealed surprisingly numerous Roman coins from the 1st–5th centuries, may also represent a road or a roadside pavement; no drainage ditches have been established at this site (Figs. 16; 17). SETTLEMENT It is located on the left bank of the River Temenica. It reached in the west to the cemetery at Pristava, while its eastern edge has as yet not been determined. The best surviving habitation remains have come to light in the west part of Trebnje. At the Rimski dvori site (Rimska cesta 18b–c), archaeologists unearthed the remains of at least four masonry buildings along a Roman road with a roadside ditch (Figs. 4-7). Building B had several rooms, though its original size and layout is not known as it had already been partially destroyed upon discovery (Fig. 5). Its longer side was parallel with the road and measured more than 16 m, while its shorter side measured at least 12 m. It was probably constructed in the early 1st and abandoned in the 2nd century.30 Building C was long and narrow, with a sub­rectangular plan (Figs. 4; 8). It faced the road with its shorter, roughly 8 m long side. Two similar, though less well preserved buildings were found on either side and only 0.8 m away from it. Building C and the building to the west of it were furnished with furnaces or hearths. 28 Josipovič 2002; Bavec 2006a, 159. 29 Bavec 2010, 384 ff, Figs. 1; 2. 30 Building B was probably emptied before being aban­doned, as the number of recovered small finds is negligible in comparison with nearby Building C. Sl. 4: Trebnje. Gradbeni ostanki na lokacijah Rimski dvori, Levji dvorec in Rimska cesta. M. = 1:2000. Fig. 4: Trebnje. Roman­period remains at the Rimski dvori, Levji dvorec and Rimska cesta sites. Scale 1 : 2000. Na lokaciji Rimski dvori (Rimska cesta 18b–c) so bili najdeni ostanki rimske ceste in obcestnega jarka ter vsaj štirih zidanih stavb (sl. 4-7). Stavba B je imela več prostorov, a zaradi delnega uničenja ni bila v celoti izkopana (sl. 5). Njena daljša stranica, ki je potekala vzporedno s cesto, je bila dolga več kot 16 m. Stavba je bila široka vsaj 12 m. Verjetno je bila zgrajena na začetku 1. st., v 2. st. pa opuščena.30 Stavba C je bila dolga in ozka ter nepravilnega, v osi nekoliko zamaknjenega pravokotnega tlorisa (sl. 4; 8). Z ožjo, okoli 8 m široko stranico je bila obrnjena proti cesti. Vzporedno in tesno ob njej (razmak le okoli 0,8 m) sta stali vsaj še dve podobni stavbi, ki sta bili precej slabše ohranjeni. V stavbi C, ob njeni severni steni in v stavbi zahodno od nje so bili odkriti ostanki peči oz. ognjišč. Severno od stavbe C je ležala tudi velika jama (okoli 1,2 m notranjega premera, globina 0,7 m), katere dno je bilo obloženo z apnenčastimi lomljenci in z nepropustno močno rdečkasto ilovico. Menimo, da gre za zbiralnik vode (sl. 8: 2; 9). V polnilu ene izmed odpadnih jam (SE 164/165; sl. 8: 165), ki jo je presekal zid kasneje zgrajenega objekta zahodno od stavbe C, je bila odkrita italska in galska sigilata iz 1. in 2. st., kar je edini tako zgoden sklop najdb. Stavba B je bila verjetno pred opustitvijo izpraznje­na, saj je količina drobnih najdb in lončenine v primerjavi z bližnjim objektom C zanemarljiva. North of Building C was a large pit, measuring around 1.2 m in interior diameter and 0.7 m in depth. Its bot­tom was lined with limestone rubble and impervious reddish loam, suggesting it was used for water collection (Figs. 8: 2; 9). The fill of one of the refuse pits (SE 164/165; Fig. 8: 165), disturbed by the wall of the later building west of Building C, held Italian and Gaulish sigillata ware from the 1st and 2nd centuries; it represents the only artefact assemblage of such an early date. Other small finds from the area of Building C are later and date it to the 4th and early 5th centuries (Fig. 10). They include imported North African sigillata bowls (Hayes 45A, 50A) and Eastern Mediterranean LR 1 amphorae.31 Of the 610 recovered coins, 26 date from the 1st to the 3rd century, others to the 4th and early 5th centuries; the latest is the coin of Honorius (408–423) (Fig. 8).32 A rectangular box (SE 125/126) had been buried at the foundations of the western building. The hoard contained the coin of Constantius II minted in 350, a bronze statuette of Hercules, a bronze bell, a whetstone, 31 Nives Varjačić analysed the imports under the men­torship of Verena Perko. 32 Alenka Miškec from the National Museum of Slove­nia determined the coins (report kept in the archives of the Institute for the Protection of Cultural Heritage of Slovenia, Regional Office Novo mesto). Sl. 5: Trebnje. Rimski dvori: stavbi A in B ter cesta in obcestni jarek (lokacija na sl. 4). M. = 1:200. Fig. 5: Trebnje. The Rimski dvori site. Buildings A, B, the adjacent road and roadside ditch (location Fig. 4). Scale 1 : 200. Sl. 6: Trebnje. Rimski dvori. Ohranjeno cestišče, pogled proti Sl. 7: Trebnje. Rimski dvori. Obcestni jarek, pogled proti vzhodu. (Foto: arhiv ZVKDS OE Novo mesto) zahodu. (Foto: arhiv ZVKDS OE Novo mesto) Fig. 6: Trebnje. The Rimski dvori site. Surviving remains of the Fig. 7: Trebnje. The Rimski dvori site. Roadside ditch, westward road, eastward view. (Photo: archives ZVKDS OE Novo mesto) view. (Photo: archives ZVKDS OE Novo mesto) 370 Sl. 8: Trebnje. Rimski dvori: stavba C in okolica (lokacija na sl. 4). Lega ognjišč, svinčenih etiket, stilusov in novcev. M. = 1:200. Fig. 8: Trebnje. The Rimski dvori site. Building C and its vicinity (location Fig. 4) with marked locations of the hearths, lead tags, styli and coins. Scale 1 : 200. 371 Sl. 9: Trebnje. Rimski dvori, stavba C. Presek domnevnega Sl. 10: Trebnje. Rimski dvori, stavba C. Fibuli s čebuličastimi zbiralnika za vodo (sl. 8: 2). zaključki, bron. Fig. 9: Trebnje. The Rimski dvori site. Cross section of the Fig. 10: Trebnje. The Rimski dvori site. Bronze crossbow presumed water collection pit at Building C (Fig. 8: 2). brooches from Building C. (Foto / Photo: arhiv /archives ZVKDS OE Novo mesto) (Foto / Photo: arhiv /archives ZVKDS OE Novo mesto) Ostale drobne najdbe z območja stavbe C in okoli nje so mlajše in jih opredeljujemo v 4. in v začetek 5. st. (sl. 10). Prepoznani so bili importi severnoafriških sigi­latnih skodel (Hayes 45A, 50A) in vzhodnosredozemske amfore LR 1.31 Med 610 novci jih 26 sodi v čas od 1. do 3. st., ostali pa so iz 4. in iz začetka 5. st. Najmlajši med njimi je Honorijev novec (408–423) (sl. 8).32 V objektu zahodno od stavbe C je bil vkopan zoglenel pravokoten zabojček (SE 125/126; sl. 8: 1), ki 31 Analizo importov je pod mentorstvom Verene Perko izvedla Nives Varjačić. 32 Novce je določila Alenka Miškec, Narodni muzej Slo­venije (poročilo hrani arhiv ZVKDS OE NM). an iron key, the handle of an oil lamp and sherds of four coarseware jars (Figs. 8: 1; 11). The wider area of Building C revealed eight com­mercial lead tags, pieces of lead used to make such tags, as well as seven styli (Figs. 8; 12–15). Two of the tags bear the name of the purchaser, the other side of one of them even the price (Figs. 13; 14). In connection with the hearth and the water collection pit, these small finds suggest a connection with textile. The Levji dvorec site (Rimska cesta 19 c–e) revealed heavily damaged drystone foundations of several build­ings (Figs. 4; 16).33 As at Rimski dvori, it appears that 33 Bavec 2007, 225: the report states an erroneous date je poleg novca Kostancija II. iz leta 350, vseboval tudi bronasti kipec Herkula, bronast zvonček, kamniti brus, železen ključ, ročaj oljenke in ostanke štirih kuhinjskih loncev (sl. 11). Na širšem območju stavbe C je bilo najdenih osem svinčenih etiket za označevanje “blaga”, svinec, iz katerega so etikete izdelovali, in sedem stilusov (sl. 8; 12–15). Na dveh etiketah je vpraskano ime naročnika, na elongated and closely spaced buildings lined a street. The modern community infrastructure shaft, but even more so the previous agricultural land use have destroyed the buildings to the degree that their layouts can only rough­ly be surmised from the remaining foundation stones and refuse pits. The walls continued to the north beyond for the ‘smithing­casting facility’. drugi strani ene od njih celo cena (sl. 13; 14). Te drobne najdbe nam v navezavi z ognjiščem in zbiralnikom za vodo morda kažejo, da je na območju stavbe C delovala suknarska delavnica. Na lokaciji Levji dvorec (Rimska cesta 19c–e) so bili odkriti močno poškodovani suhozidni temelji več objektov (sl. 4; 16).33 Kot pri Rimskih dvorih se tudi tu zdi, da so bile podolgovate stavbe tesno nanizane ob cesti. Sodobni infrastrukturni jarek, zlasti pa recentna poljedelska izraba tal sta stavbe uničila do te mere, da lahko njihove tlorise na podlagi “izropanih zidov” in odpadnih jam večinoma le še slutimo. Zidovi se proti severu nadaljujejo zunaj izkopnega polja in potekajo pravokotno na smer današnje Rimske ceste. Na severo­zahodnem delu izkopa se je ohranil del stavbe z več prostori (ohranjena velikost 20 × 15 m; sl. 16; 17). Čas nastanka in delovanja stavbe postavljamo, na osnovi odlomkov amfor (Dressel 43, 6B) in zlasti fine namiz­ne keramike iz padskih in galskih delavnic, v drugo polovico 1. in v 2. st. (sl. 16; 18; 19). Ob stavbi so ležali ostanki livarske peči z bronasto talino, v stavbi in okoli nje pa ostanki livarskih lončkov, dvajset brusnih kamnov ter jame z žganino (sl. 17; 20; 21). V eni izmed jam je bil odkrit zanimiv sklop najdb: vilice za meso, utež za tehtnico, železna zanka in keramični krožnik (sl. 17: A; 22). Stavba verjetno predstavlja obcestno delavnico, ki je izvajala manjša livarska dela v bronu, npr. popravila kovinske opreme. Večje število novcev, odkritih v ne­posredni bližini, in odlomki fine namizne keramike pa dajo slutiti, da je tu občasno delovala tudi cestna stojnica oz. mala trgovinica.34 33 Bavec 2007, 225: v poročilu je kovaško-livarski objekt napačno datiran (prim. besedilo v nadaljevanju). 34 Thiel 2001, 91. 343) from Building C. Scale 1 : 2. the excavation area and perpendicular to the modern street of Rimska cesta. Part of a multi­room building (surviving size 20 × 15 m; Figs. 16; 17) was unearthed in the north­western part of the excavation area. The sherds of Dressel 43 and 6B amphorae, and particularly the sherds of the Padana and Gaulish finewares suggest that the building dates to the second half of the 1st and the 2nd century (Figs. 16; 18; 19). Next to the building were the remains of a foundry furnace with bronze melt, while fragments of crucibles, twenty whetstones and pits filled with burnt remains came to light inside and around the building (Figs. 17; 20; 21). One of these pits contained an interesting assemblage of artefacts: a meat fork, a balance weight, an iron hook and a ceramic plate (Figs. 17: A; 22). The building was probably a roadside workshop that performed minor casting work in bronze, for example repairs. A large number of coins, found in the immediate vicinity, as well as the fineware sherds indicate the possibility of a roadside stand or small shop occasionally active in this area.34 The finds of African sigillata show that the wider area of the building was still in use in the early 5th century (Fig. 23). Also found here were sheep shears (Fig. 24). Recently, Roman building remains have been unearthed in the centre of Trebnje, under the modern street of Rimska cesta (Figs. 2: 10; 4).35 In spite of a nar­row excavation area, archaeologist have been able to identify four buildings measuring between 5 and 10 m in width, also closely spaced and located north of the main Roman road through the settlement. The location of the Roman road in this section does not correspond with the modern main road coming to Trebnje from the west (Rimska cesta). Similarly as for Building C at Rimski dvori, a coin of Honorius (408–423) represents the latest coin find.36 Weavers shears (Fig. 25) may in­dicate textile manufacture or another artisanal activity in this area as well.37 The remains associated with the Roman settlement excavated east of here are less revealing. They include a Roman drainage ditch and the remains of the underfloor heating of a large building excavated south of the parish 34 Thiel 2001, 91. 35 Kozlevčar 2015, 18; Brečić, Jovanović 2018. 36 Brečić, Jovanović 2018; Alenka Miškec from the Na­tional Museum of Slovenia determined the coins (report kept in the archives of the Institute for the Protection of Cultural Heritage of Slovenia and the Arhos d.o.o. company). 37 For the shears cf. Römer­Martijnse 1990, 242, Pl. 24: b. Sl. 16: Trebnje. Levji dvorec (lokacija na sl. 4). Ostanki temeljev in tlakov. Razširjenost amfor in fine namizne keramike. M. = 1:400. Fig. 16: Trebnje. The Levji dvorec site (location Fig. 4). Remains of buildings and paved surfaces with marked distribution of amphorae and tableware. Scale 1 : 400. 375 Sl. 17: Trebnje. Levji dvorec, detajl. Tloris livarske delavnice in domnevnega cestišča. Lega livarske peči, talilnih lončkov in brusnih kamnov. M. = 1:200. Fig. 17: Trebnje. The Levji dvorec site, detail. Foundry and presumed road with marked locations of the furnace, crucibles and whetstones. Scale 1 : 200. 376 282) from the foundry. Scale 1 : 3. >> Sl. 24: Trebnje. Levji dvorec. Škarje za striženje volne (PN 215), železo. M. = 1:2. Fig. 24: Trebnje. The Levji dvorec site. Iron sheep shears (PN 215). Scale 1 : 2. Najdbe afriške sigilate kažejo, da je bil širši prostor okoli livarske delavnice v rabi še ob začetku 5. st. (sl. 23). Najdene so bile tudi škarje za striženje volne (sl. 24). V središču Trebnjega, pod današnjo Rimsko cesto, so bili nedavno tudi najdeni rimski arhitekturni ostanki (sl. 2: 10; 4).35 Kljub ozkemu izkopnemu polju so izko­pavalci uspeli določiti štiri objekte, široke med 5 in 10 m, ki so podobno kot objekti na lokaciji Rimski dvori stali tesno drug ob drugem, severno od glavne rimske komunikacije. Njen potek tu vsekakor ne sovpada s pote­kom današnje zahodne mestne vpadnice – Rimske ceste. Podobno kot pri stavbi C na Rimskih dvorih predstavlja Honorijev novec (408–423) tudi tu najmlajšo novčno najdbo.36 Tkalske škarje bi morda lahko nakazovale suk­narsko/obrtno dejavnost tudi na tem območju (sl. 25).37 Vzhodneje so sledovi, ki jih povezujemo z rimsko naselbino, manj izpovedni. Južno od župne cerkve (parc. 236 k. o. Trebnje) so bili najdeni rimski odtočni kanal in ostanki hipokavsta nekega večjega objekta (sl. 2: 1).38 Sklenjena poselitev je morda segala do Dolenjske ulice, kjer so bili odkriti ostanki antične arhitekture (sl. 2: 6).39 Objekt C na Rimskih dvorih, ostanke stavb pod današnjo Rimsko cesto in pri Levjem dvorcu sme­mo označiti za poseben arhitekturni tip, ki ni vezan na lokalno gradbeno tradicijo. Gre za t. i. dolge hiše (nem. Streifenhaus, angl. strip-house), ki se pojavljajo v civilnih in vojaških vikusih, pa tudi v kolonijah in municipijih, skratka povsod, kjer se med seboj tesno 35 Kozlevčar 2015, 18; Brečić, Jovanović 2018. 36 Brečič, Jovanović 2018; novce je v poročilu določila Alenka Miškec, Narodni muzej Slovenije (Poročilo hrani ar­hiv ZVKDS OE NM in Arhos, d. o. o.) 37 Za škarje prim. Römer­Martijnse 1990, 242; t. 24: b. 38 Knez 1966, 507–516. Neposredno pod spomenikom “Zamolčanim žrtvam bratomorne vojne” je bil v času posta­vitve odkrit dobro ohranjen estrih z ostanki stebričkov za hipokavst. 39 Čaval, Breščak 2003, pril. 2–4. church (Lot No. 236, Trebnje cadastral municipality; Fig. 2: 1).38 The densely inhabited area may have reached to the street of Dolenjska ulica, where Roman architectural remains also came to light (Fig. 2: 6).39 Building C at Rimski dvori, the remains of the buildings under the modern Rimska cesta and those at the Levji dvorec site may be seen as a particular type of architecture not related to the local building tradition. They are strip­houses (Streifenhaus in German), which occur in civilian and military vici, but also coloniae and municipii, i.e. in places where habitation is closely associ­ated with trade and crafts.40 Strip­houses are constructed in densely­spaced groups with the short side facing the road or street. It has even been suggested that the price of renting such a building (or possibly the tax on it) may have depended on the width of the main façade. The front part of such a building, serving as a shop, usually had a wide entrance or was even open with a portico, while the workshop/storeroom and living quarters were located at the back. Such a layout can also be observed in Building C.41 SURROUNDING AREA Another habitation area, covering almost 4 ha, is located at Benečija (Fig. 2: 3) on the other, right bank of the Temenica. Geophysical investigations and trial trenching have revealed several buildings probably con­structed towards the end of the 1st and in the 2nd century. 38 Knez 1966, 507–516. A well preserved mortar floor with remains of hypocaust pillars was found directly under the ‘Zamolčanim žrtvam bratomorne vojne’ monument at the time when it was being erected. 39 Čaval, Breščak 2003, Apps. 2–4. 40 Thiel 2001, 88; I sincerely thank Jana Horvat for draw­ing my attention to the characteristic ground plans. 41 Thiel 2001, 91. prepletajo trgovina, obrt in bivanje.40 Dolge hiše vedno stojijo v skupinah, tesno druga ob drugi, z ožjo stranjo obrnjene proti komunikaciji. Zato celo predvidevajo, da je bila cena najema tovrstnih objektov (ali morda davek) odvisna od dolžine fasade, ki je bila obrnjena na glavno cesto. Prednji del stavbe, ki je bil namenjen prodaji, je bil navadno odprt s širokim vhodom (včasih celo s portikom). Zadaj ležijo delavnica ali skladišče in bivalni del.41 Takšno razporeditev prostora opazimo tudi v objektu C. BLIŽNJA OKOLICA Na Benečiji (sl. 2: 3), na desnem bregu Temenice in nasproti rimski naselbini leži še eno poselitveno ob­močje, veliko skoraj 4 ha. Z geofizikalnimi raziskavami in sondiranji je bilo ugotovljenih več objektov, ki so verjetno nastali ob koncu 1. in v 2. st. Stavbe so imele kamnite, z malto zidane temelje in opečnato kritino.42 V rimski dobi je bilo poseljeno tudi območje Tre­banjskega gradu (sl. 2: 4).43 V Kamnem Potoku (sl. 1), okoli 2,5 km severo­zahodno od Trebnjega, so bile odkrite lončene cevi (sl. 26), ki jih povezujemo z rimskodobnim vodovo­dom.44 Na hribu Brnek pri Dolenji Nemški vasi, okoli 3 km vzhodno od Trebnjega, je deloval rimski kamnolom, katerega kamen je bil uporabljen za spomenike oz. grad­bene elemente grobne arhitekture v cerkvi sv. Petra pri Jezeru (sl. 1) pri Trebnjem in v Dragi pri Beli cerkvi.45 GROBIŠČA Na Pristavi (sl. 2: 2), zahodno od naselbine, so bili odkriti žgani in skeletni grobovi, ki spadajo v obdobje od 1. do 4. st. 46 Slabe je grobove razvrstil v nekaj osnovnih oblik. Med najpreprostejše sodijo tisti, v katerih so bili sežgani ostanki umrlega z žaro in pridatki položeni v preprosto, navadno okroglo grobno jamo, ki je bila vča­sih pokrita s kamnito ploščo. Nekatere grobne jame so 40 Thiel 2001, 88; na značilne tlorise me je opozorila Jana Horvat, za kar se ji najlepše zahvaljujem. 41 Thiel 2001, 91. 42 Breščak 1990, 83. 43 Mason 1993. 44 EŠD 26499 (Ministrstvo za kulturo RS, Register ne­premične kulturne dediščine, evidenčna številka dediščine); Breščak 1977, 228. 45 Breščak, Lovenjak 2010, 308. 46 Knez 1969, 120–132; Oman 1974, 139–142 (pet gro­bov na parc. št. 216/2 k. o. Štefan je prostorsko mogoče na­tančno umestiti, enajst grobov na parceli 131 ter treh grobov na parc. 232 pa ne; gre za velike parcele, ki so bile pozneje razdeljene). Slabe 1993, 30; Knez, 1969, 131; Josipovič 2002; Bavec 2006a. The buildings had mortar­bound stone foundations and tiles as roof covering.42 The area of Trebnje Castle was also inhabited in the Roman period (Fig. 2: 4).43 Some 2.5 km northwest of Trebnje, at Kamni potok (Fig. 1), clay pipes were recovered (Fig. 26) that indicate an aqueduct.44 A Roman quarry has been documented on the hill of Brnek near Dolenja Nemška vas, located roughly 3 km east of Trebnje. It supplied the limestone used, for example, for the funerary monuments found built into the church of St Peter near Jezero (Fig. 1) and those at Draga near Bela cerkev.45 CEMETERIES Cremation and inhumation burials dating from the 1st to the 4th century were unearthed at Pristava, west of the settlement. 46 Slabe distinguished between several basic forms of graves. The simplest were the usually round pits into which the cremated remains inside an urn were placed together with grave goods; several of these pits have been found covered with a stone slab. Some were 42 Breščak 1990, 83. 43 Mason 1993. 44 EŠD (Heritage Register No.) 26499; Breščak 1977, 228. 45 Breščak, Lovenjak 2010, 301. 46 Knez 1969, 120–132; Oman 1974, 139–142 (the six graves on Lot No. 216/2, Štefan cadastral municipality, can be precisely located, the eleven graves on Lot No. 131 and the three of Lot No. 232 cannot; these are large burial plots that were later subdivided). Slabe 1993, 30; Knez 1969, 131; Josipovič 2002; Bavec 2006a. bile obložene s kamnitimi ploščami. Grobnice, navadno pravokotnega tlorisa, so imele stene zložene iz kamenja oziroma so bile zidane in včasih ometane. Notranjost je bila ponekod predeljena. Enkrat se pojavi pokop v kamniti pepelnici. Večje grobnice so bile izropane že v antiki.47 Na območju lokacije Levji dvorec sta bila odkrita delno uničen žgan grob in otroški skeletni grob s skrom­nimi pridatki (sl. 16; 27; 28). Mlajši skeletni grobovi iz druge polovice 3. in prve polovice 4. st. se pojavljajo še na skrajnem zahodnem robu nekropole na Pristavi, na kateri sicer prevladujejo žgani grobovi.48 Primarne lege ohranjenih nagrobnikov niso znane.49 V slabih 5 kilometrov oddaljenem Jezeru pri Trebnjem (sl. 1), pri cerkvi sv. Petra, je ležalo grobišče, ohranjene so sledi grobnih parcel, v cerkev pa so vzidani arhitekturni elementi in štirje nagrobniki. Grobišče je ležalo verjetno neposredno ob itinerarski cesti.50 47 Slabe 1990, 94. 48 Josipovič 2002 (kar pet tu odkritih grobov je skeletnih; med do zdaj objavljenimi, vključno z že omenjenim otroškim grobom, so bili skeletni trije). 49 Knez 1960–61, 209; Lovenjak 1998, št. 141–146. Prve omembe grobišča, npr. Petru 1960–1961, 208; Knez 1975; Breščak, Lovenjak 2010, 306–308 (z analizo zgo­dovine raziskav in epigrafskega gradiva). Za potek rimske državne ceste od Emone do Nevioduna in miljnike ob njej prim. Lovenjak 1998, 333–336, št. 178–194; 1997; 2006 in Breščak 2008. cist graves, i.e. with a lining of stone slabs. Masonry tombs were usually rectangular in plan, constructed either in the drystone technique or bound with mortar and sometimes furnished with wall plaster. In some cases, their interior was partitioned. The area also revealed one burial in an ash chest. The large tombs were already looted in Antiquity.47 A partially destroyed cremation burial and the inhumation of a child with scant grave goods were also found at the Levji dvorec site (Figs. 16; 27–28). Later inhumation burials, from the second half of the 3rd and the first half of the 4th century were unearthed at the western edge of this cemetery that otherwise mainly con­sists of cremations.48 The Trebnje area revealed several tombstones, but their original locations are not known.49 A more distant cemetery is located some 5 km to the east of Trebnje, at the church of St Peter in Jezero (Fig. 1). Traces of burial plots have survived, while sev­eral blocks of composite funerary architecture and four tombstones are built into the church. The cemetery was presumably located at the main road towards Siscia.50 47 Slabe 1990, 94. 48 Josipovič 2002 (as many as five of the graves found here are inhumations, three of which including the child’s burial have already been published). 49 Knez 1960–61, 209; Lovenjak 1998, Nos. 141–146. 