103
Documenta Praehistorica XLIII (2016)
Becoming sedentary|
The seasonality of food resource exploitation in the
Mesolithic-Neolithic Danube Gorges
Vesna Dimitrijević 1, Ivana ?ivaljević 2 and Sofija Stefanović 2
1 Laboratory for Bioarchaeology, Department of Archaeology, Faculty of Philosophy, University of Belgrade, RS
vdimitri@f.bg.ac.rs
2 BioSense Institute, University of Novi Sad, RS
sofija.stefanovic@biosense.rs
Introduction
With their long temporal sequence, spanning be-
tween c. 9500 to 5500 cal BC (corresponding to the
regional Mesolithic, Mesolithic-Neolithic Transforma-
tion phase and the Early/Middle Neolithic) (Bori≤,
Dimitrijevi≤ 2009; Bori≤ 2011), open-air sites in the
Danube Gorges (North-Central Balkans) provide great
opportunities for studying and understanding vari-
ous aspects of Mesolithic-Neolithic transition. In par-
ticular, intensified building activities in the Late Me-
solithic and Transformational Mesolithic-Neolithic
phases (cf. Bori≤, Dimitrijevi≤ 2009; Bori≤ 2011)
have been interpreted as archaeological indicators of
increased sedentism resulting from developing social
complexity of local hunter-gatherer-fisher communi-
ties (cf. Voytek, Tringham 1989; Chapman 1993;
Radovanovi≤ 1996; Radovanovi≤, Voytek 1997) and
from merging local traditions with new economic
and social practices accompanying ‘Neolithic’ life-
ways (cf. Bori≤ 2007; 2011; Jovanovi≤ 2008). Ob-
viously, there is no universal mode of hunter-fisher-
gatherer existence, as opposed to a presumed seden-
tary lifestyle of food-producing communities, nor
can a clear-cut change from mobility to sedentism be
assumed. Moreover, degrees of sedentism are often
difficult to assess and divide into neat categories due
to the complexity of human lifestyles and residen-
tial practices (cf. Whittle 2001). One of the starting
points, however, most commonly involves determin-
ing whether the sites in question were occupied sea-
sonally or year-round. This feature can be addressed
by looking at subsistence patterns, i.e. the seasons
of exploitation of most important animal species, as
has been demonstrated by a number of studies,
which range from simple observations of the pres-
ABSTRACT – In this paper, we investigate whether the Mesolithic-Neolithic sites in the Danube Gorges
were occupied seasonally or all year round by looking at animal skeletal remains. The hunting sea-
sons of most important game animals have been determined on the basis of antler and teeth
growth, supplemented by looking into the presence of migratory fish and birds. The patterns of food
resource exploitation seem to indicate year-round occupation of the settlements, and suggest that a
significant degree of sedentism existed in the Danube Gorges prior to, and independently of, the
adoption of animal and plant husbandry.
IZVLE∞EK – V ≠lanku na podlagi analize ∫ivalskih ostankov raziskujemo, ali so bila mezolitsko-neo-
litska najdi∏≠a na obmo≠ju soteske D∫erdap poseljena sezonsko ali ≠ez celo leto. Na podlagi rasti ro-
govja in zob smo dolo≠ili lovno sezono za najpomembnej∏o divjad, podatke pa smo dopolnili z anali-
zo selitve rib in ptic. Vzorci rabe raznih virov prehrane ka∫ejo na celoletno poselitev na naselbinah;
sklepamo tudi, da se je na obmo≠ju D∫erdapa delno sedentaren na≠in ∫ivljenja pojavil lo≠eno od in
pred pojavom prvih doma≠ih ∫ivali in kultiviranih rastlin.
KEY WORDS – seasonality; archaeozoology; sedentism; Danube Gorges; Mesolithic-Neolithic
DOI> 10.4312\dp.43.4 
Vesna Dimitrijević, Ivana ?ivaljević and Sofija Stefanović
104
ence/absence of seasonally available fauna, to more
sophisticated analytical procedures, including the
determination of the individual ages and the season
of death of exploited animals, skeletochronology, or
growth-increment studies (cf. Pike-Tay et al. 2004),
or stable isotope balances in teeth and bones (Blaise,
Balasse 2011).
In this study, we present estimated seasons of cap-
ture on the basis of the results of ageing of faunal
skeletal elements (teeth and antler) of economical-
ly important wild game species, supplemented by
quantifications of fish and bird remains (some of
them available seasonally and others more or less
all year round) from sites at Vlasac, Lepenski Vir,
and Padina. By no means exhaustive, this study aims
to offer additional data relevant to understanding
subsistence and settlement patterns, and to shed
more light on the relationship between hunting, fish-
ing and increased sedentism in the Danube Gorges.
Archaeological background
The Padina, Lepenski Vir, and Vlasac sites are situ-
ated in the Upper Gorge (Gospo∂in Vir, or Lady’s
Whirlpool) of the Danube Gorges (Fig. 1), on river-
ine terraces sloping towards the Danube. The initial
rescue excavations of the sites were undertaken in
the 1960s and 1970s (Srejovi≤ 1969; 1972; Srejo-
vi≤, Letica 1978; Radovanovi≤ 1996; Jovanovi≤
2008), prior to the raising of the water level due to
dam construction, and new revisory excavations of
the preserved portion of the Vlasac terrace were car-
ried out in 2006–2009 (Bori≤ 2006; Bori≤ et al.
2008; 2014). The earliest traces of human presence
in these locations date to the period of c. 9500–
7400 cal BC (the Early Mesolithic) (Bori≤, Dimitrije-
vi≤ 2009; Bori≤ 2011), and are manifested by occa-
sional burials (some in a seated position), several
stone constructions (Bori≤, Miracle 2004; Jovanovi≤
2008), as well as caches of animal bones originating
predominantly from wild game and fish (Bori≤, Di-
mitrijevi≤ 2006), but also from canids displaying
mixed wolf and dog features, suggesting that the ini-
tial stages of dog domestication took place during
this time (Dimitrijevi≤, Vukovi≤ 2015).
The few and fragmentary traces of Early Mesolithic
occupation are mostly due to their disturbance and
devastation by later (Late Mesolithic and Transfor-
mation phase) building and occupational activities
(Bori≤ 2011). The Late Mesolithic phase (c. 7400–
6300/6200 cal BC) is particularly well documented
at Vlasac (Bori≤ 2011), where numerous burials, se-
veral dug-out dwelling structures, stone-lined rectan-
gular hearths and a large number of chipped stone,
bone and antler artefacts have been discovered (Sre-
jovi≤, Letica 1978; Nemeskéri 1978; Kozłowski,
Kozłowski 1982; Roksandi≤ 1999; 2000; Bori≤ et
al. 2008; 2014), as well as an enormous assemblage
of animal bones (Bökönyi 1978). This phase also
produced evidence of large-scale exchange networks,
witnessed by occasional finds of good-quality flint of
non-local origin, and the occurrence of marine gas-
tropods Cyclope neritea and Columbella rustica in
some of the burials (Cristiani, Bori≤ 2012; Bori≤ et
al. 2014; Srejovi≤, Letica 1978).
The Transformation/Early Neolithic phase (c. 6300/
6200–5900 cal BC) (Bori≤, Dimitrijevi≤ 2007; Bori≤
2011), coinciding with the emergence of the first
farming communities in the Balkans, saw the estab-
lishment of the first settlements with elaborate tra-
pezoidal-base dwellings, distinctive sandstone sculpt-
ed boulders and the introduction of pottery at Le-
penski Vir and Padina (Srejovi≤ 1969; 1972; Rado-
vanovi≤ 1996; Gara∏anin, Radovanovi≤ 2001; Jo-
vanovi≤ 2008), but with unchanged subsistence stra-
tegies dependent on hunting and fishing (Bori≤, Di-
mitrijevi≤ 2006). The first domesticates (apart from
Fig. 1. The Danube Gor-
ges area with relevant
sites mentioned in the
text. The position of the
Danube Gorges within
a wider area – the Bal-
kan peninsula shown
in the lower left corner.
Images from Google
Earth.
Becoming sedentary| The seasonality of food resource exploitation in the Mesolithic-Neolithic Danube Gorges
105
dogs) make an appearance only after c. 6000/5900
cal BC (Bori≤, Dimitrijevi≤ 2007), after most of the
dwellings had been abandoned and new elements of
material culture (domed ovens, ground stone tools,
bone spoons), along with certain individuals of non-
local origin appear in the area (Bori≤ 2011; Bori≤,
Price 2013). Large-scale animal husbandry, however,
was never fully introduced into the Upper Gorge;
the sites continued to be in use until c. 5500 cal BC,
but probably for specialised activities, most notably
fishing (Bonsall 2008; Bori≤ 2011).