50 For the first records of the cemetery, see e.g. Petru DRUŽBA SOCIETY Prvi del imena naselbine, praetorium, pomeni, da je tu stala utrjena obcestna postaja, drugi del je povezan z lokalnim keltskim prebivalstvom, Latobiki.51 Postaja je ležala na pol poti med Emono in Neviodunom in je sodila v ager Nevioduna ter z njim v provinco Zgornjo Panonijo.52 Posamezni napisi pričajo o lokalnem prebivalstvu, ki je imelo staroselske korenine, npr. v gentilnem imenu donatorja Maronija (Maro) na Herkulu posvečenem napisu iz konca 2. ali začetka 3. st.53 Na napisih 1. st., najdenih v širši okolici Trebnjega, v Štatenberku inKamnju pri Šentrupertu se pojavljajo keltska imena.54 V Trebnjem je bilo odkritih osemnajst posvetilnih napisov, ki so jih postavili konzularni beneficiariji – be-neficiarii consularis. Starejša skupina oltarjev, ki so bili posvečeni samo Jupitru, sodi na konec 2. st. (sl. 29), pri čemer je najstarejši po omembah konzulov datiran v leto 158 (ali v 195). Mlajša skupina je datirana po konzulskih datumih med letoma 217 in 257. Ti so posvečeni hkrati Jupitru (v enem primeru dolihenskemu Jupitru) in geni­ju kraja – genius loci. Beneficiariji so služili v 10. ali v 14. dvojni legiji, ki sta imeli sedež v Zgornji Panoniji. Večina teh napisov (mdr. sl 29) je bila naključno odkritih pri gradnji hleva gostilne Opara (takrat hišna št. 26, danes Goliev trg 13) in pri hiši Zoré (Rimska cesta 7a?) kot tudi v neposredni bližini župne cerkve (sl. 2: 1), na podlagi česar je Šašel tam domneval beneficiarsko postojanko.55 Odkriti so bili še posvetilni napis Herkulu,56 po­svetilo dolihenskemu in heliopolitanskemu Jupitru57 ter tri posvetila Mitri.58 O templju, ki je bil posvečen doli­henskemu Jupitru, priča gradbeni napis iz l. 196–197.59 Kljub temu da je Pretorij Latobikov ležal na ozemlju Nevioduna, sta bila v cerkev sv. Petra v Jezeru vzidana nagrobnik Gaja Avrelija Firmina, dekuriona Emone60 ter nagrobnik dekurionov in duumvirov Celeje – Gaja Longinija Severina in njegovega sina Longinija Avita. 51 Šašel Kos 1995, 151. 52 Za zgodovino in epigrafske spomenika Nevioduna in njegovega območja glej Lovenjak 1998; 2003. 53 Šašel Kos 1995, 160; Lovenjak 1998, 228 ss; CIL III, 10786. 54 V cerkvi sv. Martina v Štatenberku vzidan napis Pub­lija Nertomarija Kvarta in žene Meite (CIL III, 107494 = AIJ 236) kaže na staroselski izvor; prim. Lovenjak 1998, 180–182. V Kamnju pri Šentrupertu napis Lukija Bajbija Sempronija ženi Eskingi razkriva žensko keltsko ime ter nedosledno, de­loma celo nepravilno rabo latinščine; Lovenjak 1997, 75, sl. 14; 1998, 157–158. 55 Šašel 1975b, 232; Šašel Kos 1995; Lovenjak 1998, št. 116–133. 56 Lovenjak 1998, št. 115. 57 Lovenjak 1998, št. 135. 58 Lovenjak 1998, št. 136–137, 139. 59 Šašel 1993, 34; Lovenjak 1998, št. 134. 60 Lovenjak 1998, 151 ss. The first part of the settlement’s name, praetorium, signifies a fortified roadside station, while the second part refers to the local Celtic population, the Latobici.51 The station was located half way between Emona and Neviodunum; it formed part of the latter’s territory and hence the province of Upper Pannonia.52 Epigraphic evidence reveals a local population of indigenous roots. An example is a man with the genti­licium Maro, who dedicated an altar to Hercules towards the end of the 2nd or the beginning of the 3rd century.53 The inscriptions recovered in the wider area of Trebnje, Štatenberg and Kamnje near Šentrupert, dating to the 1st century, reveal Celtic names.54 Eighteen altars put up by the beneficiarii consul-aris were found at Trebnje. The early group, dedicated exclusively to Jupiter, dates to the late 2nd century (Fig. 29), with the consuls mentioned in the inscrip­tions indicating that the earliest altar dates to the year 158 (or 195). The consul dates for the late group of altars, dedicated to both Jupiter (in one case Jupiter Dolichenus) and genius loci, span from 217 to 257. Beneficiarii consularis served in legiones X or XIV Geminae, stationed in Upper Pannonia at this time. Most of the altars (for example that on Fig. 29) were found by chance while building the stable next to the Opara Inn (then at Trebnje 26, today Goliev trg 13), during the construction of the Zoré house (Rimska cesta 7a?) and in the immediate vicinity of the par­ish church (Fig. 2: 1); this led Jaroslav Šašel to posit a station of the beneficiarii consularis at that location.55 Apart from these, other altars have also been found and dedicated to Hercules,56 to Jupiter Dolichenus and Jupiter Heliopolitanus,57 as well as three dedications to 1960–1961, 208; Knez 1975, 230. For an analysis of the his­tory of research and the epigraphic evidence, see Breščak, Lovenjak 2010, 295–301. For the location of the main Roman road from Emona to Neviodunum and the milestones along this road, cf. Lovenjak 1998, 333–336, Nos. 178–194; 1997; 2006 and Breščak 2008. 51 Šašel Kos 1995, 151. 52 For the history and epigraphic evidence of Neviodu­num, see Lovenjak 1998; 2003. 53 Šašel Kos 1995, 160; Lovenjak 1998, 228ff; CIL III, 10786. 54 The inscription for Publius Nertomarius Quartus andhis wife Meita, built into the church of St Martin in Štatenberk (CIL III, 107494 = AIJ 236), shows an indigenous origin of the deceased; cf. Lovenjak 1998, 180–182. The inscription for Lu­cius Baibius Sempronius for his wife Escinga reveals a Celtic female name and an inconsistent, at places even incorrect use of Latin; Lovenjak 1997, 75, Fig. 14; 1998, 157–158. 55 Šašel 1975b, 232; Šašel Kos 1995; Lovenjak 1998, Nos. 116–133. 56 Lovenjak 1998, No. 115. 57 Lovenjak 1998, No. 135. Mithras.58 A building inscription from 196–197 reveals the existence of a temple of Jupiter Dolichenus.59 Despite the fact that Praetorium Latobicorum formed part of the territory of Neviodunum, the church of St Peter in Jezero reveals the presence of officials serv­ing in other towns. Built into the church are pieces of the funerary monuments for Gaius Aurelius Firminus, decurio at Emona,60 and for Gaius Longinius Severinus and his son Longinius Avitus, both decuriones and duumviri of Celeia. The monuments date to the late 2nd or the first half of the 3rd century.61 There is also a statue of a recumbent lion adorning the staircase leading up to Trebnje Castle, which once also stood in a cemetery.62 Translation: Andreja Maver Nagrobnika sta datirana v konec 2. ali v prvo polovico 3. st.61 V stopnišče ob gradu vzidan kip leva je nekoč ravno tako pripadal neki prestižni grobni arhitekturi.62 58 Lovenjak 1998, Nos. 136–137, 139. 59 Šašel 1993, 97; Lovenjak 1998, No. 134. 61 Breščak 2009, 72 ss. 60 Lovenjak 1998, 151 ff. 62 Lovenjak 1998, št. 147. 61 Breščak 2009, 72 ff. 62 Lovenjak 1998, No. 147. Kratice / Abbreviations ANSl = Arheološka najdišča Slovenije. – Ljubljana, 1975. ILJug = A. et J. Šašel, Inscriptiones Latinae quae in Jugoslavia inter annos MCMXL et MCMLX repertae et editae sunt (Situla 5), Ljubljana 1963. RINMS = M. Šašel Kos, The Roman Inscriptions in the National Museum of Slovenia / Lapidarij Narodnega muzeja Slovenije (Situla 35), Ljubljana 1997. BAVEC, U. 2006a, Pristava pri Trebnjem. – Varstvo spomeni­kov. Poročila 39–41 (2000–2004), 159. BAVEC, U. 2006b, Trebnje – arheološko območje Trebnje. – Varstvo spomenikov. Poročila 39–41 (2000–2004), 224 –225. BAVEC, U. 2007, Trebnje – arheološko najdišče Praetorium Latobicorum. – Varstvo spomenikov. Poročila 43 (2006), 224–226. BAVEC, U. 2008, Trebnje – arheološko najdišče Praetorium Latobicorum. – Varstvo spomenikov. Poročila 44 (2007), 285–286. BAVEC, U. 2009, Rimljani ob veliki reki – poskus opisa pose­litve Posavja v času od 1. do 4. stoletja. – V / In: Ukročena lepotica. Sava in njene zgodbe, 51–68, Sevnica. BAVEC, U. 2010, Trebnje – arheološko najdišče Praetorium Latobicorum. – Varstvo spomenikov. Poročila 46 (2009), 383–385. BAVEC, M., M. MURKO, P. PREDAN 2007, Poročilo o ar­heološkem vrednotenju z izkopom strojnih testnih jarkovna parcelah številka 222/3, 222/4, 222/2, 222/1 k.o. Štefan. –Poročilo / Report; arhiv ZVKDS, OE Novo mesto (neo­bjavljeno / unpublished). BREČIĆ, J., A. JOVANOVIĆ 2018, Poročilo o arheoloških raz­iskavah na območju zamenjave vodovodnih cevi v Trebnjem 2014–2015. – Poročilo / Report; arhiv ZVKDS, OE Novo mesto (neobjavljeno / unpublished). BREŠČAK, D. 1977, Kamni potok. – Varstvo spomenikov 21, 228. BREŠČAK, D. 1990, Benečija pri Trebnjem, Trebnje. – V / In: Arheološka najdišča Dolenjske. Ob 100-letnici arkeoloških raziskovanj v Novem mestu, Arheo – Dossier Dolenjska, 83–84, Ljubljana. BREŠČAK, D. 2008,Trebnje – arheološko območje Trebnje. – Varstvo spomenikov. Poročila 44 (2007), 286–287. BREŠČAK, D. 2015, Graves with weapons from Verdun near Stopiče / Grobovi z orožjem z Verduna pri Stopičah. – V / In: J. Istenič, B. Laharnar, J. Horvat (ur. / eds.), Evidence of the Roman army in Slovenia / Sledovi rimske vojske na Slovenskem, Katalogi in monografije 41, 75–123. BREŠČAK, D., M. LOVENJAK 2010, The tombstone of two town magistrates of Celeia in the region of Neviodunum / Nagrobnik dveh mestnih veljakov iz Celeje na območju Nevioduna. – Arheološki vestnik 61, 295–310. BREŠČAK, D., A. WATERS 1990, Benečija pri Trebnjem. – Arheološki pregled 29, 122–124. ČAVAL, S., D. BREŠČAK 2003, Arheološka izkopavanja in nadzor na Dolenjski ulici v Trebnjem. – Poročilo / Report; arhiv ZVKDS, OE Novo mesto (neobjavljeno / unpubli­shed). FODOREAN, F. G. 2017, Praetorium and the Emona–Siscia– Sirmium–Tauruno road in the ancient geographical and epigraphic sources (Praetorium in cesta Emona–Siscia– Sirmium–Tauruno v antičnih geografskih in epigrafskih virih). – Arheološki vestnik 68, 337–348. GABROVEC, S. 1956, Najstarejša zgodovina Dolenjske. – Novo mesto. JOSIPOVIČ, D. 2002, Pristava pri Trebnjem. Poročilo o izvaja­nju zavarovalnih arheoloških raziskav na trasi rekonstrukcije kanalizacije. – Poročilo / Report; arhiv ZVKDS OE Novo mesto (neobjavljeno / unpublished). JOVANOVIĆ, A. 2018a, Poročilo o predhodnih arheoloških raziskavah na lokaciji Trebnje OPPN Kulturni center Trebnje – Občina Trebnje, parc. št. 230, 360/5, 231/3, 348, k.o. Trebnje. – Poročilo / Report; arhiv ZVKDS OE Novo mesto (neobjavljeno / unpublished). JOVANOVIĆ, A. 2018b, Poročilo o predhodni arheološki razis­kavi na območju Trebnje – gradnja pločnika ob lokalni cesti LC 425 051 Trebnje center – Štefan, parc. št. 16/2, 369/25, k.o. Trebnje. – Poročilo / Report; arhiv ZVKDS OE Novo mesto (neobjavljeno / unpublished). KNEZ, T. 1960–1961, Pristava pri Trebnjem. – Varstvo spo­menikov 8, 209–211. KNEZ, T. 1966, Staroslovanski grobovi v Trebnjem (Začasno poročilo). – Arheološki vestnik 17, 507–516. KNEZ, T. 1969, Novi rimski grobovi na Dolenjskem (Neue römische Gräber in Dolenjsko [Unterkrain]). – Razprave 1. razreda SAZU 6, 107–160. KNEZ, T. 1975, Jezero. – V / In: ANSl, 230. KOVAČ, O. 1910, Poročilo o izvedeni predhodni arheološki raz­iskavi, arheološko dokumentiranje strojno izkopanih testnih jarkov na območju parc. št. 232/1, 232/5 k.o. Štefan, Pristava pri Trebnjem, Občina Trebnje. – Poročilo / Report; arhiv ZVKDS, OE Novo mesto (neobjavljeno / unpublished). KOZLEVČAR, M. 2015, Ob izgradnji vodovoda našli ostan­ke iz rimskih časov. – Glasilo občanov, februar 2015, 18, Trebnje. LOVENJAK, M. 1997, Novi in revidirani rimski napisi v Slo­veniji (Die neuen und revidierten römischen Inschriften Sloweniens). – Arheološki vestnik 48, 131–142. LOVENJAK, M. 1998, Inscriptiones Latinae Sloveniae 1. Ne-viodunum. – Situla 37. LOVENJAK, M. 2003, Municipium Favium LatobicorumNeviodunum. – V / In: M: Šašel Kos, P. Scherrer (ur. / eds.), The Autonomous Towns of Noricum and Pannonia / Die autonomen Städte in Noricum und Pannonien. Pannonia 1, Situla 41, 93–105. LOVENJAK, M. 2006, Rimski miljniki na Dolenjskem. Rimska državna cesta Emona­Neviodunum­Siscia. – V / In: Rimske ceste in projekt avtocest, Delno dopolnjen separat iz revije Rast, št. 2 (104), april 2006, 39–47. MASON, P. 1993, Grad Trebnje. Arheološke raziskave 1993. – Poročilo / Report; arhiv ZVKDS OE Novo mesto (neob­javljeno / unpublished). MELIK, A. 1959, Posavska Slovenija. – Slovenija. Geografski opis 2. Opis slovenskih pokrajin 3, Ljubljana.MÜLLNER, A. 1879, Emona. Archäologische Studien aus Krain. – Laibach. OMAN, J. 1974, Pristava pri Trebnjem. – Varstvo spomenikov 17–19/1, 139–142. PETRU, P. 1960–1961, Jezero pri Trebnjem. – Varstvo spome­nikov 8, 208. PETRU, P. 1982, Rimska cesta Neviodunum­Celeia in vloga Brestanice v starejših arheoloških obdobjih. – V / In: Brestanica. Zbornik člankov in razprav, 13– 22, Brestanica. PREMERSTEIN von, A., S. RUTAR 1899, Römische Strassen und Befestigungen in Krain. – Wien. RÖMER­MARTIJNSE, E. 1990, Römerzeitliche Bleietiketten aus Kalsdorf, Steiermark. – Österreichische Akademie der Wissenschaften, Phil.­Hist. Klasse, Denkschriften Bd. 205. RUTAR et al. 2012 = RUTAR, G., K. UDOVČ, D. MLEKUŽ, N. VERŠNIK, T. MULH, M. MIHELIČ, M. HORVAT, B. NADABATH 2012, Ocena arheološkega potenciala območja OPN za Občino Trebnje, metode 1–6. – Poročilo / Report; ar­hiv ZVKDS OE Novo mesto (neobjavljeno / unpublished). SLABE, M. 1990, Pristava pri Trebnjem, Trebnje. – V / In: Arheološka najdišča Dolenjske. Ob 100-letnici arheoloških raziskovanj v Novem mestu, Arheo – Dossier Dolenjska, 93–94, Novo mesto. SLABE, M. 1993, Antična nekropola na Pristavi pri Trebnjem / The Roman cemetery at Pristava near Trebnje. – Vestnik. Zavod RS za varstvo naravne in kulturne dediščine 12, Ljubljana. ŠAŠEL, J. 1975a, Rimske ceste v Sloveniji (viae publicae). – V / In: ANSl, 74–88. ŠAŠEL, J. 1975b, Trebnje. – V / In: ANSl, 231–232. ŠAŠEL, J. 1993, Dokument o rimskem templju v Trebnjem na Dolenjskem ­ novo odkritje. – V / In: M. Slabe, Antična nekropola na Pristavi pri Trebnjem / The Roman cemetery at Pristava near Trebnje, Vestnik 12, 34, 97, Ljubljana. ŠAŠEL, J. 1977, Viae militaris. – V / In: Studien zu den Mi-litärgranzen Roms II. Beihefte der Bonner Jahrbücher 38, Bonn, 235–244 (= Opera selecta, Situla 30, Ljubljana 1992, 459–468). ŠAŠEL KOS, M. 1995, The beneficiarii consularis at Praetorium Latobicorum. – In / V : Römische Inschriften – Neufunde, Neulesungen und Neuinterpretationen. Festschrift für Hans Lieb, Arbeiten zur römischen Epigraphik und Altertums­kunde 2, 149–170, Basel. ŠAŠEL KOS, M. 1997a, Pošta v antiki. – V / In: Pošta na Slo­venskem, 18–43, Ljubljana. ŠAŠEL KOS, M. 1997b, The Roman Inscriptions in the National Museum of Slovenia. – Situla 35. ŠAŠEL KOS, M. 1999, Cestni postaji Atrans in Pretorij La­tobikov. – In / V: B. Aubelj (ur. / ed.), Zakladi tisočletij. Zgodovina Slovenije od neandertalcev do Slovanov, 238–240, Ljubljana. THIEL, A. 2001, Streifenhäuser. – V / In: T. Fischer (ur. / ed.), Die römischen Provinzen. Eine Einführung in ihre Archälo­gie, 88–91, Stuttgart. TIRAN, A. 2015, Arheološko dokumentiranje ob gradnji MLN d.o.o. – Trebnje/Pristava parc. št. 1252/11, k.o. Štefan. – Po­ročilo / Report;, arhiv ZVKDS, OE Novo mesto (neobjav­ljeno / unpublished). TRUHLAR, F. 1975, Stara pota in poskus rekonstrukcije nek­danje prometne mreže.– V / In: ANSl, 99–104. UDOVČ, K. 2010, Poročilo o predhodnih arheoloških raziska­vah na območju industrijske cone v Trebnjem. – Poročilo / Report; arhiv ZVKDS, OE Novo mesto (neobjavljeno / unpublished). Uroš Bavec Zavod za varstvo kulturne dediščine Slovenije Območna enota Novo mesto Skalitskega ulica 1 SI­8000 Novo mesto uros.bavec@zvkds.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 387–402 ROMULA – RIBNICA Irena LAZAR Izvleček Rimska postaja Romula je v antičnih virih znana in omenjena na dveh itinerarjih rimske dobe, na Tabuli Peutingeriani in Antoninskem itinerariju. Postajo je na podlagi naselbinskih ostankov na območje današnje vasi Ribnica (pri Jesenicah na Dolenjskem) lociral Peter Petru. Zgrajena je bila na strateško izjemno ugodni točki, prehodu iz dolin Krke in Save v Panonsko nižino. Nove in obsežne raziskave (2001–2004) so odkrile obsežen naselbinski del z izstopajočim obrambnim objektom, potek rimske ceste Emona–Siscia in velik del zahodnega grobišča (129 grobov). Med najdbami velja izpostaviti del posvetilnega napisa Silvanu Avgustu, ki ga je dal postaviti suženj združenja zakupnikov carine, in napis ROMVLA na svinčeni ploščici za blago. Najdbi sta pomembni za opredelitev statusa naselbine in potrditev njenega imena ter lokacije. Ključne besede: Zgornja Panonija, Ribnica, Romula, rimska doba, naselbina, cestna postaja, carinska postaja, cesta, most, svetišče, utrjena stavba, grobišče Abstract Roman road station Romula is known from ancient sources and mentioned on two Roman itineraries, the Tabula Peutingeriana and Itinerarium Antonini. On the basis of the settlement remains, the station was located on the area of present village Ribnica (near Jesenice in Dolenjska) by Peter Petru. It was built on a strategical point, the transition of the valleys of Krka and Sava Rivers to the Pannonian plain. New and extensive research (2001–2004) revealed a large settle­ment area with a prominent defensive building, the Roman road Emona–Siscia and a large part of the western necropolis with 129 graves. It is important to single out the find of a votive inscription to Silvanus Augustus, dedicated by the slave of the customs officers’ society, and an inscription ROMVLA on a lead tablet. Both finds are important for the definition of the administrative status of the settlement and for the confirmation of its name and location. Key words: Pannonia Superior, Ribnica, Romula, Roman period, settlement, road station, custom station, road, bridge, temple, fortified building, cemetery https://doi.org/10.3986/9789610502586_21 IME Ribnica (pri Jesenicah na Dolenjskem; Petru 1975, 259)oz. tudi Ribnica pri Brežicah, kot se je ime najdišča uveljavilo v času arheoloških raziskav ob gradnji avtocest (2001–2004), je kot območje rimske obcestne postaje Romula med arheološkimi spomeniki Posavja v Registru kulturne dediščine.1 ANTIČNO IME IN UTEMELJITEV LOKACIJE Romula je v antičnih virih znana in omenjena na dveh itinerarjih rimske dobe (Šašel 1975) in sicer na Ta-buli Peutingeriani in v Itinerarium Antonini. Kot postajo na glavni cesti via publica Aquileia–Emona–Siscia–Sir­mium jo Tabula Peutingeriana (IV, 3) postavlja na točko 10 rimskih milj od municipija Neviodunum in 14 milj od postaje Quadrata, Itinerarium Antonini s konca 3. st. pa jo kot postajo, označeno pod številko 274,4 na cesti II Aquileia–Senia–Siscia, postavlja na lokacijo 10 rim-skih milj od kraja Bibium in 14 rimskih milj od postaje Quadrata (Šašel 1975, 76–77). V delu Cosmographia anonimnega geografa iz Ravene je Romula omenjena med postajami vzdolž cest med Dravo in Savo. Navedena je pod številko 15, med postajo Fines (14) in krajem Nomiduni (16), nekdaj rimski Neviodunum (Šašel 1975, 80). Rimska postaja Romula je bila v arheološki stroki deležna precej pozornosti, predvsem zaradi vprašanja svoje lokacije. Alfonz Müllner (1879, 104) jo je nedoka­zano umeščal na polje med Mokricami in Jesenicami na Dolenjskem, druga možna lokacija so bile tudi Prilipe pri Čatežu pri Brežicah. Postajo Romula je na podlagi naselbinskih ostan­kov na območje današnje vasi Ribnica pri Jesenicah lociral Peter Petru (1961, 193–202). Ob gradnji avtoceste Bratstva in enotnosti (1958–1960) so bili namreč na območju vasi Ribnica odkriti in delno raziskani os­tanki arhitekture oziroma naselbine, z ostanki stavbe s hipokavstom, in grobišča z več deset dobro ohranjenimi grobovi s številnimi grobnimi pridatki, ostanki izropane grobnice in grobnih parcel (Petru 1969, 20). Med pomembnejše najdbe zadnjih izkopavanj (2001–2004) sodi napis v kurzivni pisavi na svinčeni ploščici za blago (sl. 10; Lovenjak 2005, 43). Dvostranski napis na ploščici na eni strani beremo kot ROMVLA / ANIONIS. Ena od možnosti je, da gre v tem primeru za ime naselja in pomeni “iz Romule”, v drugi vrstici pa je izpisano ime lastnika Anio-nis v 2. sklonu. Tako inter-pretiran napis ROMVLA bi lahko nedvoumno potrdil pravilnost lociranja naselbine. 1 Ministrstvo za kulturo RS, Register nepremične kultur­ne dediščine, evidenčna številka EŠD 9335. NAME Ribnica (near Jesenice in Dolenjska; Petru 1975, 259) or Ribnica pri Brežicah, as the site came to be known during archaeological research when the high­way system was under construction (2001–2004), is now listed among the Posavje archaeological monuments in the Register of Cultural Heritage as the area of the Ro­man road station Romula.1 ANCIENT NAME AND LOCATION Romula was twice mentioned in ancient sources, i.e. in two Roman itineraries (Šašel 1975): the Tabula Peutin­geriana and Itinerarium Antonini from the end of the 3rd century. The former (IV, 3) places it as a road station on the main via publica through Aquileia–Emona–Siscia– Sirmium, 10 miles from the Neviodunum municipium and 14 miles from the Quadrata station, while the latter marks it under number 274.4 on the II Aquileia–Senia– Siscia road, 10 Roman miles from Bibium and 14 miles from the Quadrata station (Šašel 1975, 76–77). In his Cosmographia, an anonymous geographer from Ravenna mentions Romula as one of the stations along the roads between Drava and Sava. It is listed under number 15, between the Fines station (14) and the Nomiduni settlement (16), formerly the RomanNeviodunum (Šašel 1975, 80). The Roman station Romula received considerable attention in the archaeological community, mainly due to the question of its location. Müllner (1879, 104) placed it without proof on the field between Mokrice and Jesenice na Dolenjskem; another possible location was Prilipe near Čatež not far from Brežice. The location of the station was finally pinpointed on the basis of settlement remains by Petru (1961, 193–202) in the area of today’s village of Ribnica near Jesenice. During the construction of the motorway in the 20th century (1958–1960), remains of architecture or of a settlement were discovered and partly researched in the area of the village of Ribnica: the remains of a building with a hypocaust and a cemetery with dozens of well-preserved graves with numerous grave goods, remains of a robbed tomb and grave plots (Petru 1969, 20). Among the most important finds of the last excava­tion (2001–2004) was the inscription in cursive script on a lead tablet for cloths (Fig. 10; Lovenjak 2005, 43). The two-sided inscription on the tablet reads ROMVLA / ANIONIS on one side. One possibility is that the name signifies the settlement “Romula” and the second line shows the name of the owner of the Anio-nis in the geni­tive case. So interpreted, the inscription ROMVLA could clearly confirm the exact location of the settlement. 1 Ministrstvo za kulturo RS, Register nepremične kultur­ne dediščine, No. EŠD 9335 GEOGRAFSKA LEGA Rimska postaja Romula je bila zgrajena na strateško izjemno ugodni točki na prehodu iz dolin Krke in Save v Panonsko nižino oziroma ob vznožju Gorjancev, na prehodu v ravnino reke Save (Savus; sl. 1). Na najožjem delu terase med Savo in pobočjem Gorjancev je potekala tudi via publica Emona–Siscia. Z vrhov Kincelj in Škof­ljančeva gabrina na obronkih Gorjancev se širi pogled na zahod do Čateža pri Brežicah in na vzhod do Zagreba, na severu pogled sega do Bizeljskega hribovja, južno stran pa varuje pobočje Gorjancev. Naselbina je bila zgrajena na ozkem delu, kjer je bilo mogoče nadzorovati tudi rečni promet po Savi, ki je bila v rimskem obdobju pomembna transportna komunikacija. Del naselja je bil tudi potok Gračenica,2 ki je prečkal osrednji del naselbi­ne, sekal rimsko cesto in se izlival v Savo. Cestna postaja Romula, ki je zaradi svojega položaja povezovala vhod v Italijo z Balkanskim polotokom, je bila zato označena v že omenjenih itinerarjih. ZGODOVINA RAZISKAV V začetku 20. stoletja so na območju arheološkega najdišča pri obdelavi njivskih površin kmetje pogosto naleteli na ostanke rimske arhitekture, ceste, grobov in posamičnih najdb (Petru 1975, 259). Z načrtnimi razis­kavami območja Romule so začeli v teku trasiranja in gradnje avtoceste Bratstva in enotnosti med letoma 1955 in 1960. Ob gradnji avtoceste so bili na tem območju odkriti ostanki arhitekture oziroma del naselbine (k. o. Podgračeno – ostanki rimskega poslopja s hipokavstom, k.o. Velika dolina stavbni kompleks in prefurnij) in gro­bišča z ostanki grobnice in grobnih parcel (k. o. Velika dolina). Vzhodna nekropola, na kateri so odkrili 41 grobov, je ležala severno od ceste Emona–Siscia (Kolšek, Petru 1958–1959; Petru 1961; 1962; 1969). Nove in obsežne raziskave na najdišču so potekale med letoma 2001–2004, v izvedbi Zavoda za varstvo kul­turne dediščine, OE Novo mesto, ob gradnji dolenjskega dela nove avtoceste Ljubljana–Obrežje.3 Rezultati teh raziskav so prinesli precej novosti, med drugim odkritje obsežnega naselbinskega dela Romule in njenega obsega ter poteka trase rimske ceste Emona–Siscia. Raziskan je bil tudi velik del zahodnega grobišča z dobro ohranjenimi grobnimi konstrukcijami, grobnimi inventarji in štirimi grobnimi parcelami, skupno je bilo odkritih 129 žganih in skeletnih grobov (Breščak 2003; 2006). 