The faunal assemblages
The assessment of the nature of animal exploitation
and quantification of faunal remains from the Danube
Gorges are hindered by several factors, namely the
inadequate recovery techniques and selective collec-
tion imposed by time constraints during the course
of the excavations, and the fragmentary state of fau-
nal assemblages from some of the sites (Bori≤ 2001;
2002; Bori≤, Dimitrijevi≤ 2006; Dimitrijevi≤ 2000;
2008). The faunal assemblages from the old excava-
tions of Lepenski Vir (1965–1970), Vlasac (1970–
1971), and Padina (1968–1970) were collected en-
tirely by hand, using collection grids consisting of
4x4m quadrants. Furthermore, the enormous assem-
blage from the original excavations of Vlasac, com-
prising nearly 30 000 identified specimens was dis-
carded after the initial analysis by Sándor Bökönyi
(1978), with only a small sample of bird (Tab. 5)
and fish remains preserved (Bori≤ 2002.App. 5; Di-
mitrijevi≤ et al. submitted). Part of the faunal assem-
blage from Lepenski Vir, mainly from first excavation
seasons (up to 1969, cf. Bori≤, Dimitrijevi≤ 2006)
was also discarded after the preliminary analysis by
Bökönyi (1969; 1970; 1972). The preserved part of
the assemblage originated from later campaigns
(1968–1970), and consisted of animal bones pri-
marily related to the floors of trapezoidal-base dwel-
lings, and areas between and underneath them (Di-
mitrijevi≤ 2000; 2008). Only the faunal assemblage
from the new (2006–2009) excavations of Vlasac, as
well as the one collected in 1992–1996 at the down-
stream site of Schela Cladovei (Fig. 1) (Bartosiewicz
et al. 1995; 2001; 2008), were collected by water
sieving and flotation, in addition to hand collection
(Bori≤ et al. 2014; Dimitrijevi≤ et al. submitted).
The environment of Padina, Lepenski Vir and Vla-
sac, which were situated on riverine terraces in a
narrow belt between the river and the forest, pro-
vided both good hunting and fishing conditions. The
dense forests surrounding the sites were the natur-
al habitat of various wild game species which were
hunted for meat, as well fur-bearing animals which
were probably hunted for fur or as vermin. The vi-
cinity of the river and its large whirlpools enabled
fishing for both freshwater species (available most
of the year) and seasonally available anadromous
species, as well as the exploitation of waterfowl.
In terms of wild game at all three sites, the most nu-
merous remains are of red deer (Cervus elaphus)
which was most commonly hunted, followed by wild
boar (Sus scrofa), roe deer (Capreolus capreolus),
and aurochs (Bos primigenius) (cf. Bökönyi 1969;
1970; 1972; 1978; Clason 1980; Dimitrijevi≤ 2008).
In terms of fishes, the remains of various cyprinids
(Cyprinidae), Wels catfish (Silurus glanis), migrato-
ry sturgeons (Acipenseridae), and huchen (Hucho
hucho) have been discovered at all three sites, but
in varying quantities (Ωivaljevi≤ in prep.). As expect-
ed, the hand-collected, water-sieved and flotated
sample from new excavations of Vlasac contained
many more remains of smaller cyprinids and exhib-
ited the greatest species diversity in comparison to
hand-collected assemblages from Lepenski Vir and
Padina (Tab. 4). However, in spite of the same col-
lection methods employed at the latter two sites,
their faunal assemblages also exhibited significant
differences, namely in the absolute predominance of
catfish observed at Padina. The Lepenski Vir assem-
blage contained a significant number of cyprinid, cat-
fish and huchen remains, but also a much greater
number of sturgeon remains in comparison to Pa-
dina and Vlasac (Tab. 4).
The role of hunting vs. fishing in subsistence strate-
gies is difficult to assess on the basis of a straight-
forward comparison of quantified mammal and fish
remains, given the differences in their anatomy and
size, and biases affecting their survival and recovery
(cf. Wheeler, Jones 1989; Takács, Bartosiewicz
1998; Bartosiewicz, Bonsall 2004; Bori≤ 2001). The
analyses of carbon (δ13C) and nitrogen (δ15N) iso-
tope ratios in human bone collagen have suggested
that the Mesolithic inhabitants of the Danube Gorges
consumed considerable amounts of fish, with a gra-
dual broadening of the dietary spectrum to include
more terrestrial resources at the onset of the Neoli-
thic (Bonsall et al. 1997; 2004; Grupe et al. 2003;
Bori≤ et al. 2004), whereas the analysis of the sul-
phur (δ34S) isotope ratio has suggested that there
have been significant inter- and intra-site variabili-
ties in dietary practices (Nehlich et al. 2010). On
the basis of faunal data, it can be hypothesised that
both hunting and fishing made significant contribu-
Vesna Dimitrijević, Ivana ?ivaljević and Sofija Stefanović
106
tions to the diet; the former mainly oriented to-
wards red deer and the latter exhibiting more intra-
site (and possibly diachronic) variety.
Apart from a general emphasis on the importance of
annual cycles in shaping human lives in the Danube
Gorges (cf. Srejovi≤ 1969; 1972), the subject of sea-
sonality was first addressed by Anneke T. Clason
(1980) in her study of Star≠evo and Padina animal
remains. In the description of mammal remains, she
specified some specimens as seasonal markers, made
some remarks on the seasonal availability of bird
and fish, and hypothesised that “Padina was perma-
nently inhabited during all seasons of the year”,
and furthermore, that this was probably also true
for Vlasac and Lepenski Vir. A discussion on the in-
fluences of seasonal cycles on life in the Mesolithic
Danube Gorges is given by Bori≤ in his thesis (Bo-
ri≤ 2002). Here, and in several later references, sea-
sonal estimates determined on the basis of particu-
lar specimens of mammal remains were also noted
(Dimitrijevi≤ 2000; 2008; Bori≤, Dimitrijevi≤ 2006;
2009).
Materials and methods
The seasonal schedule of food resource acquisition
in the Danube Gorges has been determined on the
basis of skeletal elements which are sensitive to the
question of seasonality. This included determining
the kill season of red deer and roe deer on the basis
of antler growth (cf. Schmid 1972), and ageing of
red deer, roe deer, fox and wild boar on the basis
of tooth eruption and wear (cf. Habermehl 1985;
Brown, Chapman 1991; Carter 2006; Carter, Mag-
nell 2007) (Tabs. 1–3). Although radiographs were
not made, the fragmentary state of some of the man-
dibles and loose teeth enabled ageing on the basis
of root formation stages (after Brown, Chapman
1991; Carter 2006; Carter, Magnell 2007). Other
mammal species were not taken into account, either
because they were not represented by juvenile ani-
mals, are not seasonal breeders, or they give birth
several times a year. The breeding and reproduction
patterns of dogs in particular could have changed
significantly as a result of domestication, and could
have extended throughout several seasons.
In this respect, the most reliable season indicators
are milk and erupting permanent teeth of juvenile
animals, given that they follow a fairly standardised
schedule, whereas tooth wear does not occur at an
even pace (Bartosiewicz 1989; Pike-Tay et al. 2004).
Less reliable season indicators are shed antlers; even
if they are dropped during limited periods of the
year (February-March in the case of red deer and Oc-
tober-November for roe deer, cf. Schmid 1972), they
could have been collected at any time and/or curat-
ed, especially since they were extremely valued as
raw material and often structurally deposited.
We assumed that red deer give birth at the begin-
ning of June (cf. Carter 2006); similarly, the birth-
ing season of recent red deer populations in the re-
gion is described as taking place “at the end of May
– beginning of June” (Grupa autora 1991). Roe
deer give birth in May (Grupa autora 1991; Haber-
mehl 1985; Carter 2006). The farrowing season of
wild swine is seasonally less restricted, occuring in
March-April in recent wild pig populations in the
Balkans (Grupa autora 1991), whereas in Sweden
and Germany (Carter, Magnell 2007), as well as in
Turkey (Bull, Payne 1982) it occurs between March
and May. Wild sows may breed twice a year, thus
substantially widening the farrowing season; how-
ever, 65–90% of births occur between March and
May, mostly March and April in Sweden and Ger-
many (Carter, Magnell 2007 and references there-
in). Similarly, in the Balkans, 85% of births take
place in March and April (Grupa autora 1991). Fox
cubs are born in March (Habermehl 1985).
The observations on the hunting seasons of most
important game animals were supplemented by
looking into the seasonal availability of migratory
fish and birds and quantifying their remains (Tabs.
4–5). As noted by László Bartosiewicz and Clive
Bonsall (2004) and Bartosiewicz et alii (2008), the
presence/absence criteria is highly probabilistic,
given that the presence of a certain species might
signal occupation during a particular season, but its
absence does not imply the lack of occupation at
other times. For example, prior to the Iron Gates’
dams construction, anadromous sturgeons – beluga
(Huso huso), Russian sturgeon (Acipenser guelden-
staedtii), ship sturgeon (Acipenser nudiventris), and
stellate sturgeon (Acipenser stellatus), undertook
their bi-annual spawning migrations from the Black
Sea at particular times of the year (in early spring
and autumn), but with inter-specific and individual
differences in the start and peak dates depending on
water temperature. Also, the populations migrating
in autumn were known to overwinter in the river
(Risti≤ 1977). It should be noted, however, that the
main sturgeon fishing seasons in the Danube Gorges
in more recent times corresponded with their migra-
tion patterns, with most catches occurring in spring,
followed by autumn (cf. Petrovi≤ 1998. [1941]; for
Becoming sedentary| The seasonality of food resource exploitation in the Mesolithic-Neolithic Danube Gorges
107
different data from Hungary, see Bartosiewicz et
al. 2008).
The investigation of the presence of migratory birds
as seasonal indicators is also based on the assump-
tion that their behaviour in the past roughly mir-
rored their present migration patterns. In addition,
it should be noted that bird remains at sites could
have been accumulated not only by humans and
dogs, but wild predators as well, having in mind that
no bird bones bore butchering marks. This hypoth-
esis does not seem likely for certain species – parti-
cularly birds of prey, which nested in the high cliffs
of the Gorges – but cannot be completely ruled out.
Therefore, the data we present here includes the sug-
gested seasons of kill of the most important game
animals, which could vary to some extent due to
slight or greater variations in the birthing seasons of
the species and their ontogeny, as well as the most
probable seasons of exploitation of migratory spe-
cies, assuming their past migration patterns broadly
correspond to those observable in more recent times.
Results
Vlasac
As opposed to the 640m2 area excavated in 1970–
1971 (Srejovi≤, Letica 1978), the preserved portion
of the Vlasac terrace encompassed by the 2006–
2009 excavations (326m2) covered the southern-
most peripheral area of the original settlement (Bo-
ri≤ et al. 2014), with much less evidence of domes-
tic activities or food preparation and consumption
(Dimitrijevi≤ et al. submitted). This feature is cer-
tainly manifested in the quantity and preservation
of animal remains from the two areas of the site, in
particular concerning mammal remains, which are
least likely to suffer from bias in recovery techni-
ques. The total number of mammal bone specimens
identified to the species/genus/family level from the
1970–1971 excavations equaled 11188 (Bökönyi
1978), as opposed to 801 specimen identified to the
species/genus/family level and 10955 unidentified
mammal remains from the 2006–2009 excavations
(Dimitrijevi≤ et al. submitted). In terms of the most
important game animals, the NISP (number of iden-
tified specimens) of red deer, wild boar and roe deer
in the 1970–1971 assemblage equaled 6732, 1185
and 510, respectively (Bökönyi 1978), as opposed
to their NISP in the 2006–2009 assemblage (209, 86
and 47, respectively) (Dimitrijevi≤ et al. submitted).
However, due to more sophisticated collection tech-
niques, the new excavations of Vlasac enabled the
recovery of even the most fragile skeletal elements,
including the deciduous teeth which were used in
this study (Tab. 1), as well as a plentitude of fish re-
mains (Tab. 4).
Most of the mammal remains used in this study origi-
nated from the 2006–2009 excavations of Vlasac,
and included a fragmented red deer skull, a fragment
of a roe deer skull with an antler, as well as decidu-
ous and permanent teeth of red deer, roe deer, wild
boar and fox. Given that the largest part of the fau-
nal assemblage from 1970–1971 was discarded, in
this study we were able to include only 9 unshed
and 31 shed antlers of red deer, and 3 unshed ant-
lers of roe deer, which were originally packed with
bone and antler artefacts and thus saved (Tab. 1).
The fragmented red deer skull VL 29/1 (Fig. 2) ori-
ginated from a young male, and on the basis of par-
tially preserved antlers still on pedicles, it was deter-
mined that it was captured between September and
February. The skull had been used (possibly, for a
significant amount of time), most likely in a series of
ritualistic activities, given that it exhibited a num-
ber of modifications (numerous cuts and damage to
the cranial roof and flattened occipital condyles)
which could not be explained by standard butcher-
ing practices (Dimitrijevi≤ et al. submitted), and
that it had been deposited at the top of a sequence
of several inhumations and cremations (Bori≤ 2010;
Bori≤ et al. 2014.Tab. 1: the skull is described as ori-
ginating from context 19). Unshed antlers collected
over the course of previous excavations indicate the
same broad (autumn-winter) time span, whereas, as
Fig. 2. Fragmented male red deer skull (VL 29/1)
from Vlasac, with fragmented antlers attached to
pedicles, captured between September and Febru-
ary.
already noted, the use of shed antlers (even though
shedding occures within a short time span) as sea-
sonal indicators should be taken with caution, as it
is possible they were collected later.
The stages of teeth eruption and wear indicate sev-
eral seasons of capture of red deer, some with more
precision than others. In particular, a 3rd and 4th
decidious premolar from a fragmented left maxilla
(VL 216/15) (Fig. 3) pointed to an age of 7–8 months,
indicative of a December-January capture; whereas a
left 3rd permanent incisor with an open root and
milk incisors (left and right 2nd and 3rd) and right
canine (VL 237/4) (Fig. 4) from a mandible that had
not been preserved pointed to an approximat eage
of 15–18 months or a kill that occurred between Au-
gust and November.
A fragment of a roe deer skull with an antler frag-
ment on a pedicle (VL 314 x. 23), as well as three
unshed antlers collected in previous excavations in-
dicated that the capture seasons occurred between
May and October (cf. Schmidt 1972). Assuming the
end of May as the season when roe deer give birth
(cf. Habermehl 1985), one decidious premolar (3/4)
(VL 51/1) indicated an age of one month, i.e. a June
capture.
As already noted, the farrowing season for wild
swine is generally from March to April (cf. Grupa
autora 1991; Carter, Magnell 2007); in our study,
the ageing and the estimation of season is given
counting from April (Tab. 1). Two teeth – an upper
3rd deciduous premolar (VL 237/5) and an upper 2nd
deciduous premolar (VL 282/3) – produced fairly re-
liable season estimates, an April-May and September-
October kill respectively.
Fox teeth (an upper deciduous canine and a germ of
the first molar, VL 222/22–23) were included in this
study, as they provided a fairly reliable kill season
(May), given that this stage of teeth eruption is in-
dicative of a three-month-old individual (cf. Haber-
mehl 1985).
Although most of the elements utilised in ageing
offered very broad season estimates spanning over
several months (Tab. 1), the aforementioned spec-
imens provided fairly accurate and short seasons of
death. They signal particular events which took place
in December-January, April-May, May, June and Sep-
tember-October, i.e. are indicative that hunting did
not take place during a limited period of the year.
Both the fragment of a roe deer skull with an antler
(VL 314 x. 23) and the red deer skull (VL 29/1) have
been directly dated, producing a date of 7131–6823
cal BC at 95% probability (OxA-21962) and 6006–
5838 cal BC at 95% probability (OxA-16544) respec-
tively (Bori≤ et al. 2014). At the same time, these
two dates are currently the oldest and the youngest
obtained from the new excavations of Vlasac, thus
dating the span of activities in the peripheral zone
of the settlement, with Transformation phase activ-
ity (post c. 6200 cal BC) mainly related to the inter-
ment of the latest burials in the successive burial
zone closed by the red deer skull (Bori≤ et al. 2008;
2014). A wild boar tooth (VL 282/3) was found in
a context (a concentration of artefacts beneath the
floor of a dug-out dwelling) dated with two absolute
dates: 7034–6692 cal BC at 95% probability (OxA-
24809) and 7035–6698 cal BC at 95% probability
(OxA-24810) (Bori≤ et al. 2014). From the previous
1970–1971 excavations, encompassing the central
area of the settlement, a roe deer skull with antlers
found on the floor of Building 2 (Srejovi≤, Letica
1978) produced a date of 7047–6699 cal BC at 95%
probability (OxA-16216) (Bori≤ et al. 2008). There-
fore, it can be concluded that the majority of these
events occurred during the last century of the 8th
millennium and the first half of 7th millennium BC,
and could be taken as representative of Late Mesoli-
thic resource exploitation patterns at the site.
Vesna Dimitrijević, Ivana ?ivaljević and Sofija Stefanović
108
Fig. 3. 3rd and 4th decidious premolar from a fragmented red deer left maxilla (VL 216/15) from Vlasac,
indicating an age of 7–8 months or a December-January capture: a lingual view; b buccal view; c roots,
apical view.
Becoming sedentary| The seasonality of food resource exploitation in the Mesolithic-Neolithic Danube Gorges
109
Similarly to mammal remains, the fish faunal assem-
blage from 2006–2009 displayed significant differ-
ences in comparison to the one collected in 1970–
1971. Only common carp (Cyprinus carpio), catfish,
pike (Esox lucius), unidentified cyprinids and a great
number of unidentified fish bones were reported by
Bökönyi (1978). The absence of sturgeon was clear-
ly a result of author-related bias, given that sturgeon
remains were in fact identified in the preserved fau-
nal sample from 1970–1971 (Bori≤ 2002.App. 5), as
well as in the sample from new excavations of Vla-
sac (Ωivaljevi≤ in prep.) (Tab. 4). Albeit few, the re-
mains of migratory beluga, Russian sturgeon and
stellate sturgeon are indicative of spring and autumn
fishing at the site, given that sturgeon remains as a
rule tend to be underrepresented due to their large-
ly cartilaginous skeleton (Brinkhuizen 1986; Barto-
siewicz et al. 2008). The exploitation of anadro-
mous Black Sea shad (Alosa immaculata) would
have been even more seasonally restricted, as the
species is present in the Danube only between the
end of March and the beginning of June (Tab. 4).