2 Domačini ga poimenujejo tudi Gračnica ali Grajena, v Atlasu Slovenije (Ljubljana 1986) je poimenovan kot Grače­nica. 3 Arheološka izkopavanja je vodil Danilo Breščak, s so-delovanjem številnih kolegov, ki so vodili delo po sektorjih. Iz poročila (Breščak, 2004) povzemamo tudi mnoge podatke v tem prispevku. GEOGRAPHIC POSITION The Roman station Romula was built on a strategi­cally exceptionally favourable point at the passage from the valleys of the Krka and Sava Rivers to the Pannonian Plain at the foot of the Gorjanci hills, at the passage into the plain of the Sava River (the Savus; Fig. 1). On the nar­rowest part of the terrace between the Sava River and the Gorjanci slope ran the via publica Emona–Siscia. From the peaks of Kincelj and Škofljančeva gabrina hills on the outskirts of Gorjanci hills, the view spreads west to Čatež near Brežice and east to Zagreb, in the north to Bizeljsko hills, while the south side is protected by the slopes of Gorjanci. The settlement was built on a narrow section where river traffic along the Sava River – an important transport communication in the Roman period – could be supervised. Part of the settlement was also the stream Gračenica,2 which crossed the central part of the settlement, and flowed into the Sava river, crossing the Roman road as well. This is why the Romula road station, which linked, due to its position, the entrance to Italy with the Balkan Peninsula was marked in the aforementioned itineraries. RESEARCH HISTORY In the beginning of the 20th century, farmers often stumbled upon the remains of Roman architecture, road, graves and individual finds in the area of the archaeo­logical site, when they were ploughing their fields (Petru 1975, 259). Targeted research of the Romula area began during the construction of the old motorway between 1955 and 1960; during the construction, remains of architecture or parts of a settlement were discovered (remains of a Roman building with a hypocaust in the cadastre municipality Podgračeno and a building complex and a praefurnium in the cadastre municipal­ity Velika dolina) as well as a cemetery with remains of tombs and grave plots (cadastre municipality Velika dolina). The eastern necropolis, in which 41 graves were discovered, lay north of the Emona–Siscia road (Kolšek, Petru 1958–1959; Petru 1961; 1962; 1969). New and extensive research at the site was carried out between 2001 and 2004 under the supervision of ZVKDS OE Novo mesto, during the construction of the Dolenjska part of the new Ljubljana–Obrežje highway.3 The results of this research have brought a lot of new discoveries, including the extensive residential area of Romula and the course of the Roman Emona–Siscia road. A large part of the western cemetery with well­ 2 Locals refer to it as Gračnica or Grajena; it is named Gračenica in the Atlas Slovenije (Ljubljana 1986). 3 Archaeological excavations were supervised by Danilo Breščak with numerous colleagues who supervised the works in different sectors. Several data in this paper are surmised from his report (Breščak, 2004). Sl. 1: Ribnica. Lega rimske naselbine Romula, stavbe 1-4. Fig. 1: Ribnica. The position of the Roman settlement Romula, Buildings 1-4. (Arhiv / Archive: ZVKDS OE Novo mesto; izvedba / elaborated by A. Ogorelec) PRAZGODOVINA Na območju arheološkega najdišča so že bile znane prazgodovinske najdbe (Petru 1975, 259), več prazgodovinskih najdb pa je bilo dokumentiranih v teku izkopavanj v letih 2001–2004. Njihov razpon sega od bakrene do mlajše železne dobe. Najdbe so bile od­krite na skoraj celotnem območju najdišča, večinoma v sekundarni legi, znotraj mlajših plasti. V celoti gledano gradivo kronološko ni homogeno in sodi v različna obdobja prazgodovine. Najstarejše najdbe so datirane v čas eneolitika, kot najmlajše pa so opredeljene najdbe lončenine iz mlajše železne dobe. Večjo koncentracijo najdb in ohranjene strukture so zasledili na vzhodnem območju najdišča. Dve jami sta interpretirani kot vodnjaka oz. zbiralnika za vodo. Manjši zbiralnik je delno presekal večjega, starejšega, ki je bil nepravilnega tlorisa. Na dnu so odkrili ostanke lesene konstrukcije lijakaste oblike. Vzorec lesa je bil pregledan z radiokarbonsko analizo in datira zbiralnik za vodo v srednjo bronasto dobo. preserved grave constructions, grave inventories and four grave plots were also explored. Altogether, 129 cremation and inhumation graves were discovered (Breščak 2003; 2006). PREHISTORY In the area of the archaeological site, prehistoric finds were already known (Petru 1975, 259), and many more were documented also during the excavations in 2001–2004. Ranging from the Copper age to the La Tene period, they were discovered in almost the entire area of the site, mostly in the secondary position, within later layers. Overall, the material is chronologically not homogeneous and belongs to different periods of prehis­tory. The earliest finds date to the period of Eneolithic, whereas the latest finds are pottery fragments from the La Tene period. A greater concentration of finds and survived structures were observed in the eastern area of the site. Two pits are interpreted as wells or water tanks, the smaller one partially cutting into the larger, earlier one with an irregular ground plan. Remains of wooden, funnel-shaped structure were discovered at the bottom. Wood samples were examined by radiocarbon analysis, which dated the water tank to the Middle Bronze Age. RIMSKA DOBA EPIGRAFSKI VIRI Napisi z območja Romule so vključeni v korpuse epigrafskega gradiva: AIJ 253; ILJug 1116, -1117, -1118, -1119, -1120 in ILSl 1, 54, -74, -79, -97, -98, -99; najdbe z zadnjih izkopavanj pri tem še niso upoštevane.4 V elektronsko bazo UBI ERAT LUPA pa je vključen med zadnjimi izkopavanji odkrit posvetilni napis sužnja združenja zakupnikov carine (lupa 22475, brez foto­grafije). Epigrafske najdbe, ki verjetno izvirajo z nasel­binskega območja Ribnice, omenjajo že od začetka 20. stoletja. Od tu verjetno izvira že prej odkriti žrtvenik, ki je bil vzidan v gradu Mokrice: posvetilno aro Jupitru je dal postaviti suženj Rufin (ILSl 1, 54 = AIJ 253). Nagrobnik Tiberija Klavdija Evdija iz 1. st. sta dala postaviti osvobojenec, katerega ime ni ohranjeno, in Tiberij Barbij Primus (ILSl 1, 74). Del drugega nagrob­nika je bil najden leta 1958 v Podgračenem (ILSl 1, 79), leta 1961 pa deli treh nagrobnikov na njivi K. Zanjug (ILSl 1, 97–99). Dva izmed njih kažeta izredno kako­vostno izvedbo črk (ILSl 1, 97 in 98), vsi pa so zelo fragmentarno ohranjeni. Vsi ti nagrobniki izvirajo z zahodne nekropole. Med novejšimi najdbami velja posebej izpostaviti že omenjeni posvetilni napis Silvanu Avgustu (lupa 22475),5 ki ga je dal postaviti Fajder Cecilijan ([Ph]aeder Caecilianus), suženj združenja zakupnikov carine (sl. 2), in napis ROMVLA na eni od svinčenih ploščic za blago (sl. 10; Lovenjak 2005, 42). Ploščice (najden je bil tudi večji kos prepognjene svinčene pločevine) na eni strani odpirajo vprašanje obstoja proizvodnje in obdelave blaga (lat. fullonica) v naselbini, kar bi bilo glede na bližino re-ke in zadostno količino vode mogoče. Zgoraj omenjena ploščica pa je predvsem pomembna zaradi zapisa imena naselbine oz. postaje. NUMIZMATIČNI VIRI Numizmatične najdbe z izkopavanj v obdobju 2001–2004 je obdelala Alenka Miškec in predstavila v poročilu (2004). V času izkopavanja je bil na najdišču odkrit 1001 novec, od tega je 969 rimskih (6 republi­kanskih, 97 iz obdobja do konca 1. st., 99 iz 2. st., 98 iz 3. st. in 630 iz 4. st.). Najdena sta bila tudi dva keltska srebrnika in srednjeveški srebrnik iz 12. st. Iz obdobja od 17. do 20. stoletja je izviralo 18 novcev. V objavi Milana Lovenjaka 1998 (ILSl 1) je navedena vsa starejša literatura in prve objave epigrafskih spomenikov, zato jih tu ponovno ne navajamo. 5 Napis je interpretiral Milan Lovenjak, Filozofska fakul­teta Univerze v Ljubljani. Interpretacijo glej še v zaključku besedila. ROMAN PERIOD EPIGRAPHIC SOURCES The inscriptions from Romula are included in the epigraphic corpora: AIJ 253; ILJug 1116, -1117, -1118, -1119, -1120 in ILSl 1, 54, -74, -79, -97, -98, -99; the latest excavations not taken into account yet.4 A dedicatory inscription of a slave to the association of publicani was discovered during the last excavations and listed in the electronic database UBI ERAT LUPA (lupa 22475, without photo). Epigraphic finds, which probably originate from the Ribnica settlement area, have been mentioned since the early 20th century. An altar built into the Mokrice castle probably originated from Ribnica. It was erected by the slave Rufinus (ILSl 1, 54 = AIJ 253). The tombstone of Tiberius Claudius Eudius from the 1st century was built by his freedman and by Ti­berius Barbius Primus (ILSl 1, 74). A part of the second tombstone was found in 1958 in Podgračeno (ILSl 1, 79). Three tombstones were found in 1961 on the field of Zanjug (ILSl 1, 97–99). Two of them show a high-quality lettering (ILSl 1, 97 and 98), all of them survived only in fragments. All these tombstones come from the western necropolis. Among the more recent finds, the partial inscrip­tion for Silvanus Augustus should be mentioned (lupa 22475),5 set up by [Ph]aeder Caecilianus, a slave to the association of publicani (Fig. 2). The inscription ROM­VLA was discovered on one of the lead tablets (Fig. 10; Lovenjak 2005, 42). Together with a large piece of folded lead sheet, the lead tablets raise the question of the existence of production and processing of textile in the settlement (lat. fullonica), which would be possible due to the proximity of the river and sufficient quantities of water. The above-mentioned tablet is especially important because of the name of the settlement or the station. NUMISMATIC SOURCES Numismatic finds from the 2001–2004 excava­tions were processed and reported on (2004) by Alenka Miškec. 1001 coins were discovered on the site during this excavation, 969 of which were Roman (6 Republican, 97 from the period until the end of the late 1st century, 99 from the 2nd century, 98 from the 3rd century and 630 from the 4th century). Two Celtic silver coins were also found and one medieval silver coin from the 12th 4 In Lovenjak’s publication 1998 (ILSl 1) all previous lit­erature and first publishing of epigraphic sources are listed, which is why they are not listed here. 5 The inscription was interpreted by Milan Lovenjak, Fi­lozofska fakulteta, Univerza v Ljubljani. 0 10 cm Sl. 2: Ribnica. Posvetilni napis Silvanu Avgustu, ki ga je dal postaviti suženj združenja zakupnikov carine F. Cecilijan. Fig. 2: Ribnica. Votive inscription for Silvanus Augustus dedi­cated by F. Caecilianus, slave of the sociorum publici portorii - custom association. (Arhiv / Archive: ZVKDS OE Novo mesto) OBSEG NASELJA Rimska postaja je nastala na ravninskem prehodu med reko Savo in skrajnim severnim robom Gorjancev, in sicer na najožjem delu. Zrasla je na zgornje savski te­rasi in pobočjih potoka Gračenica. Zaradi naravnih meja na severu in jugu se je naselje lahko širilo le na vzhod in zahod, vzdolž rimske ceste. Z raziskavami med letoma 2001 in 2004 določena dolžina postaje je tako znašala približno 600 m, širina pa le med 60 do 125 m (sl. 1). Vzhodna meja naselja je bila določena s položajem vzhodne nekropole. Na vzhodnem delu je bila meja naselbine omejena z leseno konstrukcijo, neke vrste ograjo oz. palisado z nizom stebrov, postavljenih v smeri sever–jug. Položaj nekropole potrjuje, da se poselitev ni širila izven meje palisade. Zahodna meja naselbine je bila določena z raziskanim območjem zahodne nekropole. CESTA Trasa rimske ceste Emona–Neviodunum–Siscia je bila potrjena z arheološkimi raziskavami najprej leta 1958, ko so odkrili vzhodno nekropolo Romule, ki je v ozkem pasu ležala vzdolž ceste (Petru 1969). Med izkopavanji v letih 2001–2004 pa je bila trasa ceste po-century. 18 coins were from the period between the 17th and 20th century. THE EXTENT OF THE SETTLEMENT The Roman station was built on narrowest point of the plateau pass between the Sava River and the far north edge of Gorjanci hills. It developed on the upper Sava terrace and the slopes of the Gračenica stream. Due to natural restrictions to the north and south, the settle­ment could spread only to the east and west, along the Roman road, reaching the length (measured during re­search between 2001 and 2004) of approximately 600 m, while its width was only between 60 and 125 m (Fig. 1). The eastern boundary of the settlement was deter­mined by the position of the eastern necropolis; here, the settlement was enclosed by a wooden construction, some kind of fence or palisade with a series of wooden pillars, following the north–south direction. The posi­tion of the necropolis confirms that the settlement did not spread beyond the palisade. The western border of the settlement was determined by the investigated area of the western necropolis. THE ROAD The route of the Roman road Emona–Nevio­dunum–Siscia was first confirmed by archaeological research in 1958, when the eastern Romula necropolis was discovered, which lay in a narrow strip along the road (Petru 1969). During the 2001–2004 excavations, the route of the road was again documented in three places, which enabled the reconstruction of its course through the settlement (Fig. 1). In the western part of the site, the road was docu­mented along the Roman cemetery in the total length of 110 m. The road was built as a gravel fill, and steeply sloped to each side from the centre. The main part of the road was 6 m wide and no more than 0.20 m thick in its preserved state; its level corresponded to the level of the graves beside it, which suggests that it couldn’t have been significantly thicker (Fig. 3). The fill was predominantly fine carbonate pebbles, in places it consisted of gravel and debris. Here and there, the edge of the road was fortified with larger stones. In the central part of the settlement, only a short section of the road was documented; here, the gravel was 0.5 meter thick. In the eastern part of the site, the road was uncovered in the length of 30.8 m. The body of the road was 6 m wide, together with erosion resi­dues 6.5 m. It was in poor state; the gravel fill was 0.15 m thick, underneath it was a 0.20 m thick layer of road foundation consisting of large, closely packed stone blocks mixed with sand. >> Sl. 3: Ribnica. Površina rimske ceste odkrita ob zahodni nekropoli. Fig. 3: Ribnica. The surface of the Roman road discovered near the western necropolis. (Arhiv / Archive: ZVKDS OE Novo mesto) novno dokumentirana na treh mestih, kar je omogočilo rekonstrukcijo njenega poteka skozi naselbino (sl. 1). Na zahodnem delu najdišča je bila cesta dokumenti­rana ob rimskem grobišču v skupni dolžini 110 m. Cesta je bila grajena kot prodnato nasutje, v sredini je bila napeta in se je proti robovom izklinila. Osrednji del ceste je bil ši­rok 6 m, debelina ohranjenega nasutja ni presegala 0,20 m. Njen nivo se je ujemal z nivojem vkopanih rimskih gro­bov ob njej, iz česar lahko sklepamo, da v osnovi ni bila bistveno debelejša (sl. 3). Nasutje so pretežno sestavljali drobnejši beli karbonatni prodniki, ponekod pa je bila sestavljena iz proda in zaplat drobirja. Na posameznih mestih je bil rob ceste utrjen z večjimi kamni. V osrednjem delu naselja je bila cesta dokumentira­na le v kratkem odseku, prodnato nasutje je bilo tu debelo 0,5 m. Na vzhodnem delu najdišča je bila cesta odkrita v dolžini 30,8 m. Jedro ceste je bilo široko 6 m, z erozijskimi ostanki pa 6,5 m. Njena ohranjenost je bila slaba, prodnato nasutje je bilo debelo 0,15 m, pod nasutjem pa je ležala 0,20 m debela plast podlage za cesto, ki so jo sestavljali močno zbiti oglati kamni, pomešani s peskom. Glede na stratigrafsko lego je gradnja ceste datirana v 1. st. Brez dvoma pa je bila v uporabi skozi celoten obstoj naselja, to je v obdobju med 1. in 4. st. MOST V najstarejši fazi poselitve je bil južno od rimske ceste postavljen most, ki je vodil preko potoka na zahodnem delu naselbine in je služil kot notranjakomunikacija. Žal je bil zaradi srednjeveških in novo­veških gradenj uničen in dokumentiran le v skromnih ostankih. STAVBE V okviru raziskane naselbine velja izpostaviti štiri objekte, ki so bili odkriti v vzhodnem in osrednjem delu naselja. V vzhodnem delu, tako rekoč na vzhod­nem robu naselbine, je severno ob rimski cesti ležal odlično zidan in izstopajoč objekt 4 (sl. 1: 4; 4). Stavba je bila pravokotnega tlorisa, dolžine 18,5 in ohranjene širine 10,5 m. Jugozahodna stran objekta je bila namreč deloma uničena ob gradnji ceste leta 1958. Objekt je bil orientiran v smeri severozahod–jugovzhod, grajen According to its stratigraphic position, the con­struction of the road dates to the 1st century. It was undoubtedly used throughout the entire existence of the settlement, that is, between the 1st and 4th century. THE BRIDGE In the earliest settlement stage, the Roman bridge was located south of the Roman road, leading across the stream in the western part of the settlement and serving as an internal communication. Due to medieval and early modern construction it was unfortunately destroyed and documented only in modest remains. THE BUILDINGS Four buildings need special attention; they were discovered in the eastern and central part of the settle­ment. In the eastern part, almost on the eastern edge of the settlement, a well-built, striking Building 4 was located (Figs. 1: 4; 4) on the northern side of the Roman road. The building had a rectangular ground plan, it was 18.5 m long and 10.5 m wide in its survived extent. The southwestern side of the building was partly destroyed during the con­struction of the road in 1958. The building was oriented northwest-southeast and was built from mortar-bound stones. The foundations of the walls were 0.5 to 1.5 m deep, depending on the terrain, and the walls above the je bil iz kamnov, vezanih z malto. Temelji zidov so bili visoki od 0,5 do 1,5 m, odvisno od podlage, zidovi pa so se nad nekoliko razširjenimi temelji ohranili v višini ene do dveh vrst kamenja. Širina zidov je bila 0,8 do 1 m. Na severovzhodni strani objekta je tekel najdaljši zid, na katerega so se pravokotno prislanjali štirje krajši zidovi. Dva sta predstavljala zunanja zidova objekta, dva pa sta delila notranjost v tri prostore (osrednji širine 9,5 m; stranska širine 2,9 m). Stranska prostora sta bila s prečnim zidom deljena še na manjše prostore. Na zunanji strani objekta so bili na vseh treh stranicah pravilno razporejeni kontraforji, pet na daljši in po dva na krajših stranicah. Glede na širino zidov in dodatne ojačitve s kontraforji lahko sklepamo, da gre za močno konstrukcijo, ki je lahko nosila veliko višino. O namembnosti zgradbe, ki v tem trenutku ver­jetno predstavlja najpomembnejši objekt raziskane naselbine, ni neposrednih dokazov. Glede na izredno trdno gradnjo, najdbe vojaške opreme (sulične osti, osti kopja, puščična ost) in položaj objekta, ki kontrolira vhod v naselbino z vzhodne strani, pa lahko sklepamo o varovalni in morda delno vojaški namembnosti stavbe. Nastanek objekta je na osnovi keramičnih najdb mogoče postaviti v drugo polovico 2. st. (Vidrih Perko 2004). V osrednjem delu sta bila na sedlu na vzhodni stra­ni potoka Gračenica raziskana objekta 1 in 2 (sl. 1: 1,2; 5). Na območju objekta 1 je najprej stala lesena stavba, kasneje pa je na istem mestu zrasla kamnita stavba pra­vokotnega tlorisa velikosti 13 × 11 m (notranji premer 10 × 11m), ki je bila glede na širino temeljev (med 70 in 80 cm) in zidov (med 55 in 60 cm) visoka vsaj med 6 in 8 m. Orientirana je bila v smeri severovzhod–jugozahod, skladno s potekom rimske ceste, severovzhodni vogal ni bil ohranjen. Stavba je bila grajena iz enakomerno velikih apnenčastih kamnov, vezanih z malto. Zunanje in notranje lice zidov je bilo izdelano iz pravilno oblikovanih Sl. 4: Ribnica. Vzhodni del rimske na­selbine. Tloris objekta 4 s kontrafori. Fig. 4: Ribnica. Eastern part of the settlement. Ground plan of the Build­ing 4 with buttresses. (Arhiv / Archive: ZVKDS OE Novo mesto; izvedba / elaborated by A. Preložnik) slightly thicker foundations were preserved to the height of one or two rock lines. The width of the walls was 0.8 to 1 m. The longest wall was on the north-eastern side of the building; four shorter walls were perpendicular to it, two of which were the outer walls of the building, while the other two divided the interior into three rooms (the cen­tral one 9.5 m wide, the two side rooms 2.9 m wide). The side rooms were further divided by transverse walls into smaller rooms. On the outside of the building, buttresses were lined at even spacing on all three sides: five along the longest side and two along the short ones. Judging by the thickness of the walls and additional reinforcement with buttresses, it is safe to assume that this was a strong construction, capable of supporting a tall building. There is no direct evidence of the purpose of this building, which is, at this moment, probably the most important object of the explored settlement. Judging by its extremely solid construction, some finds of military equipment (lance, spearheads, arrowheads) and its con­trolling position at the entrance to the settlement from the eastern side, it is safe to speculate about the protec­tive or possibly military purpose of the facility, whose construction can be placed in the second half of the 2nd century based on ceramic finds (Vidrih Perko 2004). In the central part, Buildings 1 and 2 (Figs. 1: 1,2; 5) were examined on the saddle on the eastern side of the Gračenica stream. In the area of Building 1, a wooden building first lay, which was later substituted for a stone building with a rectangular ground plan measuring 13 × 11 m (internal diameter 10 × 11 m); judging from the thickness of the foundations (between 70 and 80 cm) and the walls (be­tween 55 and 60 cm), the building was at least 6 to 8 m high. Facing northeast-southwest, its floorplan followed the direction of the road; its north-east corner did not survive. The building was built of even, mortar-bound Sl. 5: Ribnica. Osrednji del rimske naselbine. Tloris objektov 1 in 2. Fig. 5: Ribnica. Central part of the settlement. Ground plans of the Buildings 1 and 2. (Arhiv / Archive: ZVKDS OE Novo mesto; izvedba / elaborated by A. Preložnik) kamnov. V južnem zidu se je ohranila konstrukcija vhoda v objekt. V zahodnem delu stavbe so se ohranili štirje pra­vokotno obdelani kamni, morda del podstavka za steber. Drobne najdbe so raznolike in segajo od fibul, igel, jagod do žrmelj, rezil, sekačev in živalskih kosti. Interpretacija objekta zato ni povsem jasna, čeprav njegova arhitektura brez dvoma kaže, da je šlo za enega glavnih upravnih objektov naselja. Prav tako so zanimiva številna kurišča, dokumentirana znotraj objekta. Doku­mentirane strukture (kurišča in peč) so bile vkopane v hodno površino ali pa so ležale na njej. Južno od objekta 1 je na oddaljenosti 3 m ležal objekt 2 (sl. 1: 2; 5), ki je bil ohranjen le deloma, v severozahodnem delu. Južni del ni bil raziskan, ker je ležal pod nasutjem za magistralno cesto. Objekt je bil postavljen v smeri severovzhod–jugozahod, skladno s limestone blocks. The outer and inner faces of the walls were made of properly shaped stones. In the south wall, the structure of the entrance to the building survived. In the western part of the building, four rectangular stones survived, perhaps parts of pillar bases. Small finds are diverse, ranging from fibulae, nee­dles, beads and querns, blades, cutters and animal bones. This makes the interpretation of the building somewhat unclear, although its architecture clearly indicates that it must have been one of the main administrative structures