However, the vast majority of fish remains from the
new excavations of Vlasac originated from yet anoth-
er anadromous species – the vyrezub (Rutilus
frisii) of the cyprinid family, mainly due to the large
number of its pharyngeal teeth found in burials, but
also in ‘loose’ contexts. Burials with pharyngeal
teeth ornaments, from both the old and new excava-
tions of Vlasac have mainly been related to the Late
Mesolithic phase, but have also been found in a
Transformation phase burial at Vlasac and settle-
ment deposits at Lepenski Vir (Cristiani, Bori≤ 2012;
Cristiani et al. 2014; Bori≤ et al. 2014; Ωivaljevi≤ in
prep.), whereas unmodified vyrezub teeth and bones
occur as early as mid-10th millenium BC (Early Me-
solithic) at Lepenski Vir (Ωivaljevi≤ in prep.). The
species has not been reported in the Middle/Lower
Danube previously, but it was probably migrating
into the river from the Black Sea at least up to the
Middle Holocene (Ωivaljevi≤ et al. submitted). Mo-
dern Black Sea populations enter the rivers in Octo-
ber, overwinter in freshwater, and return to estuar-
ies after spawning, which takes place in April and
May (Kottelat, Freyhof 2007). Therefore, it can be
hypothesised that the main fishing seasons at Vla-
sac corresponded with their autumn migration and
spring spawning season, although the possibilities
of winter catches should not be excluded. The possi-
bility of opportunistic winter exploitation of fresh-
water fish should also be allowed, although fishing
during this time would have been difficult, as most
species take refuge on the river bottom and hiber-
nate until spring. The remains of pike, one of the
few species most succesfully caught from January to
April (Tab. 4), could be indicative of winter fishing
at Vlasac.
The bird faunal assemblage from both new and old
excavations of Vlasac contained the remains of aqua-
tic birds and waterfowl, as well as birds of prey and
songbirds. Given that bird remains were few even
in the hand-collected, water-sieved and flotated sam-
ple from the new excavations (Tab. 5), it may be hy-
pothesised that their exploitation was opportunistic.
A somewhat greater frequency of imperial eagle
(Aquila heliaca) and white-tailed eagle (Haliaeëtus
albicilla) remains (Bökönyi 1978; Dimitrijevi≤ et
al. submitted), and the spatial clustering of some of
their remains around the area of Late Mesolithic
Building 2 (Bori≤ 2002.190) could suggest that these
birds of prey were possibly held in special regard.
Most of the species represented in the 1970–1971
and 2006–2009 assemblage are still resident in the
Danube Gorges, and therefore their presence cannot
be used as a seasonal indicator. The only exceptions
Fig. 4. Red deer permanent and milk front teeth (VL 237/4) from a mandible that had not been preserv-
ed from Vlasac, indicating an age of 15–18 months or an August-November kill: a right milk 2nd incisor;
b right milk 3rd incisor; c right milk canine; d a germ of the left 3rd permanent incisor; e left 2nd milk in-
cisor; f left 3rd milk incisor.
Vesna Dimitrijević, Ivana ?ivaljević and Sofija Stefanović
110
stuck to the crushed skull bones. The age of the ani-
mal has been determined fairly accurately on the
basis of its upper teeth, which were in the last stages
of decidious/ permanent dentition replacement. In
both right and left maxilla (LV 273/4), 3rd decidious
molars were still holding above the crowns of per-
manent fourth premolars, and it would have been a
matter of days before they fell out. The 2nd and 3rd
premolars, and the left 3rd molar were erupting. This
stage of dental development is related to an age of
27 months, which would signal that this individual
was caught in October (Dimitrijevi≤ 2000; 2008; cf.
Brown, Chapman 1991).
Another example is a red deer skull (LV 261/1–2)
deposited on the floor of House 22 (Dimitrijevi≤
2008.Fig. 3). At present, only antlers have been pre-
served, one of which produced a date of 6084–5926
cal BC at 95% probability (OxA-16075) (Bori≤, Di-
mitrijevi≤ 2009), but both the skull (placed upside
down) and antlers can be seen in a photograph
taken during excavation (Fig. 5). The animal was
just about to shed its antlers at the moment of cap-
ture; as a result, they parted from the skull when ex-
cavated. This occurrence is a fairly reliable indica-
tor that the kill took place in February-March, and
the deposition of the skull shortly after. Given that
the abandonment of aforementioned House 28 was
marked by a red deer skull from an individual caught
in October, these two instances indicate that the ‘life
cycles’ of each building were probably unique, end-
ing with similar preparations but at different times
throughout the year. It would seem that certain buil-
dings were in use after others fell into disuse; this
practice is also suggested by the deposition of sculpt-
ed boulders. Most boulders were initially placed in
the rear area of rectangular hearths; however, cer-
tain dwellings had empty sockets in the floor behind
the hearth area, or merely a fragment of an origi-
nally larger boulder inserted in the socket, which
suggests that some of the boulders were moved from
abandoned buildings and inserted at new locations
(Srejovi≤ 1969; Bori≤ 2005). Consequently, the data
on seasons of abandonment enables a better under-
standing of the individual histories of the houses
and the occupation of the settlement as a whole.
Another fairly accurate season indicator comes from
a context underneath of the floor of House 31, where
a cache of animal bones, including a large part of a
red deer and a brown bear skeleton, as well as re-
mains of chamois, wild boar, beluga, Russian stur-
geon, common carp, vyrezub, catfish, huchen and an
unidentified bird bone have been found (Dimitrije-
include the remains of white pelican (Pelecanus cf.
onocrotalus) which migrates to the area in October-
November, black kite (Milvus migrans) and imperi-
al eagle which are passage visitors, present from
spring to autumn, and fieldfare (Turdus pilaris)
which overwinters in the area (from November to
March) (Tab. 5). In the case of the resident cormo-
rant (Phalacrocorax carbo), teal (Anas crecca),
and mallard (Anas platyrhynchos), it should be
noted that their populations increase in winter,
joined by visitors from Northwestern Europe (cf. Ra-
∏ajski, Kiss 2004; Clason 1980.164). According to
Anne Pike-Tay et alii (2004), teal and mallard are
migratory, and are present in neighbouring Hungary
in early spring and in autumn (the former used to
breed in Hungary), whereas the great white egret
(Egretta alba) and ferruginous duck (Aythya nyro-
ca) leave the area in winter. The data on present po-
pulations in Serbia and the Danube Gorges (Ra∏aj-
ski, Kiss 2004; Gruba≠ et al. 2013) suggest that these
are resident species, so their past migratory status is
less clear.
Lepenski Vir
The hand-collected and partially preserved faunal
assemblage from Lepenski Vir contained fewer ele-
ments sensitive to seasonality (Tab. 2), but their con-
textual provenance enables a better understanding
of the use of trapezoidal-base dwellings and the site
as a whole. At least 13 buildings contained red deer
skulls with antlers (Bökönyi 1969; 1970; 1972),
which were probably deposited on the floors, mark-
ing the event of house abandonment (Dimitrijevi≤
2000; 2008). Similarly to the structurally deposited
red deer skull from Vlasac, these could have been
curated over longer periods of time, but in a couple
of cases provided a fairly reliable indication of the
kill season.
One such example comes from House 28 in the west-
ernmost (upstream) periphery of the settlement,
where a red deer skull with unshed antlers (very
thin and quite unusual in their asymmetry) had
been deposited on the floor (Dimitrijevi≤ 2000.Fig.
16; 2008.Fig. 7). The skull has been directly dated
to a range 6206–5989 cal BC at 95% probability
(OxA-16078), which corresponds with the construc-
tion and use of the trapezoidal buildings, i.e. the
phase identified as the Mesolithic-Neolithic Transfor-
mation (Bori≤, Dimitrijevi≤ 2009). The skull was
damaged during excavation, with only the antlers
and frontal, temporal and occipital parts of the skull
and the upper jaws remaining (Dimitrijevi≤ 2000.
Fig. 17), while fish and gastropod remains were
Becoming sedentary| The seasonality of food resource exploitation in the Mesolithic-Neolithic Danube Gorges
111
vi≤ 2000; Bori≤, Dimitrijevi≤ 2006; Ωivaljevi≤ in
prep.). The context has not been AMS dated, but
given that the deposition of faunal remains predat-
ed the house construction, it can be concluded that
the event(s) took place during an early settlement
phase. The unity of the assemblage is seen in the
large portions of the red deer and brown bear car-
casses, which were probably butchered in situ. The
red deer remains originated from a young individ-
ual, given its milk dentition, unfused proximal and
distal epiphyses of long bones and porous structure
of the short bones. Its age has been determined on
the basis of a left mandible (LV bb-30/2) with a 2nd
and 3rd decidious molar, alveole for the 4th decidi-
ous molar and an erupting 1st permanent molar (Di-
mitrijevi≤ 2000.Fig. 12). This stage of dental devel-
opment and wear indicate a four-month-old indivi-
dual, suggesting that the kill season was in Septem-
ber (Dimitrijevi≤ 2000; cf. Brown, Chap-
man 1991; Carter 2006). If we assume that
the whole faunal assemblage underneath
House 31 accumulated over a short period,
September would correspond with the start
of the autumn migration of beluga and Rus-
sian sturgeon (Tab. 4), slightly preceding
the start of vyrezub migration, if migratory
patterns of modern populations are taken
into account.
Another red deer mandible (LV bb-14/3)
(Fig. 6) from an unspecified context origi-
nated from a 13- to 15-month-old individ-
ual, judging from the presence of 2nd, 3rd
and 4th decidious molars with formed but
unresorbed roots (cf. Brown, Chapman
1991). This specimen is indicative of a sum-
mer kill event that took place between July
and September.
An autumn/winter kill is suggested in the
case of a red deer left maxilla (LV 1074/5)
found in the eastern part of the settlement
(sq. a/11-14). The presence of both decidious 4th
molar and the 1st permanent molar indicates a 12-
to 16-month-old individual (cf. Habermehl 1985),
captured between October and February. The same
season has been determined in the case of red deer
maxillae (LV 1296/1) found beneath the floor of
House 19. Both left and right maxilla had mixed de-
cidious/permanent dentition (D4, M1; and D3-D4,
M1, respectively) (Fig. 7), suggesting they also orig-
inated from a 12- to 16-month-old individual (cf.
Habermehl 1985). In addition, vyrezub pharyngeal
teeth were stuck to the right maxilla. The October-
February time span corresponds well with the sea-
son during which vyrezub migrate to rivers and over-
winters, but precedes its spawning season (Tab. 4).
The red deer maxilla with vyrezub teeth, as well as
the aforementioned red deer skull from House 28
Fig. 5. Red deer skull with antlers (LV 261/1–2) deposited
facing the floor of House 22 at Lepenski Vir; the kill took
place in February-March. Photo from the archive of the Com-
puter Documentation Centre, Faculty of Philosophy, Belgrade.
Fig. 6. Red deer right mandible with milk molars (LV bb-14/3) from Lepenski Vir, originating from a 13–
15 month old individual, killed between July and September: a buccal; b lingual.
with small fish bones stuck to it seem to embody
quite literally the merging role of wild game and
fish. The fish faunal assemblage from Lepenski Vir
consisted mainly of cyprinid remains (of freshwater
species and anadromous vyrezub), as well as catfish,
huchen and a small number of Black Sea shad, pike
and pikeperch (Sander lucioperca) bones, but also
a significant amount of remains of migratory stur-
geon. Having in mind their largely cartilaginous ske-
leton and preservation bias affecting their remains
(Brinkhuizen 1986; Bartosiewicz et al. 2008), it is
remarkable that sturgeon constitute nearly 20% (in
terms of NISP) of identified fish remains at Lepen-
ski Vir (Ωivaljevi≤ in prep.); it is also highly signifi-
cant that some of the sculpted boulders from this
site bore a resemblance to sturgeon features (Rado-
vanovi≤ 1997; Bori≤ 2005). At the same time, the
Lepenski Vir assemblage exhibited the greatest di-
versity of sturgeon species, including the remains of
migratory beluga, Russian sturgeon, ship sturgeon,
and stellate sturgeon, as well as those of freshwater
sterlet (Acipenser ruthenus). The first species to
enter the Danube is the large beluga, whose spring
migration commences in late winter/early spring,
reaching a peak in April and ending in late May/
June. Its autumn migration starts by late August,
peaking in October/November. The Russian sturgeon
starts its spring migration somewhat later than bel-
uga (end of February/March), which peaks in April,
and ends mid-May. The second migration starts in
August/September, peaks in late September/mid-Oc-
tober and ends by mid-November. The ship sturgeon
migrates in spring and autumn, with autumn popu-
lations remaining in the river until spawning, which
takes place from March/April until June. The stellate
sturgeon, the last species to enter the Danube, starts
its spring migration in late April/ March, reaching
a peak in May. Its autumn run begins by August and
ends by September/mid-October (Tab. 4) (cf. Dinu
2010; Petrovi≤ 1998[1941]; Risti≤ 1977; Simonovi≤
2001; Kottelat, Freyhof 2007; Bartosiewicz et al.
2008). Judging from interspecific variations in the
migration start and peak dates of sturgeon species,
migration pattern of vyrezub, as well as availabili-
ty of freshwater fish, it might be hypothesised that
fishing could have successfully taken place between
late winter and late autumn, with particular inten-
sity in spring and autumn. The remains of species
active in winter (namely pike) are few, but the pos-
sibility of smoking and drying fish (the hearths in
particular could have been used for this purpose)
and storing it for winter months should not be ex-
cluded (cf. Bonsall 2008; Bori≤ 2002; 2007). On the
basis of fish remains from AMS dated contexts from
Lepenski Vir (cf. Whittle et al. 2002; Bonsall et al.
1997; 2008; Bori≤, Dimitrijevi≤ 2009; Bori≤ 2011),
it was determined that fishing patterns remained
relatively unchanged throughout the temporal se-
quence, i.e. that the distribution and proportion of
species is more or less the same in both Mesolithic
and Transformation phase contexts at the site (Ωi-
valjevi≤ in prep.).
Although bird remains have been uncovered at Le-
penski Vir (Bökönyi 1969; 1970; 1972; Dimitrijevi≤
2008), only a single bird bone found between the
superimposed floors of Houses 20 and 33 (context
dated to 6231–6056 cal BC at 95% probability (OxA-
15998) (Bori≤, Dimitrijevi≤ 2009) has been iden-
tified to the species level, originating from mallard
(Bori≤, Dimitrijevi≤ 2006). As already noted, Anne
Pike-Tay et alii (2004) state that the species is mi-
gratory, but used to breed in Hungary; whereas Ja-
vor Ra∏ajski and Andrei Kiss (2004) note that it is
Vesna Dimitrijević, Ivana ?ivaljević and Sofija Stefanović
112
Fig. 7. Red deer left and
right fragmented maxil-
lae (LV 1296/1) found be-
neath the floor of House
19 at Lepenski Vir, origi-
nating from a 12–16
month old individual,
captured between Octo-
ber and February. A frag-
ment of vyrezub left pha-
ryngeal bone with a row
of 4 preserved teeth is
attached to the right max-
illa. a a fragment of the
right maxilla with the
two milk molars (D3 and
D4) and the first molar (M1) in eruption, lingual, attached vyrezub pharyngeal bone at the left; b
same, buccal view, vyrezub teeth seen at the right side; c a fragment of the left maxilla with the last
milk (D4) and the first permanent molar (M1), lingual; d same, buccal view.
Becoming sedentary| The seasonality of food resource exploitation in the Mesolithic-Neolithic Danube Gorges
113
resident in Serbia, with a winter population increase
due to migrant populations from the North of Europe.
Padina
The faunal assemblage from Padina, albeit entirely
preserved, contained the fewest elements for a reli-
able estimate of occupational seasons (Tab. 3). Si-
milarly to Lepenski Vir, this was mainly due to the
hand-collection of animal bone, particularly biased
towards fragile deciduous teeth and fish and bird
remains. The majority of elements indicative of sea-
sonality included shed and unshed red deer and roe
deer antlers from both Mesolithic and Transforma-
tion phase contexts. These indicate kill seasons which
occurred between September and February, deter-
mined on the basis of unshed red deer antlers, and
May and October, on the basis of unshed roe deer
antlers. The occurrence of shed antlers of red and
roe deer indicate occupation in February-March and
October-November, respectively, assuming they were
collected shortly after shedding. The only mandible
from a juvenile animal (PAD 28.8.69/1) originated
from a roe deer, and was found in a Mesolithic occu-
pational context (cf. Bori≤ 2002.App. 3). Based on
the presence of alveoli for D2-D3, D4 and a fragment
of M1 (Fig. 8), it was determined that the animal
was between four and eight months old (cf. Carter
2006), which suggested that it was captured between
September and January.
In terms of fish remains, a specific feature of the
faunal assemblage from Padina (even in comparison
with other assemblages collected by hand) is the
absolute predominance of catfish (69.9% of all iden-
tified fish remains, and 77.1% of all fish remains
identified to species) and least species diversity
(Tab. 4) (Ωivaljevi≤ in prep.). Moreover, judging
from the species distribution in AMS dated contexts
(cf. Whittle et al. 2002; Bori≤, Miracle 2004; Bori≤
2011) and in contexts attributed to particular phas-
es on the basis of stratigraphy (cf. Bori≤ 2002.App.
3), it was determined that catfish remains become
exceptionally abundant in the Transformation phase
of the settlement. This could suggest that Padina
established itself as a specialised centre oriented to-
wards catching catfish during this time (Ωivaljevi≤
in prep.). Catfish is most successfully caught during
its spawning season (from May to July) and during
summer months, when it feeds most intensely (Ri-
sti≤ 1977; Simonovi≤ 2001; Kottelat, Freyhof 2007;
Dinu 2010); however, given that the species is sta-
tionary, and hardly moves from its respective habi-
tat (Risti≤ 1977), it can be hypothesised that it was
available for most of the year. During cold seasons,
however, catfish populations would decrease in
number, as the species requires a water temperature
of at least 18°C to reproduce (cf. Bartosiewicz, Bon-
sall 2004.Tab. 7). As already suggested, the harsh
winter seasons and periods when the river was co-
vered in ice could be overcome by storing fish by
smoking and drying (cf. Bonsall 2008; Bori≤ 2002;
2007). In addition, the presence of anadromous
sturgeons and vyrezub in the faunal assemblage sug-
gests that freshwater fish exploitation at Padina was
supplemented with fishing of seasonally available
species, most notably in spring and autumn.