of the settlement. Equally interesting are many fireplaces inside the building. Documented structures (fireplaces and furnace) were dug into the walking surface or lay on it. Building 2 (Figs. 1: 2; 5) was located at a distance of 3 m south of Building 1. It survived only partially in the north-western part. Its southern part was not researched potekom rimske ceste. Ohranjene dimenzije zunanjih zidov so bile 8 × 6 m, grajeni so bili iz bolj ali manj velikih apnenčastih kamnov, vezanih z malto. Tloris objekta je zanimiv, saj konstrukcijo zahodnega dela tvorita dva vzporedna vogala – zunanji in notranji. Zunanji, širši in močnejši zid, in notranji zid sta bila ločena z 1,8 m širokim hodnikom. Tloris dvojnih zidov kaže na kom­pleksen objekt in zasnovo, ki bi jo lahko interpretirali kot del svetišča obhodnega tipa, kot jih poznamo npr. v Podkraju pri Hrastniku in Celeji (Lazar 2011). Med objektoma 1 in 2 je bila ohranjena tlakovana prodnata površina, ki je interpretirana kot antična hod-na površina. V plasti izravnave je bilo najdenih precej odlomkov rimske keramike in gradbenega materiala. Med drobnimi najdbami velja omeniti ločno fibulo, prstan, iglo in rozeto s steklenim vložkom. Oba objekta sta živela v 1. in 2. st. V centru zahodnega dela naselja pa je bil odkrit lesen objekt 3 (vel. 5,8 × 5,1 m; sl. 1: 3), postavljen na kamnitih temeljih, z odvodnim kanalom, ki je bil znotraj obložen z deskami. V objektu in severno izven njega sta bili dokumentirani dve peči. Ohranjenih je bilo veliko kosov ožgane ilovice in kosov žlindre. Objekt je stal tik ob rimski cesti in bil odprt proti njej, zato je bil interpretiran kot delavnica oz. kovačija. Najdbe datirajo stavbo v 1. in 2. st. GROBIŠČA Vzhodna nekropola (raziskave v letih 1958–1960) je ležala vzhodno od naselja, v ozkem pasu neposredno severno ob cesti Emona–Siscia (sl. 1). Odkriti žgani in skeletni grobovi (41 grobov, od tega 5 skeletnih) sodijo v čas od začetka 1. st. do 3. st. Zahodno grobišče, pri vasi Podgračeno, je bil raz­iskano v letih 2002 in 2004 v dolžini 95 m in širini do 16 m. Raztezalo se je vzdolž severne strani rimske ceste. Grobišče je od ceste ločil 1,5 m širok pas prodnatega na­sutja oz. hodne površine. Vsi grobovi so bili orientirani glede na potek ceste. Skupaj je bilo odkritih 129 grobov, od tega 105 žganih in 24 skeletnih. Poleg posamičnih grobov so bile raziskane štiri zidane grobne parcele z več posamičnimi grobovi. Dokumentirane oblike žganih grobov so raznolike: žgan grob v preprosti grobni jami, žgan grob pokrit s kamni, bustum, žgan grob z žaro, grob z žaro v grobni jami s kamnito ploščo na dnu, grob obložen z lomljenci in ploščami na dnu, grob v kvadratni skrinji iz kamnitih plošč, grob v pravokotni zidani grobnici (sl. 6). Skeletni grobovi so največkrat v obliki preproste grobne jame, sledijo pokop z oblogo tlakovcev, opečna skrinja iz tegul in en pokop v pravokotni zidani grobnici. V grobovih so bili priloženi številni keramični, jantarni, kovinski in stekleni pridatki, nekateri še posebej izstopajo zaradi kakovosti. Zahodno grobišče kaže vsaj dve fazi uporabe. Starejši grobovi, iz zgodnjega oz. prve because it lay under the fill for the main road. The building was facing northeast-southwest, following the course of the Roman road. The dimensions of the survived outer walls were 8 × 6 m; they were built of more or less large limestone mortar-bound blocks. The ground plan of the building is interesting because the construction of the western part forms two parallel – outer and inner – cor­ners. The outer, thicker and stronger wall, and the inner wall were separated by a 1.8 m wide corridor. The floor plan of the double walls points to a complex building. According to the ground plan, it could be interpreted as an ambulatory temple similar, for example, to those in Podkraj near Hrastnik and Celeia (Lazar 2011). Between Buildings 1 and 2 a gravel-paved area survived, interpreted as an ancient walking surface. A number of fragments of Roman ceramics and construc­tion material were found in the levelling layer. Among small finds, a fibula should be mentioned as well as a ring, a needle and a rosette with a glass insert. Both buildings were in use in the 1st and 2nd century. In the centre of the western part of the settlement, a wooden Building 3 (5.8 × 5.1 m; Fig. 1: 3) was discovered; built on stone foundations, it featured a drainage channel that was covered with planks on the inside. Two furnaces were documented in the northern part of the building. Judging by numerous pieces of burnt clay and slag as well as the position of the building facing the road, it was interpreted as an ironworks workshop or a smithy. Findings date the building to the 1st and 2nd century. THE CEMETERIES The eastern necropolis (researched 1958–1960) lay east of the settlement, in a narrow strip north of the Emona–Siscia road (Fig. 1). The uncovered cremation and inhumation graves (41 graves, 5 of which were in-humations) date to period from the early 1st century to the 3rd century. The western cemetery, near the village of Podgračeno, was excavated in 2002 and 2004 in the length of 95 m and width of up to 16 m. Stretching along the Roman road on its north side, it was separated from the road by a 1.5 m wide gravel-paved walking surface; orientation of all the graves respected the course of the road. A total of 129 graves were discovered, 105 of which were crema­tion graves and 24 inhumations. Apart from individual graves, four grave plots with several individual graves were explored. The documented forms of cremation graves are quite diverse: a cremation grave in a simple grave pit, a stone-covered cremation grave, a bustum, a cremation grave with an urn, an urn grave with a stone slab at the bottom, a grave pit walled with stone slabs and a stone plate on the bottom, a square chest-like grave made of stone slabs, a grave in a masonry built tomb of rectangular form (Fig. 6). Inhumation graves usually have simple grave Sl. 6: Ribnica. Del zahodne rimske nekropole s tlorisi grobnih parcel in grobov. Fig. 6: Ribnica. Part of the western Roman necropolis with ground plan of the grave plots and graves. (Arhiv / Archive: ZVKDS OE Novo mesto; izvedba / elaborated by A. Preložnik) Sl. 7: Ribnica. Zahodna rimska nekropola. Grob 7 v grobni parceli B. Fig. 7: Ribnica. Western Roman necropolis. Grave No. 7 within the grave plot B. (Arhiv / Archive: ZVKDS OE Novo mesto) polovice 1. st., so bili globlje vkopani in niso imeli kam­nite konstrukcije. Značilno je ožiganje grobne jame oz. sežig pokojnika na mestu pokopa – bustum. Mlajša faza grobišča, verjetno od druge polovice 1. st. dalje, pozna grobne konstrukcije, grobne parcele, tlakovanje, ustrine, dokumentirani so bili tudi ostanki ceste. Dve grobni parceli, označeni kot A (3,90 × 4,10 m) in B (3,60 × 4,10 m; sl. 6; 7), sta ležali neposredno ena poleg druge in sta bili pravilne, skoraj kvadratne oblike. Omejeni sta pits, sometimes there are brick-paved graves, graves in a form of a chest made of tegulae and one grave was found in a rectangular masonry built tomb. A number of ceramic, amber, metal and glass goods were found in the graves, some of them of particularly striking quality. The western cemetery shows at least two stages of use. The earlier graves from the early 1st century were dug deeper and did not have any stone construction. The cremation of the deceased at the burial site – bus­bili s kamnito ograjo iz apnenčastih lomljencev. Apnenec je bil iz kamnoloma pri Podgračenem. Na območju vzhodne nekropole je bila dokumen­tirana ustrina z več ostanki keramike (Petru 1969), na zahodni nekropoli je bilo dokumentiranih 12 ustrin, katerih oblike in dimenzije so bile različne. Manjše so merile 0,5 × 0,6 m (ustrina 6), večje pa do 1,5 × 2,0 m (ustrina 11), njihova globina je znašala od 5 do 30 cm. Dve večji ustrini, 3 in 11, bi lahko bili ostanek sežiga para pokojnikov. Med najdbami na ustrinah so poleg sežganih odlomkov kosti prevladovali ostanki konstrukcije ob sežigu – žeblji, gradbeni material, in značilni grobni pridatki: odlomki keramike, stekla, deli osebne noše. Posamični grobovi in predmeti so bili že objavljeni, med njimi jantarne najdbe iz grobov (Križ 2017) in izbrani stekleni predmeti (Lazar 2017). PREMIČNI OSTANKI POSEBNEGA POMENA Med arheološkim gradivom izstopajo nekatere najdbe iz naselbine in grobišč. Iz naselja izvira bronasta plastika Lara (sl. 8), zavetnika in varuha hiše in polj. Kakovostno izdelana bronasta plastika je datirana v 1. st. Med izjemnimi steklenimi predmeti iz naselbine velja omeniti v kalup pihano čašo z napisom mojstra << Sl. 8: Ribnica. Bronast kipec Lara – predstavljen je v frontalnem pogledu v dinamični pozi in značilnem plesnem koraku, stoji na desni nogi, leva je pokrčena in iztegnjena nazaj. Oblečen v kratko, prepasano tuniko, ki mu valovi v gibanju. V dvignjeni levici nad glavo drži pivski rog – riton, desnica je pokrčena in spuščena ob telesu. 1. st. Fig. 8: Ribnica. Bronze statue of a Lar – he is represented fron­tally, in a dynamic pose and a typical dancing step. He stands on his right leg, the left one is bent and extended backwards. He is dressed in a short belted tunic, wavy due to his move­ment. In his extended left he holds a drinking horn - rhyton, the right hand is slightly bent and turned down. 1st century. (Arhiv / Archive: ZVKDS OE Novo mesto) tum – was applied. The later phase, probably from the second half of the 1st century on, has grave constructions, grave plots, paving, ustrinae, even some remains of a road have been documented. Two grave plots, almost completely square shaped (Figs. 6; 7), designated A (3.90 × 4.10 m) and B (3.60 × 4.10 m), lay directly next to each other; they had a stone wall of limestone quarry stones. The limestone was from the quarry in Podgračeno. In the area of the eastern necropolis, an ustrina with several ceramic residues was documented (Petru 1969); in the western necropolis 12 ustrinae of different shapes and dimensions were documented, smaller ones measuring 0.5 × 0.6 m (Ustrina 6) and larger ones up to 1.5 × 2.0 m (Ustrina 11). Their depth ranged from 5 to 30 cm. Two major ustrinae, 3 and 11, could be the remains of the cremation of a deceased couple. Apart from burnt bone fragments in the ustrinae, remains of cremation construc­tion were found (nails, construction material), as well as typical grave items, such as fragments of ceramics, glass, parts of personal wear. Individual graves and objects have already been published, among them amber finds from graves (Križ 2017) and selected glass objects (Lazar 2017). SMALL FINDS OF SPECIAL IMPORTANCE Some finds from the settlement and cemeteries stand out among the archaeological material. In the settlement, a high-quality bronze statue of a Lar (Fig. 8) was found, the patron and the guardian of the house and fields. It dates to the 1st century. Among the exceptional glass objects from the settlement, the mould blown beaker with the signature of master Ennion (Fig. 9) is particularly striking; working in the east, Ennion extended his commercial network to the west and the Adriatic (Lazar 2006, 331–332). This was the first find of such a prestigious vessel in Slovenia (Lazar 2005b, 40), dating from 35 to 45 AD. An equally rare find is the bottom of a larger rect­angular bottle with the inscription of Sentia Secunda (Lazar 2005a, 42), which can be reconstructed, based Sl. 9: Ribnica. Risba Enionove čaše – V kalup pihana steklena čaša z napisom mojstra Eniona, ki je deloval na vzhodu, najver­jetneje v Sidonu, svojo trgovsko mrežo pa razširil tudi na zahod. Gre za prvo najdbo teh izdelkov pri nas, ki sodijo med najbolj prestižno stekleno posodje svojega časa. Čaša iz rumeno obarvanega stekla ima v kvadratnem okviru ohranjen napis v grščini MNH.H O A.O PAZ.N “Naj se ohrani ime kupca”. Napis na drugi strani ENNI / .NE. / OIHCE N “Enion me je izdelal”, ni ohranjen. Datirana je med letoma 35 do 45 n. št. Fig. 9: Ribnica. Drawing of the Ennion cup – the mould-blown glass cup with the inscription of the master Ennion who was active on the East, most probably in Sidon, and spread his trade network to the west. This is the first find of one of these most prestigeous glass vessels of its time found in Slovenia. The cup made of yellowish glass bears the Greek inscription in a square frame MNH.H O A.O PAZ.N “Let the buyer be remembered”. The inscription on the other side ENNI / .NE. / OIHCE N “Ennion made me” is not preserved. The cup is dated between 35 and 45 AD. (Risba / Drawing: A. Fortuna; M. / Scale = 1:2) Eniona (sl. 9), ki je deloval na vzhodu, trgovsko mre­žo pa razširil tudi na zahod in Jadran (Lazar 2006, 331–332). Gre za prvo najdbo tega prestižnega posodja pri nas (Lazar 2005b, 40), datirano v čas med letoma 35 in 45 n. št. Prav tako redka je najdba dna večje pravokotne ste­klenice z napisom Sentie Sekunde (Lazar 2005a, 42) na dnu, ki ga lahko na osnovi redkih primerjav rekonstru­iramo v napis SENTIA SECUNDA FACIT AQUILEIAE – izdelala Sentia Secunda iz Akvileje. Izdelek sodi v 1. st. V grobovih zahodne nekropole izstopajo jantarni pridatki, ki s svojo kakovostno izdelavo sodijo med izjemne in posebne najdbe. V rimski kulturi so jantarni predmeti (lat. sucinum, electrum ali glaesum) predstavljali prestiž in naj bi imeli magične in zdravilne lastnosti. V grobu 1 je bil priložen jantarni medaljon v podobi psice, ki ščiti svoje mladiče (Križ 2017, 131, št. 4). Jantarne predmete so vsebovali tudi grob 15 (ib., 62, 130, št. 1–3), grob 25 (ib., 58, 131–133, št. 5, 6, 9, 10) in grob 16 (ib., 60, 133, št. 11). on rare comparisons, as SENTIA SECUNDA FACIT AQUILEIAE – “Made by Sentia Secunda from Aquileia”. The product belongs to the 1st century. In the graves of the western necropolis, outstanding and rare amber items of particular quality stand out. In Roman culture, amber items (lat. sucinum, electrum or glaesum) stood for prestige and were thought to have magical and healing properties. In Grave 1, an amber medallion was placed in the image of a she-dog protect­ing her young (Križ 2017, 131, Nr. 4). Amber objects were also found in Graves 15 (Križ 2017, 62, 130, Nr. 1–3), 25 (Križ 2017, 58, 131–133, Nr. 5, 6, 9, 10) and 16 (Križ 2017, 60, 133, Nr. 11). DEVELOPMENT OF THE SETTLEMENT Archaeological research has defined the events in the Romula area in two overall stages of the settlement life; an earlier phase dating to the 1st and 2nd century, and the later phase in the 3rd and 4th century. RAZVOJ NASELJA Arheološke raziskave so v grobem opredelile do-gajanje na območju Romule v dve glavni fazi življenja naselbine; starejša faza, ki obsega 1. in 2. st., ter mlajša faza, zamejena med 3. in 4. st. Zahodni del naselja je sodeč po raziskanih apneni­cah, ki so ležale pod rimskimi plastmi, živel že ob koncu 1. st. pr. n. št. Naslednjo fazo, datirano v 1. st., določajo odpadne jame in ustrine, v nadaljevanju pa ta del naselja postane aktiven del naselbine z gospodarskimi objekti; arhitekturni ostanki starejše faze sodijo v 1. in 2. st., mlajša faza pa v 3. in 4. st. V osrednjem delu naselja, ki je hkrati tudi najširši del najdišča, so severno od rimske ceste ležali kultni in verjetno upravni objekti, s površino ob potoku in teraso, južno od ceste pa delavnice oz. gospodarski objekti. Kul­tni in upravni objekt sta trenutno datirana v 1. do 2. st. Vzhodni del naselbine, sodeč po ostankih verjetno lesene arhitekture neposredno ob prometnici, je prav ta­ko živel že od 1. st. dalje (Petru 1961, 194). To potrjujejo tudi najstarejši grobovi vzhodne nekropole, ki sodijo v prvo polovico in sredino 1. st. (Petru 1969). V 2. st. pa je na skrajnem vzhodnem robu zgrajen močan zidan objekt, ki varuje in kontrolira dostop do naselbine in živi dalje v 3. in 4. st. Enak časovni razpon in dve glavni fazi kažeta tudi nekropoli. Najzgodnejši grobovi so datirani v prvo polovico oz. sredino 1. st., z največjo količino pokopov v času od druge polovice 1. do konca 2. st., skeletni gro­bovi pa se pojavijo od 3. st. dalje. Slednji so po številu in vsebini skromni in verjetno že odražajo upad pomena in vloge naselja. STATUS NASELJA Romula je na Tabuli Peutingeriani zapisana kot poš­tna in obcestna postaja. Z zadnjimi izkopavanji je bilo odkrito novo epigrafsko gradivo, ki podpira verjetnost, da je na območju postaje Romula verjetno delovala tudi carinska postaja. Gre za že omenjeni posvetilni napis Silvanu Avgustu, ki ga je dal postaviti Fajder Cecilijan (sl. 2), suženj združenja zakupnikov carine ([Ph]aeder / soc(iorum) p(ublici) p(ortorii) ser(vus) / Caecilianus / 6 Silvan(o) Aug(usto) / sa(crum). Napis ROMVLA (sl. 10) na svinčeni ploščici za blago (Lovenjak 2005, 42) izpričuje ime naselbine in pravilnost njenega lociranja. 6 Najden je bil v sektorju IV, v osrednjem delu naselbine. Napis je interpretiral M. Lovenjak, glej op. 5; lupa 22475. Judging from the investigated lime-kilns under the Roman layers, the western part of the settlement existed already at the end of the 1st century BC. The next phase, dated to the 1st century, is determined by rubbish pits and ustrinae; later, this part of the settlement becomes active with economic facilities. Architectural remains of the earlier phase belong to the 1st and 2nd centuries and the later phase to the 3rd and 4th centuries. In the central part of the settlement, which is also the widest part of the site, the cult building and what was most likely an administrative building with a terrace along the stream lay on the north side of the road; south of the road, commercial buildings and workshops were built. The cult and administrative building were dated to the 1st to 2nd century. Judging by the remains of probably wooden archi­tecture right by the road, the eastern part of the settle­ment also lived from the 1st century onwards (Petru 1961, 194). This is confirmed by the earliest graves of the eastern necropolis, which belong to the first half and the middle of the 1st century (Petru 1969). In the 2nd century, a large building was built on the far eastern edge, which protects and controls access to the settlement and remained in use in the 3rd and 4th century. The same time span and two main phases are also exhibited in the necropolises. The earliest graves are dated to the first half or the middle of the 1st century, the largest amount of burials being from the second half of the 1st century to the end of the 2nd century; inhumation graves appeared in the 3rd century. The latter are scarce in number and modest in content and probably already reflect the decline of the settlement and its importance. STATUS OF THE SETTLEMENT Settlement is recorded as Romula in the Tabula Peutingeriana, i.e. a road or postal station. The new exca­vations have uncovered new epigraphic material, which supports the idea that there was a customs station in the area of Romula. The epigraphic material in question is the inscription dedicated to Silvanus Augustus,6 erected by Phaeder Cecilianus (Fig. 2), a slave to the association of publicani ([Ph]aeder / soc(iorum) p(ublici) p(ortorii) ser(vus) / Caecilianus / Silvan(o) Aug(usto) / sa(crum). The inscription ROMVLA (Fig. 10) on the lead tablet (Lovenjak 2005, 42) attests the name of the settlement and the correctness of its location. 6 It was discovered in sector IV, in the central part of the settlement. The inscription was interpreted by Lovenjak, see note 5; lupa 22475. DRUŽBENI STATUS POSAMEZNIKOV Med napisi izstopa nagrobnik Tiberija Klavdija Evdija (Tiberius Claudius Eudius) iz 1. st., ki s svojo imensko formulo tria nomina govori o polnopravnem rimskem državljanu (ILSl 1, 74). Njegov kognomen Eu-dius je zabeležen samo na tem napisu (ILSl 1, 74), zato domnevajo, da gre za grškega sužnja, ki mu je pravice podelil cesar Klavdij (Petru 1960–1961, 38). Na istem nagrobniku je ohranjeno tudi ime osvobojenca Tiberija Barbija Prima, ki je dal spomenik postaviti. Ohranjeno je tudi ime sužnja združenja zakupni­kov carine, ki je posvetil napis Silvanu Avgustu, in ime sužnja Rufina, ki je dal postaviti aro Jupitru (ILSl 1, 54). ETNIČNA PRIPADNOST Z izjemo že omenjenega grškega kognomna Eudius, ki ga drugje ne poznamo (ILSl 1, 74), drugih direktnih pričevanj o etnični pripadnosti prebivalcev Romule za sedaj ne poznamo. Ne gre pa prezreti značilnih grobnih pridatkov zahodne in vzhodne nekropole – hišastih žar, ki brez dvoma pričajo o avtohtonem prebivalstvu in nji­hovi kulturi oziroma izročilu širšega območja današnje Dolenjske in sosednje Hrvaške oz. Zagorja (Petru 1971; Gregl 1997; Križ 2003, 24; Križ, Stipančič, Škedelj Petrič 2009, 176, 357–360). Sl. 10: Ribnica. Svinčena ploščica za blago z napisom ROM­VLA / ANIONIS. Verjetno gre za ime naselja in ga beremo “iz Romule”, v drugi vrstici pa ime lastnika Anio-nis v 2. sklonu. Fig. 10: Ribnica. Lead tablet for cloth with an inscription ROMVLA / ANIONIS. Most probably this is the name of the settlement and can be read as “from Romula”, in the second line we read the name of the owner Anio-nis in the 2nd case. (Risba / Drawing: A. Fortuna; M. / Scale = 2:1) SOCIAL STATUS OF INDIVIDUALS The funerary monument to Tiberius Claudius Eudius from the 1st century stands out among the in­scriptions. Its tria nomina formula reveals a full-fledged Roman citizen (ILSl 1, 74). His cognomen Eudius is only recorded in this inscription (ILSl 1, 74), so it is as­sumed that this was a former Greek slave awarded his rights by Emperor Claudius (Petru 1960–1961, 38). The same inscription also records the name of the freedman Tiberius Barbius Primus who had the monument built. The name of the slave to the association of publi­cani is also preserved, who dedicated an inscription to Silvanus Augustus, and the name of the slave Rufinus, who erected an altar dedicated to Jupiter (ILSl 1, 54). ETHNICITY With the exception of the Greek cognomen Eudius, which is unknown elsewhere (ILSl 1, 74), there are no other direct testimonies to the ethnicity of the inhabitants of Romula. However, one shouldn’t neglect the typical grave items of the western and eastern necropolises – the house urns, which undoubtedly testify to the autochtho-nous population and their culture or the traditions of the wider area of today‘s Dolenjska and neighbouring Zagorje in Croatia (Petru 1971; Gregl 1997; Križ 2003, 24; Križ,Stipančič, Škedelj Petrič 2009, 176, 357–360). Translation: Gregor Pobežin Okrajšave / Abbreviations AIJ = V. Hoffiler, B. Saria, Antike Inschriften aus Jugoslawien 1: Noricum und Pannonia Superior. – Zagreb 1938. ANSl = Arheološka najdišča Slovenije. – Ljubljana 1975. ILJug = A. et J. Šašel, Inscriptiones Latinae quae in Jugoslavia inter annos MCMLX et MCMLXX repertae et editae sunt (Situla 19), Ljubljana1978. ILSl 1 = M. Lovenjak, Inscriptiones Latinae Sloveniae 1. Nevi-odunum (Situla 37), Ljubljana 1998. lupa = UBI ERAT LUPA, F. und O. Harl, http://lupa.at/ (Bild­datenbank zu antiken Steindenkmälern). BREŠČAK, D. 2003, Ribnica na Dolenjskem. – V / In: Zemlja pod vašimi nogami. Arheologija na avtocestah Slovenije. Vodnik po najdiščih, Zbirka Dnevi evropske kulturne dediščine, 230–232, Ljubljana (= The Earth Beneath Your Feet. Archaeology on the Motorways in Slovenia. Guide to Sites, Ljubljana 2004). BREŠČAK, D. 2004, Poročilo o arheološkem zaščitnem izkopa­vanju na arheološkem najdišču Ribnica 2001–2004 na trasi AC Krška vas-Obrežje. – Poročilo / Report; arhiv ZVKDS OE Novo mesto (neobjavljeno / unpublished). BREŠČAK, D. 2006, Romula in rimska cesta. Poštna in carin-ska postaja ob rimski cesti Emona–Neviodunum–Siscia. – V / In: Rimske ceste in projekt avtocest, Delno dopolnjen separat iz revije Rast, št. 2 (104), april 2006, 63–65. ERIČ, M. 2004, Tehnično poročilo o grafični pripravi dokumen­tacije in vizualizaciji izbranih objektov na najdiščih Ribnica in Zagorica. – Poročilo / Report; Oddelek za arheologijo, Filozofska fakulteta Univerze v Ljubljani (neobjavljeno / unpublished). GREGL, Z. 1997, Rimske nekropole sjeverne Hrvatske / The Roman Cemeteries of Northern Croatia. – Zagreb.KOLŠEK, V., P. PETRU 1958–1959, Ribnica. – Varstvo spo­menikov 7, 303–305. KRIŽ, B. 2003, Draga pri Beli Cerkvi – antična