Bird remains are few in the faunal assemblage from
Padina, which is suggestive of opportunistic exploi-
tation; however, it is probably also a consequence of
the sampling bias. The faunal assemblage consisted
of the remains of six resident and three migratory
species, some of them available in the Danube Gor-
ges area for fairly short and seasonally limited pe-
riods of time (Tab. 5). Such is the case with the shell-
duck (Tadorna tadorna), which is present only in
March and September-October, and white pelican,
present in October-November. The black-throated
diver (Gavia arctica) is present in the area from No-
vember to March (Ra∏ajski, Kiss 2004), whereas in
Central Europe it migrates in spring and autumn (Cla-
son 1980). As already noted, the numbers of some
of the resident populations (such as cormorant) in-
crease in winter (cf. Clason 1980; Ra∏ajski, Kiss
2004), which tentatively suggests they could have
been exploited more frequently during this time.
Discussion
An important issue to be addressed here is whether
permanent, year-round occupation can be distingui-
shed from a series of repeated visits, of varying du-
Fig. 8. Roe deer mandible with alveoli for milk mo-
lars (D2-D3), last milk molar (D4) and a fragment
of the first permanent molar (PAD 28.8. 69/1) from
a Mesolithic occupational context at Padina. The
animal was between 4–8 months old, suggesting
the capture occurred between September and Janu-
ary.
Vesna Dimitrijević, Ivana ?ivaljević and Sofija Stefanović
114
ration, occurring at different seasons throughout
the year? (cf. Whittle 2001; Conneller et al. 2012).
This question, of course, remains very difficult to
answer archaeologically. The patterns of food re-
source exploitation at Vlasac, corroborated with
archaeological evidence, could be indications of the
first scenario. The intensity of occupation in the Late
Mesolithic phase (c. 7400–6300/6200 cal BC) is evi-
denced by several dug-out dwellings vaguely resem-
bling the trapezoidal shape in plan, numerous stone-
lined hearths which were constructed in the open or
were originally incorporated into dwellings that left
no trace in the archaeological record, over hundred
buried individuals and an enormous assemblage of
tools and animal bones (cf. Srejovi≤, Letica 1978;
Nemeskéri 1978; Bökönyi 1978; Kozłowski, Koz-
łowski 1982; Roksandi≤ 1999; 2000; Bori≤ et al.
2008; 2014). The faunal assemblage from the new
2006–2009 excavations of the site, which contained
most elements suggestive of kill seasons (namely,
erupting teeth of young animals), seems to indicate
year-round occupation of the settlement. The exploi-
tation patterns of red deer, the most important game
animal, do not suggest any significant seasonal hia-
tus. Similarly, wild boar and roe deer hunting does
not seem restricted to a particular season. Although
most elements utilised in this study indicated a rel-
atively broad time span, those that provided most
precise and temporally restricted seasonal estimates
point to hunting events which took place in all sea-
sons. The fishing activity is indicative of spring and
autumn months, judging from the presence of ana-
dromous vyrezub and various sturgeon species, but
fishing at other times was certainly possible, given
that freshwater species represented in the sample
would have been available for longer periods of
time. Moreover, the fish could have been stored for
winter by smoking and drying.
The hand-collected faunal assemblages from Lepen-
ski Vir and Padina offered much less information for
precise season determination, but tentative estimates
do not point to a particular season of hunting. Fur-
thermore, it can be suggested that resource exploita-
tion patterns observed in the Late Mesolithic would
have continued in the Transformation phase settle-
ments (c. 6200–5900 cal BC). Several fairly precise
season estimates obtained from specimens related
to the construction and use of elaborate trapezoidal-
base buildings at Lepenski Vir offer important in-
sights into the life-cycles of particular buildings and,
consequently, of the settlement as a whole. It is evi-
dent that some of dwellings were abandoned at dif-
ferent times of the year, while others remained in
use; this feature is also suggested by the secondary
position of some of the sculpted boulders. The signi-
ficance of both hunting and fishing activities, which
took place simultaneously for most of the year, is
particularly illustrated by caches of mammal and
fish bones stuck together and formed over a short
period of time. It remains unclear to what extent the
observed differences in preferences for fish species
at Lepenski Vir and Padina would influence the na-
ture of occupation of these sites; the latter oriented
mostly towards catfish available for most of the year,
and the former exhibiting greater species diversity
and a significant amount of seasonably available
sturgeon. Whereas the environment of Lepenski Vir
and other sites would have provided substantial
means for occupation lasting throughout the year,
the exploitation of seasonally available fish could
have been vital in (re)creating memories, events
and specific temporal perspectives (Bori≤ 2002).
Conclusion
This study offers additional data supporting the hy-
pothesis that a significant degree of sedentism in
the Danube Gorges existed already in the Late Me-
solithic, prior to, and independently of, the adoption
of animal and plant husbandry. Furthermore, recent
studies of other Mesolithic contexts in Europe pro-
vide evidence of practices involving prolonged stays
of human groups in certain locations; for example,
the construction (involving a significant amount of
time and effort) of substantial timber structures at
Starr Car in Britain (Conneller et al. 2012) and large-
scale food storage and fish fermentation, implying
a delayed-return subsistence strategy, at Norje-Sun-
nansund in Southern Scandinavia (Boethius 2016).
Moreover, ‘full-scale’ sedentism commonly associated
with food-producing communities could have also
taken many different forms, including shifting culti-
vation crops or herds, and varying levels of mobili-
ty of individuals involved in such practices (Whittle
2001). These studies further problematise the rela-
tionship of sedentism and subsistence strategies, and
add to our understanding of the complexity and
diversity of forager and food-producing lifestyle and
residential practices. In the Danube Gorges, evidence
of Early Mesolithic occupation is relatively scarce,
but the presence of several burials and stone con-
structions indicate a prolonged stay at riverine ter-
races. The higher resolution of dates, architectural
features and artefacts related to the Late Mesolithic
occupancy signal that by this time these locations
were a home to fairly sedentary populations prac-
ticing an exclusively hunter-gatherer-fisher lifestyle.
Becoming sedentary| The seasonality of food resource exploitation in the Mesolithic-Neolithic Danube Gorges
115
Their environment enabled the successful exploita-
tion of wild game and fish for most, if not the whole,
of the year.
Therefore, it can be argued that the construction of
later, Mesolithic-Neolithic Transformation phase set-
tlements at Lepenski Vir and Padina took place in
a context where continuous human occupation and
relationship with particular locales had been long
established. It can also be hypothesised that these
settlements, with more elaborate and permanent
dwellings, would have followed similar patterns of
occupation. Although the changes observed in the
Transformation phase largely coincide with the
emergence of the first food-producing communities
in the region (and contacts with distant communi-
ties are evident already in the Late Mesolithic), cer-
tain (but not all) elements of Neolithic lifestyles were
incorporated only after “they made sense in local
terms” (Bori≤ 2007). Most notably, animal husban-
dry had fairly little impact on the economy and life-
style of Upper Gorge communities, with the remains
of the first domesticates appearing only after most
of the trapezoidal buildings had been abandoned.
It can be assumed then, that local traditions, estab-
lished in a particular environment and involving
locally available resources, merged with new cultur-
al elements in the later parts of the sequence, played
substantial roles in settlement practices in the Da-
nube Gorges context.
This paper is a result of the Project “BIRTH: Births,
mothers and babies: prehistoric fertility in the Balkans
between 10 000–5000 BC”, funded by the European
Research Council (ERC) under the European Union’s
Horizon 2020 research and innovation programme
(Grant Agreement No. 640 557). We thank Du∏an Bo-
ri≤ for the opportunity to study the faunal remains
from the 2006–2009 excavations of Vlasac, Igor Aske-
yev for his help with determining some of the fish
bones, and Mihael Budja for his kind invitation to
participate in the 22nd Neolithic Seminar and for the
superb organisation of the conference. We would also
like to thank two anonymous reviewers for their in-
sightful and constructive comments on earlier drafts
of this paper.
ACKNOWLEDGEMENTS
References
Bartosiewicz L. 1989. Variability of tooth eruption in cat-
tle. Acta Veterinaria Hungarica 37(4): 303–317.
Bartosiewicz L., Bonsall C. 2004. Prehistoric fishing along
the Danube. Antaeus 27: 253–272.
Bartosiewicz L., Bonsall C. and Sisu V. 2008. Sturgeon fish-
ing along the Middle and Lower Danube. In C. Bonsall,
V. Boroneant and I. Radovanovi≤ (eds.), The Iron Gates
in Prehistory. New perspectives. British Archaeological
Reports IS 1893. Archaeopress. Oxford: 39–54.
Bartosiewicz L., Boroneant V., Bonsall C. and Stallibrass S.