nekropola / Dra­ga near Bela Cerkev – Roman cemetery (Katalog razstave / Exhibition catalogue). – Novo mesto. KRIŽ, B. 2017, Jantar – dragulji Baltika v Novem mestu / Am­ber – Jewels of the Baltic in Novo mesto. (Katalog razstave / Exhibition catalogue). - Novo mesto. KRIŽ, B., P. STIPANČIČ, A. ŠKEDELJ PETRIČ, 2009, Ar-heološka podoba Dolenjske / The Archaeological Image of Dolenjska. (Katalog stalne arheološke razstave / Catalogue of the permanent archaeological exhibition). - Novo mesto. LAZAR, I. 2005a, Some interesting new finds from Slovenia. Fragment of the base of the bottle with the inscription SENTIA SECUNDA. – Instrumentum 21, 41–42. LAZAR, I. 2005b, Some interesting new finds from Slovenia. ENNION beaker. – Instrumentum 21, 40–41. LAZAR, I. 2006, Glass finds in Slovenia and neighbouring areas. – Journal of Roman Archaeology 19, 329–342. LAZAR, I. 2011, The world of gods and religious life in Roman Celeia. – V / In: I. Lazar (ur. / ed.), Religion in Public and Private Sphere. Acta of the 4th International Colloquium The Autonomous Towns of Noricum and Pannonia, An-nales Mediterranei, 23–38, Koper. LAZAR, I. 2017, Bogastvo in raznolikost rimskega stekla v Ribnici. – V / In: I. Lazar, B. Uroš, Romula – odsevi rimskega stekla, 7–11, Brežice. LOVENJAK, M. 2005, Roman lead tablet from Ribnica with an inscription in cursive writing. – Instrumentum 21, 42–43. MIŠKEC, A. 2004, Opredelitev novcev z najdišča Ribnica (Krško, Velika dolina, Spodnja Ribnica). – Poročilo / Report; ZVKD Slovenije, OE Novo mesto, (neobjavljeno / unpublished). MÜLLNER, A. 1879, Emona. Archäologische Studien aus Krain. – Laibach. PETRU, P. 1961, Podoba antičnega podeželja na Dolenjskem in v Spodnjem Posavju v luči najdb na avtomobilski cesti. – V / In: Dolenjski zbornik 1960–1961, 193–202, Novo mesto. PETRU, P. 1960–1961, K trem novim napisom iz Spodnjega Posavja (Zu drei neuen Inschriften aus dem unteren Sava­gebiet). – Arheološki vestnik 11–12, 27–45. PETRU, P. 1962, Ribnica. – Varstvo spomenikov 8 (1960–61), 250. PETRU, P. 1969, Rimski grobovi iz Dobove, Ribnice in Petrušn­je vasi (Römische Gräber aus Dobova, Ribnica in Petrušnja vas). – Razprave SAZU 6, 20–85. PETRU, P. 1971, Hišaste žare Latobikov (Hausurnen der Lato-biker). – Situla 11. PETRU, P. 1975, Ribnica. – V / In: ANSl, 259. ŠAŠEL, J. 1975, Rimske ceste v Sloveniji. – V / In: ANSl, 74–81. VIDRIH PERKO, V. 2004, Arheološka keramika v postizkopa­valnih in muzeoloških procesih. Študijski primer Ribnica na Dolenjskem. – Argo 47/2, 16–27. Irena Lazar Univerza na Primorskem Fakulteta za humanistične študije Titov trg 5 SI-6000 Koper irena.lazar@fhs.upr.si Manjša rimska naselja na slovenskem prostoru / Minor Roman settlements in Slovenia, Opera Instituti Archaeologici Sloveniae 40, 2020, 403–420 PRIMERJALNI PREGLED MANJŠIH RIMSKIH NASELIJ MINOR ROMAN SETTLEMENTS – COMPARATIVE OVERVIEW Jana HORVAT Izvleček V zaključnem poglavju knjige Manjša rimska naselja na slovenskem prostoru smo naselja primerjali po enotnih kriterijih ter opredelili štiri glavne skupine, ki verjetno predstavljajo najpomembnejše tipe manjših rimskih naselij na obravnavanem prostoru. To so: naselja podobna mestom, obcestne postaje, naselja odmaknjena od glavnih poti in višinska naselja, omejena na območje Notranjske. Vsa so nastala v 1. oziroma 2. st. in živela do prve polovice 5. st. Naselja iz prvih dveh skupin ležijo ob glavnih prometnih povezavah. V njih najdemo kakovostno arhitekturo, stavbe javnega značaja in epigrafske spomenike, med prebivalstvom izstopajo priseljenci iz Italije in elita, tudi staroselskega porekla. Za naselja iz drugih dveh skupin so, poleg oddaljenosti od glavnih prometnih povezav, značilne skromnejše stavbe in verjetna prevlada avtohtonega prebivalstva. Ključne besede: Italija (10. regija), Norik, Zgornja Panonija, Slovenija, rimska doba, poselitev, manjša naselja Abstract The conclusion of the book takes a comprehensive and comparative look at the minor Roman settlements in Slovenia using uniform criteria. The comparison reveals four main groups that likely represent the most significant types of minor Roman settlements in the area: minor urban settlements, roadside stations, remote settlements and hilltop settlements, the last limited to the region of Notranjska. Most were established in the 1st or 2nd centuries and abandoned in the first half of the 5th century. The settlements of the first two groups were located along the main roads and boasted buildings of quality construction, public buildings and epigraphic monuments revealing a population that included immigrants from Italy and members of the local elite, some of whom were indigenous in origin. The settlements of the third and fourth groups were removed from important roads, their buildings more modest in character and their populations presumably largely autochthonous. Keywords: Italy (Regio X), Noricum, Upper Pannonia, Slovenia, Roman period, minor settlement https://doi.org/10.3986/9789610502586_22 Knjiga Manjša rimska naselja na slovenskem prosto­ru je posvečena tistim strnjenim naseljem, ki niso imela avtonomnega statusa. Ležala so na širšem prostoru jugo­vzhodnih Alp, ki je bil razdeljen med 10. regijo Italije ter provinci Norik in Zgornjo Panonijo (sl. 1). V antičnih pisnih virih, med katerimi so najpo­membnejši Tabula Peutingeriana, Itinerarium Antonini in Itinerarium Burdigalense, so podatki o malih naseljihtega območja izjemno redki. Še vedno aktualen pregled na podlagi pisnih virov je skupaj z vprašanji lociranja podal Jaroslav Šašel.1 Poznejši arheološki pregledi pode­želske poselitve pa niso zajeli najnovejših, a neobjavlje­nih podatkov arheoloških zaščitnih raziskav.2 V knjigi je predstavljenih dvajset naselij s kakovost­nejšimi arheološkimi podatki oziroma naselij, v katerih so potekale moderne arheološke raziskave. Poglavja zgoščeno podajajo problematiko posameznih naselij, v zaključku pa želimo s primerjavo med njimi povzeti glavne ugotovitve in pogledati, kako je napredovalo naše vedenje o poselitvi prostora. V nadaljevanju so povzeta dejstva o posameznih naseljih. Natančne reference je treba poiskati v ustrez­nem poglavju; tu je navedena zgolj tista literatura, ki dopolnjuje ali drugače osvetljuje posamezne predsta­vitve. IZHODIŠČA Za razumevanje manjših naselij in sistema pose-litve, ki so ga sooblikovala, se lahko opremo predvsem na arheološke podatke, ki pa so zaradi neenakomerne stopnje raziskanosti zelo različne kakovosti. Na podlagi dejstev, ki so bila na voljo pri večini naselij, smo določili osnovne kriterije za primerjavo: - status naselja na podlagi pisnih virov, - lega v prostoru, - odnos do prometnih poti, - kontinuiteta/diskontinuiteta iz prazgodovine, - izvor in status prebivalcev, - velikost, - urbanistične značilnosti in arhitektura, - gospodarske dejavnosti, - razvoj skozi čas. Velikost naselij smo ocenili po znanem obsegu naselbinskih ostankov, pri čemer gre zgolj za boljši ali slabši približek (razpr. 1). Urbanistične značilnosti in arhitekturo naselij smo ugotavljali na podlagi gradbenih ostankov, gradbenih napisov in arhitekturnega okrasa. Pomembna se zdi prisotnost ali odsotnost monumentalnih stavb in zgradb 1 Šašel 1975. Npr. Sagadin 1995; Guštin et al. 1996; Horvat 1999; Maggi, Žbona Trkman 2007; Bavec 2009; Horvat, Sagadin 2017; Bugar, Guštin 2018; Horvat 2019. The book entitled ‘Minor Roman settlements in Slovenia’ deals with the settlements without an autono­mous status located in the wider area of the south-eastern Alps, in the Roman provinces of Noricum and Upper Pannonia (Fig. 1). These settlements are only rarely mentioned in an­cient records, notably the Tabula Peutingeriana, ItinerariumAntonini and Itinerarium Burdigalense. Jaroslav Šašel pro­vided an overview of these sources in terms of what can be gleamed of the settlements and their location, a work that remains relevant to this day.1 Settlements other than urban centres were later discussed in several other overviews, but these do not include the latest as yet unpublished results of rescue archaeological investigations.2 This book presents twenty settlements that have yielded considerable archaeological evidence through modern archaeological research. They are discussed in­dividually in several chapters, while the conclusion offers a comprehensive and comparative view of the whole and assesses the current knowledge of the settlement pattern in the area during the Roman period. The references in the overview only serve to ad­ditionally or more broadly illuminate the sites. The references pertinent to individual settlements are given in respective contributions. BASIC PREMISES Our knowledge of the minor settlements and the settlement system they helped create is mainly based on available archaeological data rather than written texts. Having said that, archaeological evidence varies from settlement to settlement due to the differences in the degree of investigation. The data available for the great majority of settlements have led us to formulate the basic comparative criteria: - status of a settlement as related in written texts, - location, - relationship towards the lines of communication, - continuity/discontinuity from prehistory, - origin and status of inhabitants, - size, - characteristics of urban design and architecture, - economic activities, - development through time. Size was estimated on the basis of the known extent of habitation remains (Tab. 1). The characteristics of urban design and architecture have been assessed using the known building remains, building inscriptions and decorated architectural mem­ 1 Šašel 1975. 2 E.g. Sagadin 1995; Guštin et al. 1996; Horvat 1999; Maggi, Žbona Trkman 2007; Bavec 2009; Horvat, Sagadin 2017; Bugar, Guštin 2018; Horvat 2019. Sl. 1: Rimska naselja na območju današnje Slovenije.Fig. 1:Roman settlements on the territory of present-day Slovenia. Naselbina / Settlement Površina / Surface Opomba / Note Hektar / Hectare Obdobje / Dating Praetorium Latobicorum 16 Mengeš 15 Most na Soči 12 Fluvio Frigido / Castra 8 1.–3. st./cent. 2,7 ha – poznorimska utrdba / Late Roman fort Nauportus 7,5 1.–3. st./cent. 9,5 ha – poznoavgustejska naselbina / Late Augustan settlement (Breg 7,5 ha, Dolge njive 2 ha) 2,3 ha – poznorimska utrdba / Late Roman fort Colatio 6,5 1,5 ha – poznorimske delavnice / Late Roman workshop area Slovenska Bistrica 6 Ulaka 5 Vipava 5 Romula 4 Ajdovščina nad Rodikom 3,9 Carnium 3–6 1. st. / cent. Longaticum 2,8 Blagovica 2,3 Atrans 2,2 Gradišče pri Knežaku 2,07 0,6 ha – dodatek na zahodu / west extension Ančnikovo gradišče 0,6 Ad Pirum 0,4 Razpr. 1: Ocenjene površine manjših rimskih naselij. Table 1: Estimated surface of the minor Roman settlements. z javnimi funkcijami. O kakovostni bivalni arhitekturi lahko sklepamo na podlagi podatkov o freskah, moza­ikih, hipokavstih in kanalizaciji. Na podlagi kriterijev, med katerimi so se pokazali kot najpomembnejši lega in gradbeni ostanki, smo na­selja primerjali in razdelili v štiri skupine. Za prve tri, naselja podobna mestom, obcestne postaje in naselja oddaljena od prometnih poti, je značilna lega v nižini. Četrto skupino predstavljajo višinska naselja na ob-močju Notranjske. Dveh kratkotrajnih naselij z začetka in konca rimske dobe (Karnij, Ančnikovo gradišče) ni bilo mogoče povezati z nobeno izmed zgornjih skupin. PREMIK POSELITVE V NIŽINE Na večini obravnavanega ozemlja, z izjemo No-tranjske in Krasa, so bile po rimski zasedbi v avgu­stejskem obdobju opuščene utrjene višinske naselbine avtohtonega prebivalstva. V nižinah in blizu prometnih poti so nastala nova naselja.3 V nobeni izmed nižinskih naselbin, predstavljenih v knjigi, ni bilo opaziti neposredne kontinuitete iz bers, most importantly the presence or absence of monu­mental buildings and buildings of a public character. The quality and comfort of living can be assessed from the data on wall paintings, mosaic floors, hypocausts and sewage system. Of the criteria, location and building remains have proven to be the most significant. The set of criteria has revealed four groups of settlements. Those of the first three groups – minor urban settlements, roadside stations and remote settlements – all lie in the lowland. Those of the fourth group lie on hilltops in the Notran­jska region. Two short-lived settlements from the begin­ning and the end of the Roman period, Carnium and Ančnikovo gradišče, could be associated with none of the four groups. SHIFT OF SETTLEMENT TO THE LOWLAND Following the Roman conquest in the Augustan pe­riod, the hillforts of the local population were abandoned across Slovenia with the exception of Notranjska and the 3 Npr. Horvat 1999, 219, 248–249; Horvat 2009, 372–373. predrimskega obdobja. Tudi sporadične prazgodovinske najdbe iz rimskih plasti je ne dokazujejo (npr. Pretorij Latobikov, Romula). Pri novonastalih naselbinah gre včasih za razmero-ma majhne premike v prostoru. V Mengšu in Šmartnem sta rimski naselbini nastali na vznožju vzpetin, ki sta bili verjetno poseljeni še v poznolatenskem obdobju. Med Velikimi bukvami, ki so živele v latenskem obdobju, in Dolenjim Logatcem, z rimsko naselbino Longatik, je okoli 2 km. Predrimski Navport, ki ga nakazuje Strabo-nov opis (7, 5, 2), ne leži na območju rimske naselbine, čeprav morda ni bil daleč. Izjema je samo naselje na Mostu na Soči, ki ga na kratko obravnavamo v nadaljevanju. Leži v dolini, a na dobro zavarovani točki. V njem se poselitev nadaljuje brez prekinitve iz predrimskega obdobja. MESTOM PODOBNA NASELJA Naselja Fluvio Frigido / Kastra, Navport, Kolacio­na in Pretorij Latobikov ležijo na ključnih geografskih ali prometnih točkah, v ravninah in ob glavnih cestah (sl. 1). Pri Fluvio Frigido / Kastra se začne cesta vzpenjati čez Julijske Alpe, Navport je umeščen na drugi strani prehoda čez Alpe in na začetek plovne poti po Ljub­ljanici. Fluvio Frigido / Kastra, Pretorij Latobikov in Kolaciona so verjetno predstavljali tudi osrednja naselja manjših geografskih območij (zgornje Vipavske doline, Mirnske doline in Mislinjske doline). Ležijo razmeroma daleč od avtonomnih mest ozi­roma naselij podobnega značaja (sl. 1). Naselje Fluvio Frigido / Kastra je oddaljeno okoli 34 km od naselja Pons Sonti in 40 km od Navporta. Razdalja od Kolacione do Celeje je 43 km, do Juene pa 37 km. Od Pretorija La-tobikov do Emone je 50 km, do Nevioduna pa 45 km. Izjema je Navport, ki je le 19 km oddaljen od Emone, vendar je spadal v okvir akvilejskega agra.4 Vsa štiri naselja so omenjena v Tabuli Peutingeriani in v itinerarijih, izmed njih Kastra kot mutatio (It. Burd. 560, 2). Navport je imel status vikusa (Tacit, Ann. 1, 20, 1), v Pretoriju Latobikov je bila beneficiarska postaja. Ocenjena velikost Pretorija Latobikov je 16 ha (ali manj). Fluvio Frigido / Kastra, Navport in Kolaciona merijo med 6,5 in 8 ha (razpr. 1). Vse naselbine so nastale na novo, brez predrimske tradicije. Navport je konec 2. ali na začetku 1. st. pr. Kr. prva rimska postojanka na prostoru osrednje Slovenije. Naselje Fluvio Frigido je verjetno nastalo že konec 1. st. pr. Kr. Najstarejša gradnja v Pretoriju Latobikov sega na začetek 1. st. Prva faza Kolacione je datirana v 1. st. Navport, Kolaciona in Pretorij Latobikov so pro-padli, tako kot ostala naselja ob glavnih poteh in v nižini, v prvi polovici 5. st. Izjema je zgolj Kastra, ki leži najbolj na zahodu in se obdrži vsaj do druge polovice 5. st. 4 Razdalje so ocenjene na podlagi sodobnih kart. Kras. New settlements sprang up in the lowland and close to the lines of communication.3 Practically none of the discussed settlements in the lowland revealed evidence of a direct continuity from the pre-Roman period. There are sporadic prehistoric artefacts, but they were recovered from Roman-period layers (for example in Praetorium Latobicorum, Romula). In some cases, new settlements grew up in close proximity to old, prehistoric ones. At Mengeš andŠmartno, for example, a Roman settlement was estab­lished at the foot of a hill presumably still settled in the Late La Tene period. Only some 2 km separates the La Tene settlement at Velike bukve and Roman Longaticum at Dolenji Logatec. Archaeological evidence suggests that pre-Roman Nauportus, mentioned by Strabo (7, 5, 2), did not lie in the area of the later Roman settlement, though possibly not very far away. The only minor exception is the settlement at Most na Soči, briefly discussed below. It does lie in a valley, but on a naturally protected promontory. Human habitation here continues into the Roman period without interruption. MINOR URBAN SETTLEMENTS The settlements Fluvio Frigido / Castra, Nauportus, Colatio and Praetorium Latobicorum are located on key spots in terms of geography and the communications network, in the lowland and along major routes (Fig. 1). Fluvio Frigido / Castra stands where the main Roman road begins its ascent across the Julian Alps, Nauportus is sited on the other side of this Alpine stretch and at the beginning of the navigable route along the Ljubljanica. Fluvio Frigido / Castra, Praetorium Latobicorum and Co-latio probably also represented the central settlements of smaller geographic units (upper valley of the River Vipava, valleys of the Mirna and the Mislinja, respectively). They lie relatively far from autonomous towns and from other settlements of a similar character (Fig. 1). Flu-vio Frigido / Castra lies some 34 km from the settlement of Pons Sonti and 40 km from Nauportus. Colatio is sited 43 km from Celeia and 37 km from Iuenna. Praetorium Latobicorum is located at a distance of 50 km from Emona and 45 from Neviodunum. Nauportus is an exception in this sense, located merely 19 km from Emona, but in a different ager – that of Aquileia.4 All four settlements appear in the Tabula Peutingeri­ana, Itinerarium Antonini and Itinerarium Burdigalense, Castra as a mutatio (It. Burd. 560, 2). Nauportus had the status of a vicus (Tacitus, Ann. 1, 20, 1), Praetorium Latobicorum hosted a statio beneficiarii. Praetorium Latobicorum covered an estimated 16 ha (or less). Fluvio Frigido / Castra, Nauportus and Colatio were smaller, covering between 6.5 and 8 ha (Tab. 1). 3 E.g. Horvat 1999, 219, 248–249; Horvat 2009, 372–373. 4 The distances are estimated using modern maps. V teh naseljih so razmeroma pogoste najdbe epi­grafskih spomenikov, ki pričajo o navzočnosti družbene elite, priseljencev iz Italije in avtohtonega romanizirane­ga prebivalstva. Odkriti so bili ostanki monumentalnih stavb z arhitekturnim okrasom, stavb z javnimi funkci­jami ter bivališč višjega bivalnega standarda (hipokav-st, mozaična tla in slikan omet). Pojavljajo se sledovi različnih obrtniških dejavnosti. Drobno gradivo kaže močan dotok dobrin in kakovostnih izdelkov iz Italije in drugih delov Sredozemlja. Gre torej za naselja na podeželju, ki so bila po­membne prometne postojanke, hkrati pa verjetno tudi lokalna gospodarska in upravna središča. Zanje velja izraz, ki ga je uporabil Tacit pri omembi vikusa v Navportu: municipii instar – kot municipij (Tacit, Ann. 1, 20, 1). Fluvio Frigido / Castra V naselju so na koncu 1. st. pr. Kr. in v 1. st. po Kr. živeli tako priseljenci iz Italije kot tudi romanizirani staroselci. Okrašeni arhitekturni členi in del napisa, ki verjetno omenja sevira, pripadajo monumentalni stavbi iz 1. ali začetka 2. st. Iz druge polovice 2. st. je znan upravnik ali nadzornik javnih posesti (saltuarius). Nagrobnik aktivnega vojaka sodi v 2.–3. st. Odkrite so bile sledi železarstva, verjetno ko­vaštva. Nagrobnik z upodobitvijo zakrivljenega noža za obrezovanje vej (prva polovica 1. st.) nakazuje obstoj vinogradov v okolici. V sedemdesetih ali osemdesetih letih 3. st. je bila zgrajena utrdba z obzidjem in stolpi, po kateri je naselje dobilo novo ime Kastra. V utrdbi, ki jo je prečila cesta, so stale velike zgradbe: osrednja stavba s portikom, skla­dišča in terme. Arhitektura in drobne najdbe vojaškega značaja kažejo, da je Kastra predstavljala eno glavnih vojaških postojank v obrambnem sistemu Claustra Alpium Iuliarum na prehodu čez Julijske Alpe. Nauportus V Navportu je že od konca 2. ali začetka 1. st. pr. Kr. prevladovala skupnost priseljencev iz Italije, ki so jo kot magistri vici vodili osvobojenci velikih trgovskih družin. V avgustejskem obdobju je bila tu vsaj občasno prisotna vojska. Avtohtone prebivalce slutimo samo v kultu Ekorne, ki so ga prevzeli priseljenci. Najpozneje v zgodnjeavgustejskem obdobju je bilo na Ljubljanici zgrajeno pristanišče, na Dolgih njivah pa je stal kompleks, utrjen z obzidjem s stolpi in obrambnim jarkom. Okoli osrednjega trga s stebriščem so bili zgraje­ni nizi skladišč in tabern, v sredi je bilo verjetno svetišče. Morda se na te zgradbe nanašata napisa iz sredine 1. st. pr. Kr., ki omenjata svetišče, posvečeno Ekorni, in portik. Skladiščni prostori so bili postavljeni tudi na drugi strani Ljubljanice, na Bregu, kjer so datirani od avgustejskega obdobja do najmanj 2. st. All these settlements were established anew, none exhibited pre-Roman habitation traces. In the late 2nd or early 1st century BC, Nauportus was the first Roman post in central Slovenia. Fluvio Frigido was probably es­tablished in the late 1st century BC, the earliest building of Praetorium Latobicorum dates to the early 1st century AD, while the first phase of Colatio is broadly attributable to the 1st century AD. Similarly as other lowland settlements along the main roads, Nauportus, Colatio and Praetorium Latobi­corum were abandoned in the first half of the 5th century. Castra is an exception here, lying farthest to the west and inhabited at least into the second half of the 5th century. These settlements revealed relatively numerous epi-graphic monuments that tell of the social elite, immigrants from Italy and autochthonous Romanised population. They left behind pieces of monumental buildings with architectural decoration, buildings of public functions and houses with a higher standard of living (hypocaust, mosaic floors, wall paintings), but also traces of various crafts that were practised there. The small finds show a sizeable import of goods and quality products from Italy and other parts of the Mediterranean. They are settlements in the countryside that func­tioned as important posts along major roads, but probably also as local economic and administrative centres. They can be ascribed the term that Tacitus used when men­tioning the vicus in Nauportus: municipii instar – like a municipium (Tacitus, Ann. 1, 20, 1). Fluvio Frigido / Castra Towards the end of the 1st century BC and in the 1st century AD, this settlement was inhabited by both immigrants from Italy and Romanised local population. Pieces of decorated architectural members indicate the presence of a monumental building from the 1st or early 2nd centuries and include a fragment of an inscription pre­sumably mentioning a sevir. A tombstone, dating to the second half of the 2nd century, was erected for a saltuarius, curator of public woodlands. Another tombstone was set up for an active soldier, sometime in the 2nd–3rd century. Excavation revealed traces of ironworking, probably smithing. A tombstone with a depiction of a billhook (first half of the 1st century) suggests that winegrowing was practised in the area. In the 270s or 280s, a fort was constructed here and enclosed within a fortification wall with towers. The fort gave the settlement its name – Castra. It hosted a thor­oughfare and large buildings: porticated central building, storehouses and baths. The architecture and small finds of a military character reveal Castra as one of the main posts in the Claustra Alpium Iuliarum barrier system protecting the passages across the Julian Alps. Odkrite so bile sledi izdelovanja bronastih pred­metov ter tekstilne obrti. Svinčene ploščice pričajo o trgovini s sadikami vinske trte, z začinjenimi ribjimi omakami ter o pošiljkah volne in tunik. Razcvet Navporta v avgustejskem obdobju je bil verjetno povezan s trgovino in oskrbo vojaških enot v osrednjem Podonavju in na severnem Balkanu. Njegov pomen usahne v prvi polovici 1. st., verjetno hkrati z upadom prometa po Ljubljanici in spremembo vojaških oskrbovalnih poti. Takrat je vodilno vlogo v regiji pre­vzela kolonija Emona. Na nekoliko dvignjenem prostoru (Gradišče) je bila verjetno konec 3. st. postavljena trdnjava in v njeni bližini opazovalni stolp. Colatio Na enem izmed nagrobnih napisov iz Kolacione je omenjen duumvir municipija Celeje, na družbeno elito kaže tudi grobnica z reliefom kurulskega stola in podobo liktorja. Imena oseb so tako latinska kot tudi keltska. Odkrito je bilo svetišče iz 2. in 3. st., s stebriščem, okrašenimi arhitekturnimi členi in votivnimi spome­niki. Na južnem robu naselja leži območje obrtniških delavnic, katerih dejavnost je ostala podrobneje neo­predeljena (3.–4. st.). Stavbe so bile preproste, z enim ali dvema prosto­roma, kamnitimi temelji in leseno nadgradnjo. Srečamo tudi enoprostorno stavbo, ki jo s treh strani obdaja hod-nik. Gre za poseben tip, značilen za jugovzhodnonoriška podeželska naselja.5 Ob naselju so bile najdene monumentalne zidane grobnice z bogatim kamnitim okrasjem in razkošnimi pridatki. Praetorium Latobicorum V Pretoriju Latobikov se močan delež keltskega pre­bivalstva kaže v krajevnem imenu in v osebnih imenih. Iz druge polovice 2. in prve polovice 3. st. so znani be-neficiariji. Gradbeni napis priča o templju posvečenem Jupitru Dolihenskemu (v letih 196–197). Ob tlakovani cesti skozi naselje so bile nanizane stavbe. Najdena je bila livarska delavnica iz druge polo-vice 1. in 2. st. V 4. in na začetek 5. st. sodijo dolge ozke hiše, namenjene bivanju in obrtniškim dejavnostim. V njih so bile ugotovljene predvsem sledi tekstilne obrti. Gre za tako imenovane dolge hiše, ki so značilne predvsem za vikuse severno od Alp, v tem prostoru pa so izjema.6 Pogojno bi lahko med naselja, ki so podobna mestom, uvrstili tudi zelo slabo poznane naselbine,katerih pomen nakazujejo kamniti spomeniki v Šem­petru v Savinjski dolini, Zagradu (obe predstavljeni v knjigi) in Staršah. 5 Groh, Sedlmayer 2006, 113–115. 6 Npr. Czysz 2013, 316–342. Nauportus The settlement of Nauportus was dominated already from the late 2nd or early 1st century BC by a community of immigrants from Italy, led as magistri vici by the freedmen of affluent merchant families. In the Augustan period, the army was present here at least occasionally. As for the indigenous population, their existence can only be per­ceived in the cult of Aecorna that the immigrants adopted. By the Early Augustan period at the latest, a harbour was constructed on the River Ljubljanica and a complex at Dolge njive enclosed within a fortification wall with towers and a defensive trench. Its central porticated square was surrounded by a series of storehouses and tabernae, a sanctuary probably stood in its centre. It may be these buildings that are referred to in two inscriptions from the mid-1st century BC, mentioning a sanctuary dedicated to Aecorna and a porticus. Storehouses also stood on the opposite bank of the Ljubljanica, at Breg; these date from the Augustan period to at least the 2nd century. The settlement yielded traces of bronzework and textile production. Lead tags tell of trading with grape­vine plants and fish sauces, and of shipments of wool and tunics. The prosperity of Nauportus in the Augustan period is probably related to trading and to supplying the Ro­man army engaged in the central Danube Basin and the northern Balkans. Its importance begins to wane during the first half of the 1st century AD, probably together with the decrease in the traffic along the Ljubljanica and the shift of the army supply lines. The colony of Emona took over the leading role in the area. Probably towards the end of the 3rd century, a fort was constructed on the slightly elevated ground at Gradišče, in its vicinity also a watch tower. Colatio One of the funerary inscriptions found at this settle­ment mentions a duumvir of the municipium of Celeia. Additional evidence of the social elite present at the settle­ment is the fragment of a funerary monument depicting a sella curulis flanked by a lictor. The names in the funerary inscriptions are both Latin and Celtic. The settlement included a sanctuary dated to the 2nd and 3rd centuries, with a porticus, decorated architectural members and votive monuments. A workshop area from the 3rd–4th century was excavated at the southern edge of the settlement, but it is not certain what kind of produc­tion took place there. Houses were simple, with one or two rooms, stone foundations and a wooden superstructure. There is also a single-room house with a corridor on three sides; this is a type of housing typical of the countryside settlements in south-eastern Noricum.5 Monumental stone-built, lavishly decorated and furnished tombs stood outside the settlement. 5 Groh, Sedlmayer 2006, 113–115. Ležijo razmeroma blizu večjih središč, Šempeter približno 14 km od Celeje, Zagrad 15 km od Kolacione oziroma 22 km od Juene, Starše 11 km od Petovione. Šempeter v Savinjski dolini, kamor je nezanesljivo locirana postaja Ad Medias, leži sredi Savinjske doline in v bližini prehoda čez Savinjo. Nagrobniki z imeni in upodobitvami kažejo na družine, priseljene iz Italije, in na staroselsko keltsko prebivalstvo. Posamezniki so imeli pomembnejše funkcije v upravi mesta Celeje (duumvir, dekurion, edil). Oltar iz let 217–218 priča o obnovi svetišča Jupitra Dolihenskega. Zagrad je umeščen v najširši del Mežiške doline. Tu so bili najdeni arhitekturni členi, ki kažejo obstoj večje stavbe, in ostanki grobnic z reliefnim okrasom, ki pričajo o višjem družbenem sloju pokojnikov. Starše ob Dravi s številnimi odlomki nagrobnih napisov z latinskimi imeni in deli grobnic nakazujejo lego pomembnejšega naselja v bližini.7 OBCESTNE POSTAJE Ob magistralnih cestah so ležala še številna druga naselja (sl. 1). Nekatera izmed njih so omenjena v iti­nerarijih oziroma na Tabuli Peutingeriani.8 V knjigi so predstavljena naselja v Vipavi, na Hrušici (Ad Pirum), v Dolenjem Logatcu (Longaticum), Blagovici (morda Ad Publicanos), na Trojanah (Atrans), v Slovenski Bistrici in Ribnici (Romula). Vsa so po pisnih virih obcestne postaje, Longatik in Atrans pa sta posebej izpričana kot mansio (It. Ant. 129, 1 in 3; It. Burd. 560, 5 in 9). Carinski postaji sta bili v Romuli in Blagovici (Ad Publicanos), na Ad Pirumu beneficiarska postaja, na Atransu pa tako carinska kot tudi beneficiarska postaja. Naselja te skupine so razmeroma majhna. Sloven-ska Bistrica, Vipava in Romula imajo površino 4–6 ha, Longatik, Blagovica in Atrans 2–3 ha, Ad Pirum pa samo 0,4 ha (razpr. 1). Nastala so v rimskem času na novo – Romula mor­da že konec 1. st. pr. Kr., Vipava, Longatik in Blagovica na začetku 1. st., Atrans v 1. st., Ad Pirum in Slovenska Bistrica v 2. st. Vsa so bila opuščena v prvi polovici 5. st. Naselje Atrans, ki je ležalo na prelazu blizu meje med Italijo in Norikom, izstopa po kakovosti stavbnih ostankov in bogatem epigrafskem gradivu. Na napisih srečamo beneficiarja in več nižjih carinskih uradnikov, ki so bili suženjskega stanu. V naselbini razmeroma majhnega obsega, ki jo omejuje ozek naravni prostor, je bilo odkritih dvanajst objektov. Kljub temu da krono­logija in tlorisi niso natančno poznani, pa značilnosti, kot so lega tik ob cesti, velike dimenzije in podolgovate oblike nekaterih stavb, kakovostni bivalni prostori z ogrevanjem in slikanim ometom ter terme, nakazujejo, 7 Pahič 1977, 56–57; Pahič 1996, 136–138. 8 Šašel 1975. Praetorium Latobicorum The name of the settlement and the personal names related on its stone monuments show that it had a high share of Celtic inhabitants. Beneficiarii were present here in the second half of the 2nd and the first half of the 3rd century. There is also a building inscription that relates the construction of a temple dedicated to Jupiter Dolichenus (in 196–197). Buildings of the settlement lined the main thor­oughfare. They included a foundry from the second half of the 1st and the 2nd century. Dating to the 4th and early 5th centuries are long narrow houses that served both as living quarters and workshops; the latter mainly intended for textile production. They are strip-houses (Streifenhaus in German) that are typical of the vici north of the Alps and exceptional here.6 Three other settlements have yielded some evidence to suggest they may also be included in this category, but the evidence only consists of stone monuments. Thesesettlements are Šempeter v Savinjski dolini, Zagrad (both presented in the book) and Starše. They all lie relatively close to larger urban centres:Šempeter at a distance of roughly 14 km from Celeia, Zagrad 15 km from Colatio and 22 km from Iuenna, Starše 11 km from Poetovio. Šempeter v Savinjski dolini, which may be the site of the Ad Medias post, is located in the centre of the valley of the River Savinja, in the vicinity of the river crossing. The funerary monuments unearthed here reveal a population of immigrant families from Italy and of an indigenous Celtic population. Their members held important ad­ministrative functions in nearby Celeia (duumvir, decurio, aedile). An altar from 217–218 also tells of a renovation of a temple dedicated to Jupiter Dolichenus. Zagrad lies in the widest part of the Mežica Val­ley. The architectural members from the area belong to a large building, while the remains of relief-decorated tombs suggest that the deceased belonged to the upper echelons of society. Starše in the Drava Valley yielded numerous frag­ments of funerary monuments bearing Latin names that indicate the existence of a larger settlement in the vicinity.7 ROADSIDE STATIONS Numerous settlements lined the main Roman roads across Slovenia (Fig. 1). Some of them are mentioned in the Roman road map and itineraries.8 Those presented in this book are at Vipava, Hrušica (Ad Pirum), Dolenji Logatec (Longaticum), Blagovica (possibly Ad Publica-nos), Trojane (Atrans), Slovenska Bistrica and Ribnica 6 E.g. Czysz 2013, 316–342. 7 Pahič 1977, 56–57; Pahič 1996, 136–138. 8 Šašel 1975. da lahko v stavbah vidimo ostanke poštne, carinske in beneficiarske postaje.9 Na Atransu je bil najden tudi cesarski gradbeni napis iz leta 168, ki priča o obnovi ali novogradnji po­membnejše javne zgradbe. Cela vrsta votivnih oltarjev, posvečenih rimskim in orientalnim božanstvom, govori o obstoju vsaj enega svetišča. Odkriti so bili še odlomki kipov iz kamna in brona ter okrašeni kamniti arhitek­turni členi. Ostanki delov bronastega konja kažejo, da je na območju prelaza stal cesarski kip na konju. Najdeni sta bili tudi dve lončarski peči. V ostalih naseljih te skupine se monumentalne stavbe ali takšne z višjim bivalnim standardom pojavijo le izjemoma. Epigrafski spomeniki in sledi družbene elite so redki. V Slovenski Bistrici je bilo v neposredni bližini ces­te odkritih več velikih stavb (dimenzije od 20 do 35 m). V 2. in 3. st. sodijo dva velika lesena objekta in zidana stavba s kvadratnim notranjim dvoriščem, okoli katerega so bili razporejeni manjši prostori. Stavbe z notranjim dvoriščem se pogosto pojavljajo v obcestnih postajah, kjer so interpretirane kot gostišča s prenočišči.10 Iz 4. st. je stavba A s podolgovatim notranjim dvoriščem in manjšimi prostori ob straneh. V njeni bližini je bila odkrita kovaška delavnica. Zelo podobne velikosti in zasnove kot stavba A je tudi prva zidana faza stavbe E (3.–4. st.). Za stavbo D v obliki dvorane z dvema vrstama stebrov, prav tako datirano v 3.–4. st., domnevajo funk-cijo skladišča. Kombinacijo podobnih stavb, takšnih s podolgovatim notranjim dvoriščem in skladišč, lahko vidimo na najdišču Biedermannsdorf blizu Vindobone, kjer naj bi bile del obcestne postaje iz 4. st.11 Vse stavbe, odkrite v Slovenski Bistrici, lahko torej povežemo s prometom, domnevajo pa tudi povezavo s kamnolomi marmorja na Pohorju in s kamnoseštvom. Kljub veliki raziskani površini ni bilo stavb, ki bi kazale na višjo raven bivanja. Epigrafskih spomenikov je malo. Razdalja okoli 500 m med poznorimskimi stavbo A na eni strani ter stavbama D in E na drugi strani kaže, da gre za dela naselbine z različnimi osnovnimi funkcijami ali s še ne dovolj pojasnjenimi kronološkimi razlikami. Romula je bila postavljena na ožini ob cesti in Savi, tik preden se svet odpre v Panonsko ravnino. Na napisih so dokumentirani rimski državljani in carinski uradnik suženjskega stanu. Hišaste žare z grobišča pričajo o sta­roselskem elementu. Odkrit je bil del velike stavbe (4) iz druge polovice 2. st., z verjetno podolgovatim notranjim dvoriščem, ki je imelo ob daljših stranicah nanizane prostore. Zgradba se po obliki in dimenzijah ujema s stavbo A iz Slovenske Bistrice. Morda gre tudi tu za tip arhitekture, značilen za obcestne postaje. V Romuli so Npr. Bender 1975; Grabherr, Kainrath 2010; Czysz 2013, 306–307; Basso, Zanini 2016. 10 Npr. Czysz 2013, 307–308; Groh 2013, 174–178. 11 Talaa, Herrmann 2003; Talaa, Herrmann 2008; Groh 2013, 175–181. (Romula). According to ancient texts, all are roadside stations, while Longaticum and Atrans are expressly men­tioned as a mansio (It. Ant. 129, 1 and 3; It. Burd. 560, 5 and 9). Romula and Blagovica (Ad Publicanos) were also customs posts, Ad Pirum had a statio beneficiarii, Atrans had both a customs post and a statio beneficiarii. The settlements of this group were relatively small. Slovenska Bistrica, Vipava and Romula covered an area of 4–6 ha, Longaticum, Blagovica and Atrans 2–3 ha, Ad Pirum only 0.4 ha (Tab. 1). They were established anew in the Roman period – Romula possibly already towards the end of the 1st cen­tury BC, Vipava, Longaticum and Blagovica in the early 1st century AD, Atrans sometime in the 1st century, Ad Pirum and Slovenska Bistrica in the 2nd century. All were abandoned in the first half of the 5th century. Atrans, located on a pass near the border between Italy and Noricum, stands out in the quality of its build­ing remains and rich epigraphic evidence. Inscriptions reveal the presence of a beneficiarius and several slaves serving as lower customs officers. The settlement was of a small size, limited by a narrow natural passage. It revealed twelve buildings. Their dating and ground plans are poorly known, but features such as location directly at the road, large size and elongated plans of some, comfortable liv­ing quarters with central heating, painted walls and baths indicate that they may represent the remains of a postal, customs and beneficiarii station.9 An imperial building inscription dated to 168 relates the construction or renovation of an important public building. A series of votive altars dedicated to Roman and Oriental deities suggests the existence of at least one sanctuary. There are also fragments of sculptures made of stone and bronze, as well as fragments of stone architec­tural members, while pieces of a bronze horse reveal that an equestrian statue of an emperor stood in the area of the pass. Also found in the settlement were two pottery kilns. In the other settlements of this group, monumental buildings or those of a higher standard of living occur only rarely. Also rare are epigraphic monuments and traces of the social elite. In Slovenska Bistrica, several large buildings (20 to 35 m long) were excavated next to the Roman road. Two large wooden buildings and one of stone, with small rooms arranged around a square inner courtyard, date to the 2nd and 3rd centuries. Buildings with an in­ner courtyard are frequently found in roadside stations, where they are interpreted as inns that also provided ac­commodation.10 Building A dates to the 4th century and had a long inner courtyard with small rooms on either side. In its vicinity was a smithery. Similar in size and design is Building E in its first masonry phase (3rd–4th century). Building D, also from the 3rd–4th century, had 9 E.g. Bender 1975; Grabherr, Kainrath 2010; Czysz 2013, 306–307; Basso, Zanini 2016. 10 E.g. Czysz 2013, 307–308; Groh 2013, 174–178. izkopali še del stavbe (2) domnevno kvadratne oblike, ki je bila obdana s hodnikom in za katero domnevajo, da morda predstavlja svetišče obhodnega tipa. Najdena je bila tudi kovaška delavnica iz 1.–2. st. Tudi v Blagovici, kjer naj bi ležala carinska in poštna postaja Ad Publicanos, so vzdolž ceste stale velike stavbe. V prvi polovici 1. st. sta bili zgrajeni dve ozki in dolgi leseni stavbi, ki ju opredeljujejo kot gospodarski poslopji (hleva ali skladišči). V drugi polovici 1. st. je bila postavljena velika lesena stavba s številnimi manjšimi prostori, obdana s portikom. V 2. in v prvi polovici 3. st. so za zidavo uporabili kamen ter postavili več manjših enoprostornih in dve veliki stavbi. Za prvo, predeljeno v velike prostore, med katerimi je imel eden dvignjena tla, lahko s precejšnjo gotovostjo rečemo, da je predstavljala skladišče. Druga je interpretirana kot ograda ali hlev. V tej fazi je bilo naselje zamejeno z zidom. V četrti fazi, ki sodi v drugo polovico 3. in v 4. st., je na območju sta­rejših stavb deloval manjši železarski obrat. Po oblikah stavb lahko torej tudi v Blagovici prepoznamo ostanke obcestne postaje. V Longatiku sta bili poleg glavne ceste, ki je vodila skozi naselje, odkriti še dve stranski tlakovani cesti. Stavbe, ki so bile le delno raziskane, so imele najprej leseno, nato zidano fazo. Čeprav poznamo le približen obseg naselja iz Vi-pave in eno samo stavbo v njem, pa lega na točki, kjer se je začela cesta vzpenjati na prelaz Okra, kaže na njegovo pomembno vlogo v prometu. Po legi na visokem prelazu in z majhnim obsegom izstopa Ad Pirum. Najstarejše stavbe, ki sodijo v 2. st., lahko povežemo z obcestno postajo. Posvetilo Jupitru, ki ga je postavil beneficiar verjetno v 3. st., kaže na ob-stoj beneficiarske postaje. Sredi druge polovice 3. st.12 ali konec 3. oziroma v prvem desetletju 4. st.13 je bila na Hrušici postavljena ena izmed utrdb obrambnega sistema Claustra Alpium Iuliarum. Stavbe v notranjosti so preslabo ohranjene, da bi lahko sklepali o njihovi namembnosti. Po koncu 4. st. je bila trdnjava opuščena in redke mlajše najdbe kažejo, da je tudi promet čez prelaz močno upadel. Naselja te skupine lahko torej opredelimo kot ob-cestne postaje. So razmeroma majhna. V njih srečamo velike objekte, postavljene neposredno ob cesto, med njimi stavbe z velikimi notranjimi dvorišči, ki so obdana z manjšimi prostori, ter različne oblike skladišč oziroma velikih, podrobno nedoločljivih gospodarskih poslopij. Pogoste so tudi kovaške delavnice. Samo naselje Ad Pirum je bilo v pozni rimski dobi utrjeno in je dobilo vojaško posadko. 12 Pröttel 1996, 134–136. 13 Kos 2012, 286–287, 297, 299: novec, kovan 312–313, predstavlja terminus ante quem; Kos 2014, 127; Kos 2015, 33–35. the interior divided by two rows of columns or posts and presumably functioned as a storehouse. A similar com­bination of buildings, those with a long inner courtyard and storehouses, is known from Biedermannsdorf, a site near Vindobona, which have been interpreted as part of a roadside station from the 4th century.11 All buildings excavated in Slovenska Bistrica can be linked to traffic. The settlement was also connected with the marble quarries in the Pohorje Hills and with stone-masonry. In spite of the sizeable scale of excavations, the site revealed no buildings that would exhibit higher living standards. The site also yielded very few inscriptions. A distance of some 500 m between Late Roman Building A, on one side, and Buildings D and E, on the other, show that we are dealing with two parts of the settlement with either different basic functions or as yet unclear chrono­logical differences. Romula was located in the narrow space between the Roman road and the River Sava, just before the terrain opens onto the Pannonian Plain. The inscriptions from the site reveal the presence of Roman citizens and a slave serving as a customs officer. On the other hand, the house urns from the local cemetery represent an indigenous element. Excavations unearthed part of a large building (4) from the second half of the 2nd century, probably with a long courtyard with rooms opening to either side. In layout and size, it corresponds with Building A from Slovenska Bistrica. Here as well, it may represent archi­tecture typical of roadside stations. A part of another building (2) excavated at Romula suggests a square plan surrounded by a porticus, which may have been an am­bulatory sanctuary. The settlement also had a smithery from the 1st–2nd century. Blagovica, the presumed site of the Ad Publicanos postal and customs station, also revealed large buildings lining the road. Two narrow and long wooden buildings were constructed in the first half of the 1st century AD, interpreted as either stables or storehouses. In the second half of the 1st century, a large wooden building with nu­merous small rooms and surrounded by a porticus was constructed. Masonry buildings were constructed in the 2nd and first half of the 3rd century, consisting of several single-room and two large buildings. One of the large buildings partitioned into sizeable rooms, one with a raised floor, was almost certainly a storehouse. The second large house is interpreted as an enclosure or a stable. In this phase, the settlement was enclosed with a boundary wall. In Phase 4, dated to the second half of the 3rd and the 4th century, a small ironworks operated in the area of earlier buildings. The form of the buildings suggests that the remains at Blagovica can also be identified as those of a roadside station. Investigations of Longaticum unearthed the main road through the settlement, as well as two minor gravel 11 Talaa, Herrmann 2003; Talaa, Herrmann 2008; Groh 2013, 175–181. Antični pisni viri naštevajo tudi različna obcestna naselja, ki še niso bila natančno locirana (prim. sl. 1).14 Ob cestah je bilo odkritih veliko arheoloških najdišč iz rimske dobe, vendar je ob neizrazitih naselbinskih najdbah težko ugotoviti, za kakšno vrsto poselitve gre. Sodimo torej, da poznamo le manjši del strnjenih naselij ali postaj, ki so ležala ob glavnih cestah. NASELJA, ODMAKNJENA OD GLAVNIH POTI V ravnini in odmaknjena od glavnih prometnih po-ti ležijo rimska naselja v Mengšu, na Igu in morda tudi vŠmartnem pri Cerkljah (sl. 1). V njih ni monumentalnih stavb in tudi ne stavb, v katerih bi lahko prepoznali jav­ne funkcije. Stavbe višje kakovosti so redke, prav tako epigrafski spomeniki z omembami oseb, ki so sodile v družbeno elito. V prazgodovinski naselbini na Gobavici nad Mengšem, ki je verjetno živela do prihoda Rimljanov, je bila v avgustejskem obdobju nastanjena manjša vojaška posadka. Hkrati se je v Mengšu na vznožju hriba razvilo novo naselje. Glede na domnevan obseg 15 ha je bilo dokaj veliko. Notranja ureditev in oblika posameznih stavb, ki so bile zidane in lesene, ni poznana. Odkriti so bili močni sledovi železarstva. Tudi na utrjeni prazgodovinski naselbini na Straži pri Šmartnem je bila v drugi polovici 1. st. pr. Kr. in na začetku 1. st. po Kr. verjetno nastanjena manjša vojaška enota. Novo rimsko naselje, ki je nastalo ob vznožju hriba, je skoraj popolnoma nepoznano. Na njegovem obrobju je bil odkrit poznorimski železarski in kovaški obrat. Večji pomen naselja se kaže v domnevnih ostan­kih cerkve z mozaičnim tlakom iz konca 4. st. Velikost in ureditev rimskega naselja na Igu nista znani. Njegova posebnost so številni nagrobni spome­niki. Osebna imena kažejo, da je na Igu prevladovalo lokalno prebivalstvo brez rimskega državljanstva. Pol-nopravnih rimskih državljanov je bilo malo – po enkrat sta izpričana emonski duumvir in sevir. Na območju Iga so delovale kamnoseške delavnice, o obstoju kovinarske delavnice govori upodobitev različnih izdelkov na enem izmed nagrobnikov. Verjetno lahko v to skupino uvrstimo tudi rimsko naselbino na Mostu na Soči, postavljeno v nižini, na dobro zavarovani točki na stiku dveh dolin. V nasprotju z drugimi rimskimi naselbinami v dolinah je ležala na mestu naselbine iz starejše in mlajše železne dobe. Na prostoru, velikem okoli 12 ha, so odkrili posamezne, razpršeno postavljene preproste stavbe. Kamniti zidovi so bili vezani z malto, ometani, kritina je bila opečnata. V nekaterih stavbah so bili ostanki hipokavsta. Naselbina je živela do 4. st.15 V bližini je bilo odkrito grobišče iz 14 Šašel 1975. 