1995. Schela Cladovei: a preliminary review of the prehi-
storic fauna. Mesolithic Miscellany 16(2): 2–19.
2001. New data on the prehistoric fauna of the Iron
Gates: a case study from Schela Cladovei, Romania. In
R. Kertész, J. Makkay (eds.), From the Mesolithic to the
Neolithic. Proceedings of the International Archaeologi-
cal Conference held in the Damjanich Museum of Szol-
nok, September 22–27, 1996. Archaeolingua. Budapest:
15–21.
Blaise E., Balasse M. 2011. Seasonality and season of birth
of modern and late Neolithic sheep from south-eastern
France using tooth enamel δ18O analysis. Journal of
Archaeological Science 38(11): 3085–3093.
Boethius A. 2016. Something rotten in Scandinavia: The
world’s earliest evidence of fermentation. Journal of Ar-
chaeological Science 66: 169–180.
Bökönyi S. 1969. Ki≠menjaci (prethodni izve∏taj). In D.
Srejovi≤ (ed.), Lepenski Vir. Nova praistorijska kultura u
Podunavlju. Srpska knji∫evna zadruga. Beograd: 224–228.
1970. Animal Remains from Lepenski Vir. Science 167
(926): 1702–1704.
1972. The vertebrate fauna. In D. Srejovi≤ (ed.), Eu-
rope’s first monumental sculpture: new discoveries at
Lepenski Vir. Thames and Hudson. London: 186–189.
1978. The vertebrate fauna of Vlasac. In M. Gara∏anin
(ed.), Vlasac. A Mesolithic Settlement in the Iron Gates.
Volume II. Geology – Biology – Anthropology. Serbian
Academy of Sciences and Arts Monographies. Belgrade:
35–65.
Bonsall C. 2008. The Mesolithic of the Iron Gates. In G. N.
Bailey, P. Spikins (eds.), Mesolithic Europe. Cambridge
University Press. Cambridge: 238–279.
∴
Vesna Dimitrijević, Ivana ?ivaljević and Sofija Stefanović
116
Bonsall C., Cook G. T., Hedges R. E. M., Higham T. F. G.,
Pickard C. and Radovanovi≤ I. 2004. Radiocarbon and sta-
ble isotope evidence of the dietary change from the Me-
solithic to the Middle Ages in the Iron Gates: new results
from Lepenski Vir. Radiocarbon 46: 293–300.
Bonsall C., Lennon R., McSweeney K., Stewart C., Hark-
ness D., Boroneant V., Bartosiewicz L., Payton R. and Chap-
man J. 1997. Mesolithic and Early Neolithic in the Iron
Gates: a palaeodietary perspective. Journal of European
Archaeology 5(1): 50–92.
Bonsall C., Radovanovi≤ I., Roksandi≤ M., Cook G., Hig-
ham T. and Pickard C. 2008. Dating burials and architec-
ture at Lepenski Vir. In C. Bonsall, V. Boroneant and I. Ra-
dovanovi≤ (eds.), The Iron Gates in Prehistory. New per-
spectives. British Archaeological Reports IS 1893. Archaeo-
press. Oxford: 175–204.
Bori≤ D. 2001. Mesolithic and early Neolithic hunters and
fishers in the Danube Gorges: an analysis of archaeozo-
ological data. In R. Kertész, J. Makkay (eds.), From the
Mesolithic to the Neolithic. Proceedings of the Internatio-
nal Archaeological Conference held in the Damjanich Mu-
seum of Szolnok, September 22–27, 1996. Archaeolingua.
Budapest: 101–124.
2002. Seasons, Life Cycles and Memory in the Danube
Gorges, c. 10,000–5500 BC. Unpublished PhD thesis.
University of Cambridge. Cambridge.
2005. Body metamorphosis and animality: volatile bod-
ies and boulder artworks from Lepenski Vir. Cambridge
Archaeological Journal 15(1): 35–69.
2006. New discoveries at the Mesolithic-Early Neolithic
site of Vlasac: preliminary notes. Mesolithic Miscellany
18(1): 7–14.
2007. The House between Grand Narrative and Micro-
history: a house society in the Balkans. In R. A. Beck
(ed.), The Durable House. House Society Models in
Archaeology. South Illinois University, Center for Ar-
chaeological Investigations, Occasional Paper No. 35.
Carbondale: 97–129.
2010. Happy forgetting? Remembering and dismem-
bering dead bodies at Vlasac. In D. Bori≤ (ed.), Archaeo-
logy and Memory. Oxbow Books. Oxford: 48–67.
2011. Adaptations and transformations of the Danube
Gorges foragers (c. 13,000–5500 BC): an overview. In
R. Krauß (ed.), Beginnings – New Research in the Ap-
pearance of the Neolithic between Northwest Anato-
lia and the Carpathian Basin. Verlag Marie Leidorf
Gmbh. Rahden: 157–203.
Bori≤ D., Dimitrijevi≤ V. 2006. Continuity of foraging stra-
tegies in the Mesolithic-Neolithic transformations: dating
faunal patterns at Lepenski Vir (Serbia). Atti della Socie-
tà per la preistoria e protoistoria della regione Friuli-
Venezia Giulia XV (2004–05): 33–107.
2007. When did the ‘Neolithic package’ reach Lepenski
Vir? Radiometric and faunal evidence. Documenta
Praehistorica 34: 52–71.
2009. Apsolutna hronologija i stratigrafija Lepenskog
Vira. Starinar LVII/2007: 9–55.
Bori≤ D., French C. and Dimitrijevi≤ V. 2008. Vlasac re-
visited: formation processes, stratigraphy and dating. Do-
cumenta Praehistorica 35: 261–287.
Bori≤ D., French C. A. I., Stefanovi≤ S., Dimitrijevi≤ V., Cri-
stiani E., Gurova M., Antonovi≤ D., Allué E. and Filipovi≤
D. 2014. Late Mesolithic lifeways and deathways at Vla-
sac (Serbia). Journal of Field Archaeology 39(1): 4–31. 
Bori≤ D., Grupe G., Peters J. and Miki≤ Ω. 2004. Is the Me-
solithic-Neolithic subsistence dichotomy real? New stable
isotope evidence from the Danube Gorges. European
Journal of Archaeology 7(3): 221–248.
Bori≤ D., Miracle P. 2004. Mesolithic and Neolithic (dis)
continuities in the Danube Gorges: new AMS dates from
Padina and Hajdu≠ka Vodenica (Serbia). Oxford Journal
of Archaeology 23(4): 341–371.
Bori≤ D., Price T. D. 2013. Strontium isotopes document
greater human mobility at the start of the Balkan Neoli-
thic. Proceedings of the National Academy of Sciences
of USA 110(9): 3298–3303.
Brinkhuizen D. C. 1986. Features observed on the skele-
tons of some recent European Acipenseridae: their impor-
tance for the study of excavated remains of sturgeon. In
D. C. Brinkhuizen, A. T. Clason (eds.), Fish and Archaeo-
logy. Studies in osteometry, taphonomy, seasonality and
fishing methods. British Archaeology Reports IS 294.
Archaeopress. Oxford: 18–33.
Brown W. A. B., Chapman N. G. 1991. The dentition of
red deer (Cervus elaphus): a scoring scheme to assess age
from wear of the permanent molariform teeth. Journal
of Zoology 224(4): 519–536.
Bull G., Payne S. 1982. Tooth eruption and epiphyseal
fusion in pigs and wild boar. In B. Wilson, C. Grigson and
S. Payne (eds.), Ageing and sexing animal bones from
archaeological sites. British Archaeology Reports IS 109.
Archaeopress. Oxford: 55–71.
Carter R. J. 2006. A method to estimate the ages at death
of red deer (Cervus elaphus) and roe deer (Capreolus ca-
preolus) from developing mandibular dentition and its
application to Mesolithic NW Europe. In D. Ruscillo (ed.),
Recent advances in ageing and sexing animal bones.
Becoming sedentary| The seasonality of food resource exploitation in the Mesolithic-Neolithic Danube Gorges
117
Proceedings of the 9th ICAZ conference. Oxbow Books.
Oxford: 40–61.
Carter R. J., Magnell O. 2007. Age estimation of wild boar
based on molariform mandibular tooth development and
its application to seasonality at the Mesolithic site of Ring-
kloster, Denmark. In U. Albarella, K. Dobney, A. Ervynck
and P. Rowley-Konwy (eds.), Pigs and humans: 10,000
years of interaction. Oxford University Press. Oxford:
197–217.
Chapman J. 1993. Social power in the Iron Gates Mesoli-
thic. In J. Chapman, M. P. Dolukhanov (eds.), Cultural
Transformations and Interactions in Eastern Europe.
Aldershot. Avebury: 71–121.
Clason A. T. 1980. Padina and Star≠evo: game, fish and
cattle. Palaeohistoria 22: 141–173.
Clements J. F., Schulenberg T. S., Iliff M. J., Roberson D.,
Fredericks T. A., Sullivan B. L. and Wood C. L. 2014. The
eBird/Clements checklist of birds of the world. Version
6.9. http://www.birds.cornell.edu/clementschecklist/down
load/
Conneller C., Milner N., Taylor B. and Taylor M. 2012.