15 Maggi, Žbona Trkman 2007, 68–70; Mlinar et al. 2012. roads. The buildings of the settlement were only partially investigated, revealing a wooden phase in the beginning followed by masonry constructions. The settlement in Vipava is poorly known. The avail­able evidence allows a rough estimation of its extent; one house has also been investigated. Its location, however, on the spot where the road across the Ocra Pass began its ascent, suggests it played an important part in the traffic along this road. Standing out in its high altitude on the Hrušica Pass and in small size is Ad Pirum. Its earliest buildings, from the 2nd century, can be associated with a roadside station. A dedication to Jupiter set up by a beneficiarius probably in the 3rd century indicates the existence of a statio benefi­ 12 ciarii. Either in the middle of the second half of the 3rd,the end of the 3rd or in the first decade of the 4th century,13 a fort was constructed here that functioned as part of the Claustra Alpium Iuliarum barrier system. The buildings in its interior have survived to an insufficient degree to shed light on their function. After the end of the 4th century, the fort was abandoned and rare small finds show that traffic across the pass almost ceased. The settlements of this group have been interpreted as roadside stations. They are relatively small and consist of large buildings lining a main road. These buildings include those with large inner courtyards surrounded by small rooms, different storehouses and other, precisely undeterminable outhouses, frequently also smitheries. Only Ad Pirum was fortified in the Late Roman period and hosted an army garrison. Ancient authors mention a number of other roadside stations, but they have either not yet been discovered or cannot be positively identified as such (cf. Fig. 1).14 Ar­chaeological excavations have unearthed many sites that are associated with roads, but their habitation remains are not sufficiently diagnostic in terms of the type of set­tlement. It is safe to conclude that we thus far only know of a small portion of all the settlements or stations that lined the Roman roads. REMOTE SETTLEMENTS Located in the lowland, but away from the main roads are the Roman settlements at Mengeš, Ig and pos­sibly also Šmartno near Cerklje (Fig. 1). They revealed neither monumental buildings nor those that we may as­sume to have had a public function. Buildings of a higher standard of living are rare, as are inscriptions mentioning individuals of the social elite. 12 Pröttel 1996, 134–136. 13 Kos 2012, 286–287, 297, 299: the coin minted in 312– 313 serves as terminus ante quem; Kos 2014, 127; Kos 2015, 33–35. 14 Šašel 1975. v 1. st. pr. Kr. in 1. st. po Kr., z žganimi in skeletnimi pokopi.16 Tu najdena nagrobnika iz sredine ali druge polovice 1. st. pr. Kr., eden omenja vojaka 15. legije, pri-čata o zgodnjem priseljevanju iz Italije.17 Na nasprotnem bregu reke Idrijce sta ležala še eno grobišče in daritveno mesto, ki sta se neprekinjeno nadaljevala iz latenske v zgodnjo rimsko dobo.18 Vse kaže, da je na Mostu na Soči stalo naselje s prevlado avtohtonega prebivalstva. Podobnih naselbin avtohtonega prebivalstva, ki so bile oddaljene od glavnih prometnih poti, je bilo verjetno še več, le da arheološko še niso prepoznane. NOTRANJSKA VIŠINSKA NASELJA Nekatera prazgodovinska višinska naselja z ob-močja jugozahodne Slovenije (del Notranjske, morda tudi Kras) so živela neprekinjeno še v rimsko obdobje. Med njimi so najbolje poznana Ulaka, Gradišče pri Knežaku (obe predstavljeni v knjigi) in Ajdovščina pri Rodiku. Podobna, a slabše raziskana višinska naselja s kontinuiteto navajata v tej knjigi Andrej Gaspari in Boštjan Laharnar (s soavtoricama): Kerin nad Pivko, 16 Svoljšak, Žbona Trkman 1985. 17 Šašel 1985; Svoljšak, Žbona-Trkman 1986, 390–391. Mlinar 2005; Gaspari, Mlinar 2005, 169–170, 181; Gaspari 2007, 151–152; Laharnar, Mlinar 2015. The prehistoric settlement on the hill of Gobavica above Mengeš, which presumably existed until the arrival of the Romans, received a small army garrison in the Au­gustan period. At the same time, a new settlement sprang up at the foot of the hill, at Mengeš. Given its presumed extent of 15 ha, this settlement was quite large. The build­ings were both masonry and wooden, but their interior layout and form is unknown. Excavations also revealed considerable traces of ironworking. The fortified prehistoric settlement on the hill ofStraža near Šmartno also presumably hosted a small army unit in the second half of the 1st century BC and the early 1st century AD. The new, Roman settlement grew at the foot of the hill, but almost nothing is known of it. A Late Roman ironworking and smithing workshop was found at its outskirts. The settlement must have been significant, however, as suggested by the presumed remains of a church with a mosaic floor from the 4th century. The Roman settlement at Ig is of an unknown size and layout. It stands apart in the great number of tombstones. Personal names indicate a predominantly autochthonous population without Roman citizenship. Very few individuals had full Roman citizenships – one is known to have been a duumvir and another a sevir of Emona. Stonemasonry was practised in the area of Ig, the decoration on one of the tombstones also suggests the existence of a metalworking workshop. The Roman settlement at Most na Soči probably also falls into this group. It was set up in the lowland, on a naturally protected spot at the junction of two river valleys. As opposed to other Roman lowland settlements, it stood on the very spot of the earlier settlement from the Early and Late Iron Ages. The inhabited area of roughly 12 ha has revealed dispersed simple buildings with mortar-bound and plastered stone walls, and brick tiles as roofing, some with underfloor heating. The set­tlement lived to the 4th century.15 A cemetery from the 1st century BC and the 1st century AD was found in the vicinity, with both cremations and inhumations.16 Two tombstones are known from here, dating to the middle or second half of the 1st century BC and indicating early immigration from Italy; one mentions a soldier of the Fifteenth Legion.17 Another Roman cemetery, as well as an offering place was found on the opposite bank of the River Idrijca, which continued without interruption from the La Tene to the Roman period.18 Evidence suggests that the settlement at Most na Soči was predominantly autochthonous in population. Apart from these, there must have been numerous other settlements of the autochthonous population across 15 Maggi, Žbona-Trkman 2007, 68–70; Mlinar et al. 2012. 16 Svoljšak, Žbona-Trkman 1985. 17 Šašel 1985; Svoljšak, Žbona-Trkman 1986, 390–391. 18 Mlinar 2005; Gaspari, Mlinar 2005, 169–170, 181; Gaspari 2007, 151–152; Laharnar, Mlinar 2015. Šilentabor, Gradišče na Čepni, Gradišče nad Trnovim v Ilirski Bistrici, Gradišče nad Gornjo Košano in Sv. Pavel nad Planino (sl. 2). Površine bolj znanih naselij merijo 5 ha (Ulaka), 3,9 ha (Ajdovščina nad Rodikom) in 2,07 ha (Gradišče nad Knežakom) (razpr. 1). O pomenu Ulake kot središča avtohtonega prebival­stva poleg velikosti, osrednje geografske lege in kontinui­tete poselitve pričajo tudi rimski tabori v bližini in sledi spopada iz sredine ali druge polovice 1. st. pr. Kr. Imena oseb z rimskih nagrobnikov iz širše soseščine kažejo na avtohtono, a zgodaj romanizirano prebivalstvo. V rim-skodobni naselbini se nizi stavb koncentrično širijo okoli osrednjega nepozidanega prostora. Zgradbe višjega bival­nega standarda niso bile dokumentirane. Hiše, ki so bile vkopane v tla, s kamnitimi temelji in leseno nadgradnjo, imajo korenine v regionalnem železnodobnem stavbar­stvu. Ostanki več kovaških delavnic, žlindre in orodja kažejo, da je bila naselbina specializirana na železarsko proizvodnjo. Odkriti so bili še sledovi predelave brona, domnevajo pa tudi lončarsko delavnico. Rimskodobna Ulaka je doživela prvi vzpon konec 2. in na začetku 3. st. in drugi vzpon v sredini in drugi polovici 4. st. Novčne najdbe kažejo obstoj vsaj do začetka 5. st. Tudi Gradišče nad Knežakom je živelo neprekinje-no iz poznega latena v rimsko dobo. V drugi polovici 3. in v 4. st. je bila naselbina gosto pozidana s podolgovati-mi stavbami z več prostori, ki so bile razporejene v nizih. V bližini naselja je stal v pozni rimski dobi opazovalni stolp. Novčni obtok kaže na prekinitev poselitve še pred sredino 4. st. Na koncu predrimske in na začetku rimske dobe je bilo na Ajdovščini nad Rodikom verjetno središčeplemena Rundiktov. Življenje na gradišču se je nepreki­njeno nadaljevalo vsaj do 2. ali 3. st.19 V poznem 4. in do sredine 5. st. je bila naselbina gosto pozidana s stavbami z več prostori (sl. 3). Terenski pregledi so pokazali močne sledi predelave železa, ki jih morda lahko povežemo z zadnjo fazo življenja.20 Prostor, na katerem ležijo omenjene višinske na­selbine, se večinoma ujema z vzhodnim delom ozemlja kolonije Tergeste. Severno od tod, vzdolž ceste iz Akvi­leje v Emono, je bilo življenje na višinskih naselbinah prekinjeno v avgustejskem obdobju ali najpozneje na začetku 1. st. Opustitev starih železnodobnih naselbin konec 1. st. pr. Kr. je vidna tudi na drugih delih območja današnje Slovenije. Razlika v poselitvi med Notranjsko in ostalimi predeli je verjetno odsev različnih odnosov rimske države do avtohtonih skupnosti. Zadnji razcvet notranjskih višinskih naselbin v po­znorimskem obdobju je bržkone povezan z lego v zaledju obrambnega sistema Claustra Alpium Iuliarum, ko so bile verjetno vključene v njegovo logistično podporo. 19 Istenič 1987; Istenič 1988; Vedaldi Iasbez 1994, 272– 274; Vidrih Perko 1997; Slapšak 1997; Slapšak 2003, 247–248, 254; Štular, Lozić 2016, 163–164. 20 Mušič 1999, 356–370, 399–401. Slovenia, located away from the main roads, but they have as yet not been archaeologically detected. HILLTOP SETTLEMENTS IN THE NOTRANJSKA REGION Some of the prehistoric hillforts in south-western Slovenia (part of Notranjska, possibly also the Kras) continued to be inhabited into the Roman period. The best known among them are Ulaka, Gradišče near Knežak (both presented in this book) and Ajdovščina near Rodik. Andrej Gaspari and Boštjan Laharnar (with co-authors) cite other, less well researched hillforts with continu­ity: Kerin above Pivka, Šilentabor, Gradišče on Čepna, Gradišče above Trnovo in Ilirska Bistrica, Gradišče above Gornja Košana and Sv. Pavel above Planina (Fig. 2). The better known hillforts cover a surface of 5 ha (Ulaka), 3.9 ha (Ajdovščina above Rodik) and 2.07 ha (Gradišče above Knežak) (Tab. 1). The significance of Ulaka as a centre of the local population is visible in its size, central location and habi­tation continuity, but also Roman forts in proximity and traces of an armed conflict from the middle or second half of the 1st century BC. The names on the tombstones recovered in the wider area indicate an autochthonous population that was Romanised early on. The Roman-period settlement consists of concentric rows of buildings around the empty central space. None of the investigated buildings was found to offer a higher living standard. The houses had sunken floors, stone foundations and a wooden superstructure, as such rooted in the regional Iron Age building tradition. The remains of several smitheries, of slag and tools indicate that the inhabitants specialised in ironworking. Also recovered were traces of bronze processing, presumably even pottery making. Its first peak of prosperity came towards the end of the 2nd and in the early 3rd century, the second in the middle and second half of the 4th century; coin finds show that it was inhabited at least to the early 5th century. The settlement on Gradišče above Knežak also lived without interruption from the Late La Tene to the Roman period. In the second half of the 3rd and the 4th century, it was densely built-up with rows of long houses with several rooms. A watchtower stood near the settlement in the Late Roman period. The monetary circulation shows that the site was abandoned before the mid-4th century. At the end of prehistory and the beginning of the Roman period, the hill of Ajdovščina above Rodik was probably the centre of the Rundictes tribe. Life at this hillfort continued uninterruptedly at least to the 2nd or 3rd centuries.19 In the late 4th to the mid-5th century, the settlement was densely built-up with houses of several 19 Istenič 1987; Istenič 1988; Vedaldi Iasbez 1994, 272– 274; Vidrih Perko 1997; Slapšak 1997; Slapšak 2003, 247–248, 254; Štular, Lozić 2016, 163–164. Carnium V Kranju, na dobro zavarovanem pomolu nad sotočjem Save in Kokre, je stalo v srednje- in poznoavgu­stejskem obdobju naselje, veliko 3–6 ha, ki se je verjetno že imenovalo Carnium. Drobno gradivo kaže, da so tu živeli priseljenci iz Italije, sporadično je bila navzoča vojska. Posamezne latenske najdbe so bile najdene v istih slojih kot rimskodobne. Naselje je bilo obzidano s kamnitim obzidjem s stolpi. Stavbe v notranjosti, kate­rih tlorisi sicer niso ohranjeni, so bile večinoma lesene. Ostanki zidov in fresk kažejo tudi obstoj kakovostnejših stavb. Karnij je bil v avgustejskem obdobju ena prvih rimskih civilnih naselbin v vzhodnoalpskem svetu, ki so bile zavarovane z obzidjem in v katerih so prevla­dovali prišleki iz Italije. Po teh značilnostih je podoben Navportu, Gurini in Štalenski gori.21 Opustitev Karnija po koncu avgustejskega obdobja je verjetno povezana s spremenjenimi prometnimi tokovi ter družbenimi in gospodarskimi razmerami v regiji na začetku 1. st. Karnij je znova oživel v 4. st., ko je spet postala zanimiva dobra obrambna lega in so se vzpostavile nove prometne poti. Postavljene so bile zidane stavbe z mozaiki in ogrevanjem. Ta faza poselitve se je brez prekinitve nadaljevala v srednji vek. rooms (Fig. 3). Field surveys have revealed considerable traces of ironworking, possibly associated with the last phase of life at the site. 20 These hilltop settlements lie roughly in the east part of the territory belonging to the colony of Tergeste. To the north of here, along the road from Aquileia to Emona, life on such elevated settlements ceased in the Augustan period or the early 1st century AD at the latest. Old Iron Age hillforts were abandoned towards the end of the 1st century BC also in other parts of Slovenia. This difference in the settlement pattern between Notranjska and the rest of Slovenia may reflect a different relationship of the Roman state towards the local communities. The last period of prosperity of the Notranjska hill­top settlements may be connected with their location in the hinterland of the Claustra Alpium Iuliarum defence system, probably included in its logistic support. Carnium At Kranj, on a naturally well protected promontory overlooking the confluence of the Rivers Sava and Kokra, a settlement covering 3–6 ha stood in the Middle and Late Augustan periods that was probably called Carnium already at this early time. Small finds show that it was inhabited by immigrants from Italy with the occasional presence of the army. Individual La Tene artefacts were found in the same layers as those from the Roman period. The settlement had a stone fortification wall with towers. Most buildings in its interior, of unknown ground plans, were wooden, while remains of walls and wall paintings also reveal the presence of better constructed houses. In the Augustan period, Carnium was one of the first civilian Roman settlements in the eastern Alpine area with fortifi­cation walls and largely inhabited by Italian immigrants. In these characteristics, it resembles Nauportus, Gurina and Magdalensberg.21 Its abandonment after the end of the Augustan period is probably linked to the changes in the lines of communication, as well as the changes in the social and economic conditions in the region at the beginning of the 1st century AD. Carnium again prospered in the 4th century, when its naturally protected location and new communications again gave an important advantage. At this time, masonry buildings with mosaic floors and underfloor heating were constructed. This habitation phase lasted without inter­ruption into the Middle Ages. 20 Mušič 1999, 356–370, 399–401. 21 Horvat 2010. 21 Horvat 2010. Ančnikovo gradišče Ančnikovo gradišče je majhna utrjena naselbina (0,6 ha) na zahodnem robu petovionskega ozemlja, ki je bila postavljena na novo v zadnji tretjini 4. st. in je propadla v požaru pred sredino 5. st. Obzidana je bila s kamnitim obzidjem, v notranjosti so bile večinoma lesene stavbe, deloma pa tudi lesene s kamnitimi temelji. Poleg civilnega prebivalstva je bila tu nastanjena tudi manjša vojaška posadka. Ključna za nastanek Ančniko­vega gradišča je bila odmaknjena lega na južnih obron­kih Pohorja, ki pa je še vedno omogočala nadzor nad dolino in magistralno cesto Celeja–Petoviona. Morda je določeno vlogo pri nastanku igrala tudi neposredna bližina ležišč marmorja v soteski Bistriškega vintgarja. Ančnikovo gradišče je eno najstarejših naselij, ki so jih v pozni rimski dobi postavljali na oddaljenih višinskih legah, ko je poselitev v nižinah postopno zamirala. SKLEP Na podlagi podatkov, predstavljenih v knjigi, lahko manjša rimska naselja razdelimo v štiri glavne skupine: pomembnejša naselja, ki so bila podobna mestom, ob-cestne postaje, naselja, odmaknjena od glavnih poti, in višinska naselja z območja Notranjske. Vsa so živela od 1. oziroma 2. st. do splošne opustit­ve nižinske poselitve v prvi polovici 5. st. Naselja v prvih dveh skupinah, to je pomembnejša naselja in obcestne postaje, ležijo ob glavnih prometnih povezavah, v njih najdemo epigrafske spomenike, kakovostno arhitekturo in stavbe z javnimi funkcijami, med prebivalstvom iz­stopajo priseljenci iz Italije in elita avtohtonega porekla. Za naselja drugih dveh skupin, to je za nižinska naselja, odmaknjena od glavnih poti, in za notranjska višinska naselja, so poleg oddaljenosti od glavnih prometnih povezav značilni skromnejši stavbni ostanki in verjetna prevlada avtohtonega prebivalstva. Zdi se, da te štiri skupine predstavljajo najpo­membnejše tipe manjših rimskih naselij na obravna­vanem prostoru. Zaradi majhnega števila analiziranih naselij ter neenakomerne oziroma slabe raziskanosti rimskega podeželja nam v pregledu verjetno ni uspelo zajeti vseh oblik strnjene poselitve. Dveh kratkotrajnih naselij z začetka in konca rim-ske dobe (Karnij, Ančnikovo gradišče) ni bilo mogoče povezati z nobeno izmed zgornjih skupin. V predavgustejskem in avgustejskem obdobju izstopata Navport in Karnij kot zgodnji, dobro utrjeni oporišči rimskih naseljencev. Navport je imel ključno vlogo na poti iz Italije proti srednjemu Podonavju in je predstavljal tako trgovsko oporišče kot tudi oskrbovalno postojanko za vojaške podvige v Panoniji in na severnem Balkanu. Vloga Karnija ni jasna. Morda je povezana z Ančnikovo gradišče This was a small, 0.6 ha large fortified settlement at the western edge of the territory of Poetovio that was established anew in the last third of the 4th century. It was abandoned after a devastating fire before the mid-5th cen­tury. It had a stone fortification wall. Its interior revealed mostly wooden houses, some with stone foundations. It was inhabited by civilians, but also hosted a small army garrison. The main reason for the construction of this settlement was its remote location on the southern fringes of the Pohorje Hills and a vantage point that allowed the control over the valley below and the main road from Ce-leia to Poetovio. It may also in part be connected with the marble deposits in the immediate vicinity, in the gorge of Bistriški vintgar. Ančnikovo gradišče is one of the earliest settlements established in remote and elevated locations in the Late Roman period, at a time when settlements in the lowland gradually began to wane. CONCLUSIVE REMARKS The evidence presented for individual sites in this book suggests four main groups of settlements: more prominent minor urban settlements, roadside stations, settlements removed from the main lines of communica­tion and hilltop settlements in the region of Notranjska. They were all inhabited from at least the 1st or 2nd century to the first half of the 5th century, when lowland settlements were generally abandoned. Those of the first two groups were located along the main roads and left behind epigraphic evidence, remains of well-built houses and buildings of a public character, while their inhabitants included immigrants from Italy and members of the local elite. The settlements of the other two groups, in remote locations and on the hilltops of Notranjska, also removed from the main communications, have revealed more modest building remains and other finds that suggest a predominantly autochthonous population. The four groups appear to represent the most promi­nent types of minor Roman settlements in Slovenia. Hav­ing said that, the number of analysed settlements is small and the investigation of the Roman countryside uneven or poor, hence there are probably forms of habitations that have escaped our attention. Two of the settlements of short duration from the beginning and the end of the Roman period, namely Carnium and Ančnikovo gradišče could be ascribed to none of the four groups. Standing out in the pre-Augustan and Augustan periods are Nauportus and Carnium, solidly fortified posts of the Romans. The former played a major role on the route from Italy towards the central Danube Basin both as a trading post and a supply station for the Roman military effort in Pannonia and the northern Balkans. The rimsko strategijo zasedbe prostora severno od Emone in s prometno smerjo od Foruma Iulii naravnost proti vzhodu. Po koncu avgustejskega obdobja, ko so se po­litične in gospodarske razmere bistveno spremenile, se je pomen Navporta močno zmanjšal, naselje v Karniju pa je celo ugasnilo. Od konca 3. do začetka 5. st. je na območju preho­dov čez obrobje Julijskih Alp deloval obrambni sistem Claustra Alpium Iuliarum, ki je ščitil Italijo pred vpadi z vzhoda. V njegovem okviru so stara naselja dobila nove vloge. Zgrajene so bile trdnjave Kastra, Ad Pirum in Navport. Za trdnjavo Kastra (stari Fluvio Frigido) domnevajo, da je predstavljala osrednjo postojanko ob-rambnega sistema. Utrdba v Navportu je zapirala glavno pot na vzhodni strani prehodov, tista na Ad Pirumu pa na najvišji točki.22 Hkrati se je življenje v notranjskih višinskih naseljih močno okrepilo, kar pomeni, da so verjetno pridobila veljavo kot zaledne postojanke ob-rambnega sistema. V istem obdobju, od konca 3. st. dalje, so bile vo­jaške postojanke postavljene tudi zunaj ožjega območja Claustra Alpium Iuliarum. Proti koncu 4. st. je prišlo do večjih premikov prebivalstva na varnejše višinske točke, čeprav so naselja v ravnini še intenzivno živela. V prvi polovici 5. st. so bila naselja ob glavnih poteh opuščena. Vnovična poselitev Karnija na naravno dobro zavaro­vanem mestu sodi v to prelomno obdobje. Ančnikovo gradišče pa konec 4. in v prvi polovici 5. st. predstavlja eno prvih višinskih vojaških postojank in hkrati pribe­žališč civilnega prebivalstva.23 strategic role of Carnium is unclear; it may be linked to the Roman conquest of the area north of Emona and to securing the communication that led eastwards from Fo­rum Iulii. After the end of the Augustan period, when the political and economic situation changed substantially, the role of Nauportus decreased drastically, Carnium was even abandoned. From the late 3rd to the early 5th century, the Claustra Alpium Iuliarum defence system was set up along the passages across the fringes of the Julian Alps with the aim of protecting Italy from incursions from the east. As part of this system, old settlements were given new roles. The forts of Castra, Ad Pirum and Nauportus were constructed. Castra (earlier Fluvio Frigido) is believed to have functioned as the central post of the defence system. The fort of Nauportus guarded the main route at the east end of the Julian passages; the fort of Ad Pirum guarded the same route at its highest point.22 At the same time, the hilltop settlements of Notranjska became more densely inhabited, indicating that they gained in importance as posts in the hinterland of the same defence system. From the late 3rd century onwards, military posts were also set up outside the narrow zone of Claustra Al-pium Iuliarum. Major shifts occurred towards the end of the 4th century, when people began moving to safer, higher locations, although the lowland settlements continued to live on. Finally, in the first half of the 5th century, the settlements along the main roads were abandoned. The renewed settlement of Carnium, located on a naturally well protected promontory, falls in this transitional time. On that note, Ančnikovo gradišče is one of the first hilltop military posts from the end of the 4th and the first half of the 5th century that also served as a refuge for the civilian population.23 Translation: Andreja Maver 22 Ciglenečki 2015; 2016. 22 Ciglenečki 2015; 2016. 23 Ciglenečki 2015, 403–414. 23 Ciglenečki 2015, 403–414. BASSO, P., E. ZANINI (ur. / eds.) 2016, Statio amoena. Sostare e vivere lungo le strade romane. – Oxford. BAVEC, U. 2009, Rimljani ob veliki reki – poskus opisa rimske poselitve Posavja v času od 1. do 4. stoletja (The Romans at the great river – an atempt to describe the Roman settling of the Posavje region in the period from the 1st until 4th century). – V / In: Ukročena lepotica. Sava in njene zgodbe, 51–68, Sevnica. BENDER, H. 1975, Römische Straßen und Straßenstationen. – Keine Schriften zur Kenntnis der römischen Besetzun­gsgeschichte Südwestdeutschlands 13, Stuttgart. BUGAR, A., M. GUŠTIN 2018, The Roman Countryside of Prekmurje (Rimsko podeželje Prekmurja). – V / In: M.Janežič, B. Nadbath, T. Mulh, I. Žižek (ur. / eds.), Nova odkritja med Alpami in Črnim morjem / New Discoveries Between the Alps and the Black Sea, Monografije CPA 6, 45–63. CIGLENEČKI, S. 2015, Late Roman army, Claustra Alpium Iuliarum and the fortifications in the south-eastern Alps / Poznorimska vojska, Claustra Alpium Iuliarum in utrjena krajina v jugovzhodnih Alpah. – V / In: J. Istenič, B. La-harnar, J. Horvat (ur. / eds.), Evidence of the Roman Army in Slovenia / Sledovi rimske vojske na Slovenskem, Katalogi in monografije 41, 2015, 385–430. CIGLENEČKI, S. 2016, Claustra Alpium Iuliarum, tractus Italiae circa Alpes and the defence of Italy in the final part of the Late Roman period. – Arheološki vestnik 67, 409–424. CZYSZ, W. 2013, Zwischen Stadt und Land – Gestalt und Wesen römischer Vici in der Provinz Raetien. – V / In: A. Heising (ur. /ed.), Neue Forschungen zu zivilen Klein-siedlungen (vici) in den römischen Nordwest-Provinzen, 261–377, Bonn. GASPARI, A. 2007, The Ljubljanica River. Evidence for a Late Iron Age Ritual Site in the South-eastern Alps (Slovenia). – V / In: S. Groh, H. Sedlmayer (ur. / ed.), Blut und Wein, Keltisch-römische Kultpraktiken, Protohistoire Européenne 10, 141–154. GASPARI, A., M. MLINAR, 2005, Grave with a machaira from Most na Soči / Grob z mahairo z Mosta na Soči. – Arheološki vestnik 56, 169–186. GRABHERR, G., B. KAINRATH (ur. / eds.) 2010, Römische Raststationen und Straßeninfrastruktur im Ostalpenraum.– Ikarus 6. GROH, S. 2013, Die Infrastruktur der Bernsteinstraße in Pannonien. – V / In: S. Groh, H. Sedlmayer, C. Virág Zalka, Die Straßenstationen von Nemescsó und Sorokpolány an der Bernsteinstraße (Pannonien, Ungarn), Zentraleuropäische Archäologie 3, 173–184. GROH, S., H. SEDLMAYER 2006, Forschungen im Vicus Ost von Mautern-Favianis. – Der römische limes in Österreich 44. GUŠTIN, M., P. NOVAKOVIĆ, D. GROSMAN, B. MUŠIČ, M. LUBŠINA-TUŠEK 1996, Rimsko podeželje / Roman Countryside. – Ljubljana. HORVAT, J. 1999, Roman Provincial Archaeology in Slovenia Following the Year 1965: Settlement and Small Finds / Rimska provincialna arheologija v Sloveniji po letu 1965: poselitvena slika in drobna materialna kultura. – Arheološki vestnik 50, 215–257. HORVAT, J. 2009, Selected aspects of romanisation in western and central Slovenia. – V / In: G. Cuscito (ur. / ed.), Aspetti e problemi della romanizzazione. Venetia, Histria e arco alpino orientale, Antichita Altoadriatiche 68, 355–381. HORVAT, J. 2010, First century BC Roman fortifications in the Eastern Alps. – V / In: P. Herz, P. Schmid, O. Stoll (ur. / eds.), Zwischen Region und Reich, Region in Umbruch 3, 135–159. HORVAT, J. 2019, Roman road network and secondary settlements in the hinterland of Caput Adriae. – V / In: C. S. Fioriello, F. Tassaux (ur. / eds.), I paesaggi costieri dell’Adriatico tra Antichita e Altomedioevo, Scripta Antiqua 119, AdriAtlas 2, 77–95. HORVAT, J., M. SAGADIN 2017, Emonsko podeželje / Emona’s countryside. – V / In: Emona MM. Nastanek mesta – ur­banizacija prostora / Urbanisation of space – beginning of a town, 201–223, Ljubljana. ISTENIČ, J. 1987, Rodik – grobišče Pod Jezerom (Rodik – das Gräberfeld Pod Jezerom). – Arheološki vestnik 38, 69–136. ISTENIČ, J. 1988, Kuhinjsko posuđe egejskih tipova na nalazištu Rodik – nekropola Pod Jezerom (Aegean type cooking pottery from the Rodik – Pod Jezerom cemetery). – Diadora 10, 99–110. KOS, P. 2012, The construction and abandonment of the Claustra Alpium Iuliarum defence system in light of the numismatic material / Gradnja in opustitev obrambnega sistema Claustra Alpium Iuliarum v luči numizmatičnega gradiva. – Arheološki vestnik 63, 265–300. KOS, P. 2014, Izgradnja zapornega sistema Claustra Alpium Iuliarum / Construction of the Claustra Alpium Iuliarum fortifications. – V / In: J. Kusetič, P. Kos, A. Breznik, M. Stokin, Claustra Alpium Iuliarum, 112–132, Ljubljana. KOS, P. 2015, Ad Pirum (Hrušica) in Claustra Alpium Iulia-rum. – Vestnik 26. LAHARNAR, B., M. MLINAR 2015, Alto bacino dell'Isonzo. – V / In: F. Oriolo, G. Righi, A. Ruta Serafini, S. Vitri (ur. / eds.), Celti sui monti di smeraldo, 62–65, Trieste. MAGGI, P., B. ŽBONA TRKMAN 2007, Tra Natisone e Isonzo: il territorio in eta romana. – V / In: M. Chiaba, P. Maggi, C. Magrini (ur. / eds.), Le valli del Natisone e dell'Isonzo tra Centroeuropa e Adriatico, Studi e ricerche sulla Gallia Cisalpina 20, 59–77. MLINAR, M. 2005, Most na Soči (S. Lucia di Tolmino): a century after Carlo Marchesetti. – V / In: G. Bandelli, E. Montagnari Kokelj (ur. / eds.), Carlo Marchesetti e i cas­tellieri 1903-2003, Fonti e studi per la storia della Venezia Giulia 9, 325–338. MLINAR, M., D. SVOLJŠAK, V. VIDRIH PERKO, B. ŽBONA TRKMAN 2012, Kopalnico ima: arheološke raziskave in prezentacija rimske hiše z Mosta na Soči. – V / In: I. Lazar, B. Županek (ur. / eds.), Emona med Akvilejo in Panonijo, 257–271, Koper. MUŠIČ, B. 1999, Geophysical prospecting in Slovenia: an overview with some observations related to the natural environment / Geofizikalna prospekcija v Slovenij: pregled raziskav z nekaterimi ugotovitvami glede naravnega okolja. – Arheološki vestnik 50, 349–405. PAHIČ, S. 1977, Seznam rimskih kamnov v Podravju in Pomurju (Verzeichnis der Römersteine im slowenischen Drauland). – Arheološki vestnik 28, 13–73. PAHIČ, S. 1996, Uničujoča Drava (Die zerstörende Drava). – Ptujski zbornik 6/1, 131–155. PRÖTTEL, PH. M. 1996, Mediterrane Feinkeramikimporte des 2. bis 7. Jahrhunderts n. Chr. im oberen Adriaraum und in Slowenien. – Kölner Studien zur Archäologie der römischen Provinzen 2. SAGADIN, M. 1995, Poselitvena slika rimskega podeželja na Gorenjskem . – V / In: Kranjski zbornik 1995, 13–22, Kranj. SLAPŠAK, B. 1997, Starejša zgodovina Rodika (The Early History of Rodik). – V / In: M. Pregelj (ur. / ed.) Rodik med Brkini in Krasom, 19–64, Koper. SLAPŠAK, B. 2003, O koncu prazgodovinskih skupnosti na Krasu / The end of prehistoric communities in the Karst region. – Arheološki vestnik 54, 243–257. SVOLJŠAK, D., B. ŽBONA TRKMAN 1985, Načini pokopa v prazgodovini Posočja (Formes d‘enterrement dan la préhi­stoire de la région de Posočje). – Sahranjivanje pokojnika sa aspekta ekonomskih i društvenih kretanja u praistoriji i antici, Materijali 20, 87–89. SVOLJŠAK, D., B. ŽBONA-TRKMAN 1986, Novi napisi v Posočju (Neue Inschriften im Sočagebiet). – Arheološki vestnik 37, 385–397. ŠAŠEL, J. 1975, Topografsko-orientacijski leksikon krajevnih imen, ohranjenih v navedenih antičnih virih. – V / In: Arheološka najdišča Slovenije, 88–96, Ljubljana. ŠAŠEL, J., 1985, Zur Frühgeschichte der XV. Legion und zur Nordostgrenze der Cisalpina zur Zeit Caesars. – V / In: Römische Geschichte, Altertumskunde und Epigraphik, Festschrift für Artur Betz, Archäologisch-epigraphischeStudien 1, 547–555 (= J. Šašel, Opera Selecta, Situla 30, 1992, 469–477). ŠTULAR, B., E. LOZIĆ 2016, Primernost podatkov projekta Lasersko skeniranje Slovenije za arheološko interpretacijo: metoda in študijski primer (The Suitability of Laser Scan­ning of Slovenia Data for Archaeological Interpretation: Method and a Case Study). – V / In: R. Ciglič, M. Geršič, D. Perko, M. Zorn (ur. / eds.), Digitalni podatki, GIS v Sloveniji 13, 157–166. TALAA, D., I. HERRMANN 2003, Eine römische Straßensta­tion in Biedermannsdorf bei Wien – Vorbericht. – Fundort Wien 6, 212–224. TALAA, D., I. HERRMANN 2008, Biedermannsdorf. – Fund-berichte aus Österreich 47, 564–565. VEDALDI IASBEZ, V. 1994, La Venetia orientale e l‘Histria. – Roma. VIDRIH PERKO, V. 1997, Rimskodobna keramika z Ajdov-ščine pri Rodiku (Roman Pottery from Ajdovščina near Rodik). – Arheološki vestnik 48, 341–358. Jana Horvat Znanstvenoraziskovalni center SAZU Inštitut za arheologijo Novi trg 2 1000 Ljubljana jana.horvat@zrc-sazu.si Opera Instituti Archaeologici Sloveniae 1. Janez Dular, Slavko Ciglenečki, Anja Dular, Kučar. Železnodobno naselje in zgodnjekrščanski stavbni kompleks na Kučarju pri Podzemlju / Eisenzeitliche Siedlung und frühchristlicher Gebäudekomplex auf dem Kučar bei Podzemelj, 1995. (EUR 14.60) 2. Ivan Turk (ed.), Moustérienska »koščena piščal« in druge najdbe iz Divjih bab I v Sloveniji / Mousterian »bone flute« and other finds from Divje Babe I cave site in Slovenia, 1996. (EUR 14.60) 3. Jana Horvat (with contributions by Vesna Svetličič, Meta Bole, Metka Culiberg, Draško Josipović, Marko Stokin, Nina Zupančič), Sermin. Prazgodovinska in zgodnjerimska naselbina v severozahodni Istri / A Prehistoric and Early Roman Settlement in Northwestern Istria, 1997. (EUR 14.60) 4. Slavko Ciglenečki (with contributions by Zvezdana Modrijan, Andreja Dolenc Vičič, Ivan Turk), Tinje nad Loko pri Žusmu. Poznoantična in zgodnjesrednjeveška naselbina / Tinje oberhalb von Loka pri Žusmu. Spätantike und frühmit­telalterliche Siedlung, 2000. (EUR 14.60) 5. Janez Dular, Irena Šavel, Sneža Tecco Hvala, Bronastodobno naselje Oloris pri Dolnjem Lakošu / Bronzezeitliche Siedlung Oloris bei Dolnji Lakoš, 2002. (EUR 14.60) 6. Janez Dular, Halštatske nekropole Dolenjske / Die hallstattzeitlichen Nekropolen in Dolenjsko, 2003. (EUR 20.70) 7. Irena Lazar, Rimsko steklo Slovenije / The Roman glass of Slovenia, 2003. (EUR 27.40) 8. Anton Velušček (ed.), Hočevarica. Eneolitsko kolišče na Ljubljanskem barju / An eneolithic pile dwelling in the Lju­bljansko barje, 2004. (EUR 52.40) 9. Ivan Turk (ed.), Viktorjev spodmol in / and Mala Triglavca. Prispevki k poznavanju mezolitskega obdobja v Sloveniji / Contributions to understanding the Mesolithic period in Slovenia, 2004. (EUR 42.40) 10. Anton Velušček (ed.), Resnikov prekop. Najstarejša koliščarska naselbina na ljubljanskem barju / The oldest pile-dwel­ling settlement in the Ljubljansko barje, 2005. (EUR 40.00) 11. Andrej Gaspari (ed.), Zalog pri Verdu. Tabor kamenodobnih lovcev na zahodnem robu Ljubljanskega barja / Zalog near Verd. Stone Age hunters’ camp at the western edge of the Ljubljansko barje, 2006. (EUR 43.00) 12. Janez Dular, Sneža Tecco Hvala, South-Eastern Slovenia in the Early Iron Age. Settlement – Economy – Society / Jugovzhodna Slovenija v starejši železni dobi. Poselitev – gospodarstvo – družba, 2007. (EUR 58.00) 13. Ivan Turk (ed.), Divje babe I. Paleolitsko najdišče mlajšega pleistocena v Sloveniji. I. del: Geologija in paleontologija / Divje babe I. Upper Pleistocene Palaeolithic site in Slovenia. Part I: Geology and Palaeontology, 2007. (EUR 82.00) 14. Andrej Pleterski (with Timotej Knific, Borut Toškan, Janez Dirjec, Benjamin Štular and Mateja Belak), Zgodnjesrednje­veška naselbina na blejski Pristavi. Najdbe / Frühmittelalterliche Siedlung Pristava in Bled. Funde, 2008. (EUR 51.00) 15. Benjamin Štular, Mali grad. Visokosrednjeveški grad v Kamniku / Mali grad. High Medieval Castle in Kamnik, 2008. (EUR 51.00) 16. Anton Velušček (ed.), Koliščarska naselbina Stare gmajne in njen čas. Ljubljansko barje v 2. polovici 4. tisočletja pr. Kr. / Stare gmajne pile-dwelling settlement and its era. The Ljubljansko barje in the 2nd half of the 4th millennium BC, 2009. (EUR 56.00) 17. Jana Horvat, Alma Bavdek, Okra. Vrata med Sredozemljem in Srednjo Evropo / Ocra. The gateway between the Me­diterranean and Central Europe, 2009. (EUR 51.00) 18. Janez Dular, Marjana Tomanič Jevremov, Ormož. Utrjeno naselje iz pozne bronaste in starejše železne dobe / Ormož. Befestigte Siedlung aus der späten Bronze- und der älteren Eisenzeit, 2010. (EUR 57.00) 19. Andrej Pleterski (with contributions by Igorj Bahor, Vid Pleterski, Marko Žagar and Veronika Pflaum), Zgodnjesrednje­veška naselbina na blejski Pristavi. Tafonomija, predmeti in čas. / Frühmittelalterliche Siedlung Pristava in Bled. Taphonomie, Fundgegenstände und zeitliche Einordnung, 2010. (EUR 39.00) 20. Jana Horvat and Andreja Dolenc Vičič (with the contribution of Marjana Tomanič Jevremov and Marija Lubšina Tušek), Arheološka najdišča Ptuja. Rabelčja vas / Archaeological Sites of Ptuj. Rabelčja vas, 2010. (EUR 45.00) 21. Borut Toškan (ed.), Drobci ledenodobnega okolja. Zbornik ob življenjskem jubileju Ivana Turka / Fragments of Ice Age environments. Proceedings in Honour of Ivan Turk's Jubilee, 2011. (EUR 45.00) 22. Anton Velušček (ed.), Spaha, 2011. (EUR 47.00) 23. Slavko Ciglenečki, Zvezdana Modrijan, Tina Milavec (with contributions of Benjamin Štular, Saša Čaval and Ivan Šprajc), Poznoantična utrjena naselbina Tonovcov grad pri Kobaridu. Naselbinski ostanki in interpretacija / Late Antique fortified settlement Tonovcov grad near Kobarid. Settlement remains and interpretation, 2011. (EUR 55.00) 24. Zvezdana Modrijan and Tina Milavec (with contributions of Peter Kos, Dragan Božič, Matija Turk, Petra Leben Seljak, Borut Toškan, Janez Dirjec, Francesco Boschin and K. Patrick Fazioli), Poznoantična utrjena naselbina Tonovcov grad pri Kobaridu. Najdbe / Late Antique fortified settlement Tonovcov grad near Kobarid. Finds, 2011. (EUR 55.00) 25. Maja Andrič (ed.), Dolgoročne spremembe okolja 1, 2012. (EUR 30.00) 26. Sneža Tecco Hvala, Magdalenska gora. Družbena struktura in grobni rituali železnodobne skupnosti / Social structure and burial rites of the Iron Age community, 2012. (EUR 57.00) 27. Janez Dular, Severovzhodna Slovenija v pozni bronsati dobi / Nordostslowenien in der späten Bronzezeit, 2013. (EUR 51.00) 28. Andrej Pleterski, The Invisible Slavs, 2013. (EUR 35.00) 29. Ivan Turk, Divje babe I. Paleolitsko najdišče mlajšega pleistocena v Sloveniji. II. del: Arheologija / Divje babe I. Upper Plei­stocene Palaeolithic site in Slovenia. Part II: Archaeology, 2014. (EUR 62.00) 30. Sneža Tecco Hvala (ed.), Studia Praehistorica in Honorem Janez Dular, 2014. (EUR 56.00) 31. Jana Horvat (ed.), The Roman army between the Alps and the Adriatic, 2016. (EUR 42.00) 32. Lucija Grahek, Stična. Železnodobna naselbinska keramika / Stična. Iron Age Settlement Pottery, 2016. (61.00 EUR) 33. Drago Svoljšak, Janez Dular, Železnodobno naselje Most na Soči. Gradbeni izvidi in najdbe / The Iron Age Settlement at Most Na Soči. Settlement Structures and Small Finds, 2016. (EUR 73.00) 34. Janez Dular, Sneža tecco Hvala (eds.), Železnodobno naselje Most na Soči. Razprave / The Iron Age Settlement at Most na Soči. Treatises, 2018 (73.00 EUR) 35. Andrej Pleterski, Benjamin Štular, Mateja Belak (eds.), Izkopavanja srednjeveškega in zgodnjenovoveškega grobišča pri Župnicerkvi v Kranju v letih 1964-1970 / Excavations of Medieval and Post-Medieval cemetery at the Župna cerkev in Kranj from 1964 to 1970, Grobišče Župna cerkev v Kranju 1 / Cemetery Župna cerkev in Kranj 1, 2016. (EUR 34.00) 36. Sneža Tecco Hvala, Molnik pri Ljubljani v železni dobi / The Iron Age site Molnik near Ljubljana, 2017. (EUR 43.00) 37. Andrej Pleterski, Benjamin Štular, Mateja Belak (eds.), Arheološka raziskovanja grobov najdišča Župna cerkev v Kranju med letoma 1972 in 2010 / Archaeological investigation of graves from Župna cerkev site in Kranj between 1972 and 2010, Grobišče Župna cerkev v Kranju 2 / Cemetery Župna cerkev in Kranj 2, 2017. (EUR 39.00) 38. Andrej Pleterski, Benjamin Štular, Mateja Belak (eds.), Začetek in konec raziskovanj grobišča pri Župni cerkvi v Kranju (1953 in 2011-2013) / The beginnning and end of exploration at the Župna cerkev cemetery in Kranj (1953 and 2011-2013) Grobišče Župna cerkev v Kranju 3 / Cemetery Župna cerkev in Kranj 3, 2019. (EUR 39.00) 39. Slavko Ciglenečki, Zvezdana Modrijan, Tina Milavec (with contributions of Peter Kos, Mateja Kovač, Borut Toškan, Lucija Grahek, Darja Grosman, Julijana Visočnik), Korinjski hrib in poznoantične vojaške utrdbe v Iliriku / Korinjski hrib and late antique military forts in Illyricum, 2020. (EUR 49.00) Sneža Tecco Hvala (with contributions by Brina Škvor Jernejčič, Borut Toškan, Tjaša Tolar, Karina Grömer, Klara Kostajnšek, Tatjana Tomazo-Ravnik and Adrijan Košir) Molnik pri Ljubljani v starejši železni dobi The Iron Age site at Molnik near Ljubljana Molnik is a hill in the hinterland of Ljubljana in central Slovenia that has been known as an ar­chaeological site for quite some time. Systematic investigations ensued in the 1980s and 1990s, led by Ivan Puš, then curator in the Mestni muzej Ljubljana. They revealed ramparts and habitation terraces on the summit of the hill, as well as flat and tumulus graves on five locations around the summit that date to the Early Iron Age and some finds from later periods. The monograph presents the results of the field investigations and of numerous analyses of the site, which was in the Early Iron Age a frontier post of the Dolenjska Hallstatt community. The content is illustrated with numerous maps, plans, tables and photos, while the body of the small finds is presented in drawn form on the plates. The cultural and historical assessment of the finds from Molnik improves our knowledge of the time between the 8th and the 4th centuries on the territory of the present-day Slovenia. This knowledge is further advanced by the results of the anthropological, archaeozoological, archaeobotanical and radiocarbon analyses, as well as the spectrographic elemental microanalysis of select items presented in the third part of the book. 2017, (Opera Instituti Archaeologici Sloveniae, 36), 272 pages, 116 drawings, plans, photographs, tables, 44 plates, 20 × 29 cm, hardcover, ISBN 978-961-05-0020-9. Price: EUR 43.00 Andrej Pleterski, Benjamin Štular, Mateja Belak Arheološka raziskovanja grobov najdišča Župna cerkev v Kranju med letoma 1972 in 2010Archaeological investigation of graves from Župna cerkev site in Kranj between 1972 and 2010 Sixty-year excavation (1953 to 2013) of Župna Cerkev Cemetery in Kranj unearthed almost 3000 graves. The monograph introduces 1048 graves found between 1972 and 2010. The excavations were initially led by Andrej Valič, the curator of the Gorenjski muzej (1972 and 73, 862 graves). Milan Sagadin from ZVKD Kranj continued the dig from 1984 to 2001 (163 graves), followed by Draško Josipovič (Megalan Skupina, d. o. o., 2003 to 2010), who excavated 23 more graves. An extensive chapter presents the artefacts of Župna Cerkev Cemetery with no data about their location. The varied documentation, which has changed over the years, is the basis for the publication of all available data on the graves and artefacts kept by the Gorenjski muzej in Kranj. The introduction, a study on the nature of the data, comments on possible inconsistencies between the various types of documentation. Primarily it deals with the question of reliability of individual grave good assemblages, which is the basis for all further analysis. The discussed group of graves originates mainly from the northern exterior of the church and from the interior of the present­-day church, that is, from the place where graves intertwined with the remains of older church and non-church buildings. Their building development and changing usability will be impossible to explain without good knowledge of the cemetery. 2017, (Opera Instituti Archaeologici Sloveniae, 37), 424 pages, 1048 graves, more than 1650 artefacts, hundreds of drawings and photographs, 24 plans, 20 x 29 cm, hardcover, ISBN 978-961-05-0030-8. Price: EUR 39.00 Andrej Pleterski, Benjamin Štular, Mateja Belak, Helena Bešter Začetek in konec raziskovanj grobišča pri Župni cerkvi v Kranju (1953 in 2011-2013)The beginnning and end of exploration at the Župna cerkev cemetery in Kranj (1953 and 2011-2013) The third book on the cemetery Župna cerkev in Kranj completes the entire corpus of data on objects, graves and the cemetery as a whole. During the sixty years of field research (1953-2013) all documentation techniques that prevailed in individual periods were used, which is an ideal test of their comparative advantages and disadvantages. The route from excavations to critical publication was extremely compound and could only be completed in a very complex digital environment with the help of a number of unique information solutions. Simultaneous publishing in analogue and digital form makes it possible to use the best of both media. The three-part structure of the book introduces an overview of field documentation and its structure, and solves the problems that arise. The most extensive is the central part, which uniquely represents archaeological remains. The third part is cartographic and shows spatial distribution of individual graves. The cemetery belonging to the medieval city is the largest published medieval cemetery in Europe. More than 1000 years (from the 8th century to the end of the 18th century) of continuous burial in the same place caused the accumulation of graves, resulting in over 10,000 stratigraphic relations. Complete information is now available for further typological, chronological, population and other studies. Sky (imagination) is the limit. 2019, (Opera Instituti Archaeologici Sloveniae, 38), 408 pages, 1865 graves, more than 1300 artefacts, hundreds of drawings and photographs, 36 plans, 20 x 29 cm, hardcover, ISBN 978-961-05-0150-3. Price: EUR 39.00 Slavko Ciglenečki, Zvezdana Modrijan, Tina Milavec (with contributions by Peter Kos, Mateja Kovač, Borut Toškan, Lucija Grahek, Darja Grosman, Julijana Visočnik Korinjski hrib in poznoantične vojaške utrdbe v Iliriku Korinjski hrib and late antique military forts in Illyricum The monograph presents the results of two-year excavations (1982 and 1983) on the late-antique fortified hilltop settlement Korinjski hrib above Veliki Korini in Suha Krajina, where the remains of five defence towers and an early Christian church were explored. The settlement was initially recognized as a military post and as such represented an exception in the Eastern Alpine area. Such a definition raised some doubts and dilemmas. These dilemmas are presented - in addition to geographical outline and research history - in the Introduction. The following set presents - in text and with abundant graphic material – field reports of excavations of towers, the church, and a small cemetery. All groups of finds (non-pottery, pottery and coins) and anthropological and archaeological studies are also presented. An extensive chapter is devoted to the interpretation of architectural remains (towers, church, the fort as a whole).The conclusion also discusses the prehistoric settlement, and the results of structural survey of the ruins of church of St George just below the fort. 2020, (Opera Instituti Archaeologici Sloveniae, 39), 400 pages, 229 plans, drawings and photos, 13 graphs, 39 plates, appendix, 20 × 29 cm, hardcover, ISBN 978-961-05-0254-8. Price: EUR 49.00