Substantial settlement in the European Early Mesolithic:
new research at Star Carr. Antiquity 86(2012): 1004–
1020.
Cristiani E., Bori≤ D. 2012. 8500-year-old Late Mesolithic
garment embroidery from Vlasac (Serbia): technological,
use-wear and residue analyses. Journal of Archaeologi-
cal Science 39(11): 3450–3469.
Cristiani E., Ωivaljevi≤ I. and Bori≤ D. 2014. Residue ana-
lysis and ornament suspension techniques in prehistory:
Cyprinid pharyngeal teeth beads from Late Mesolithic bu-
rials at Vlasac (Serbia). Journal of Archaeological Science
46: 292–310.
Dimitrijevi≤ V. 2000. The Lepenski Vir fauna: bones in
houses and between houses. Documenta Praehistorica
27: 101–117.
2008. Lepenski Vir animal bones: what was left in the
houses? In C. Bonsall, V. Boroneant and I. Radovanovi≤
(eds.), The Iron Gates in Prehistory. New perspectives.
British Archaeological Reports IS 1893. Archaeopress.
Oxford: 117–130.
Dimitrijevi≤ V., Vukovi≤ S. 2015. Was the dog locally do-
mesticated in the Danube Gorges? Morphometric study of
dog cranial remains from four Mesolithic–Early Neolithic
archaeological sites by comparison with contemporary
wolves. International Journal of Osteoarchaeology 25
(1): 1–30.
Dimitrijevi≤ V., Ωivaljevi≤ I. and Bla∫i≤ S. Submitted. Fau-
nal remains. In D. Bori≤ (ed.), Vlasac: a Mesolithic site in
the Danube Gorges revisited.
Dinu A. 2010. Mesolithic fish and fishermen of the lower
Danube (Iron Gates). Documenta Praehistorica 37: 299–
310.
Gara∏anin M., Radovanovi≤ I. 2001. A pot in House 54
at Lepenski Vir I. Antiquity 75: 118–125.
Gruba≠ B., Milovanovi≤ Z. and πekler M. 2013. Ptice ∑er-
dapa. Princip pres: Zavod za za∏titu prirode Srbije – Na-
cionalni park “∑erdap” – Veterinarski specijalisti≠ki in-
stitut “Kraljevo”. Beograd – Donji Milanovac – Kraljevo.
Grupa autora 1991. Velika ilustrovana enciklopedija lov-
stva. DIP, gra∂evinska knjiga – DNEVNIK. Beograd – Novi
Sad.
Grupe G., Miki≤ Ω., Peters J. and Manhart H. 2003. Verte-
brate food webs and subsistance strategies of Meso and
Neolithic populations of Central Europe. In G. Gruppe, J.
Peters (eds.), Documenta Archaeobiologiae 1. Jahrbuch
der Staatssammlung für Anthropologie und Paläoana-
tomie. Verlag Marie Leidorf, Rahden/Westf. München:
193–213.
Habermehl K.-H. 1985. Altersbestimmung bei Wild-und
Pelztieren. Verlag Paul Parey. Hamburg, Berlin.
Jovanovi≤ B. 2008. Micro-regions of the Lepenski Vir cul-
ture: Padina in the Upper Gorge and Hajdu≠ka Vodenica
in the Lower Gorge of the Danube. Documenta Praehisto-
rica 35: 289–324.
Kottelat M., Freyhof J. 2007. Handbook of European
Freshwater Fishes. Publications Kottelat. Cornol.
Kozłowski J. K., Kozłowski S. K. 1982. Lithic industries
from the multi-layered Mesolithic site Vlasac, in Yugosla-
via. In J. K. Kozłowski (ed.), Origin of the Chipped Stone
Industries of the Early Farming Cultures in the Balkans.
Prace archeologiczne 33. Panstwowe Wydawnictwo Nau-
kowe. Warszawa-Krakow: 11–109.
Nehlich O., Bori≤ D., Stefanovi≤ S. and Richards M. P.
2010. Sulphur isotope evidence for freshwater fish con-
sumption: a case study from the Danube Gorges, SE Eu-
rope. Journal of Archaeological Science 37: 1131–1139.
Nemeskéri J. 1978. Demographic structure of the Vlasac
Epipalaeolithic population. In M. Gara∏anin (ed.), Vlasac.
A Mesolithic Settlement in the Iron Gates. Volume II.
Geology – Biology – Anthropology. Serbian Academy of
Sciences and Arts Monographies. Belgrade: 97–133.
Petrovi≤ M. 1998. [orig. 1941]. ∑erdapski ribolovi u pro-
∏losti i sada∏njosti. In D. Trifunovi≤ (ed.), Ribarstvo. Sa-
Vesna Dimitrijević, Ivana ?ivaljević and Sofija Stefanović
118
brana dela Mihaila Petrovi≤a. Zavod za ud∫benike i nas-
tavna sredstva. Beograd: 175–270.
Pike-Tay A., Bartosiewicz L., Gál E. and Whittle A. 2004.
Body part representation and seasonality: sheep/goat,
bird and fish remains from Early Neolithic Ecsegfalva
23, SE Hungary. Journal of Taphonomy 2(4): 221–246.
Puzovi≤ S. 2000. Atlas ptica grabljivica Srbije – mape
rasprostranjenosti i procene brojnosti 1977–1996. Za-
vod za za∏titu prirode Srbije. Beograd.
Radovanovi≤ I. 1996. The Iron Gates Mesolithic. Interna-
tional Monographs in Prehistory. Ann Arbor.
1997. The Lepenski Vir culture: a contri-bution to in-
terpretation of its ideological aspects. In D. Srejovi≤, M.
Lazi≤ (eds.), Antidoron Dragoslavo Srejovi≤ completis
LXV annis ab amicis, collegis, discipulis oblatum. Cen-
tar za arheolo∏ka istra∫ivanja. Beograd: 87–93.
Radovanovi≤ I., Voytek B. 1997. Hunters, fishers or far-
mers: sedentism, subsistence and social complexity in the
∑erdap Mesolithic. In A. van Gijn, C. Bakels and M. Zvele-
bil (eds.), Ideology and Social Structure of Stone Age
Communities in Europe. Analecta Praehistorica Leiden-
sia 29: 19–31.
Ra∏ajski J., Kiss A. 2004. Ptice Banata. Gradski muzej Vr-
∏ac. Vr∏ac.
Risti≤ M. 1977. Ribe i ribolov u slatkim vodama. Nolit.
Beograd.
Roksandi≤ M. 1999. Transition from Mesolithic to Neoli-
thic in the Iron Gates Gorge: Physical Anthropology Per-
spective. Unpublished PhD thesis. Department of Anthro-
pology. Simon Fraser University. Vancouver.
Roksandi≤ M. 2000. Between foragers and farmers in the
Iron Gates Gorge: physical anthropology perspective. ∑er-
dap population in transition from Mesolithic to Neolithic.
Documenta Praehistorica 27: 1–100.
Schmid E. 1972. Atlas of Animal Bones: For Prehisto-
rians, Archaeologists and Quaternary Geologists. Else-
vier Publishing Company. Amsterdam, London, New York.
Simonovi≤ P. 2001. Ribe Srbije. NNK Internacional. Bio-
lo∏ki fakultet Univerziteta u Beogradu. Zavod za za∏titu
prirode Srbije. Beograd.
Srejovi≤ D. 1969. Lepenski Vir: Nova praistorijska kultu-
ra u Podunavlju. Srpska Knji∫evna Zadruga. Beograd.
1972. Europe’s first monumental sculpture: new disco-
veries at Lepenski Vir. Thames and Hudson. London.
Srejovi≤ D., Letica, Z. 1978. Vlasac. Mezolitsko naselje u
∑erdapu (I arheologija). Srpska akademija nauka i umet-
nosti. Beograd.
Takács I., Bartosiewicz L. 1998. Data on the history of fish
exploitation in Hungary. Internet Archaeology 4. http://
intarch.ac.uk/journal/issue4/takacs_index.html
Voytek B. A., Tringham R. 1989. Rethinking the Mesoli-
thic: the case of South-East Europe. In C. Bonsall (ed.),
The Mesolithic in Europe. Papers Presented at the Third
International Symposium Edinburgh 1985. John Donald
Publishers Ltd. Edinburgh: 492–499.
Wheeler A., Jones A. K. G. 1989. Fishes. Cambridge Ma-
nuals in Archaeology. Cambridge University Press. Cam-
bridge.
Whittle A. 2001. From mobility to sedentism: change by
degrees. In R. Kertész, J. Makkay (eds.), From the Mesoli-
thic to the Neolithic. Proceedings of the International Ar-
chaeological Conference held in the Damjanich Museum
of Szolnok, September 22–27, 1996. Archaeolingua. Buda-
pest: 447–461.
Whittle A., Bartosiewicz L., Bori≤ D., Pettitt P. and Richards
M. 2002. In the beginning: new radiocarbon dates for the
Early Neolithic in Northern Serbia and South-East Hun-
gary. Antaeus 25(2002): 63–117.
Ωivaljevi≤ I. in prep. Ribolov na ∑erdapu u ranom holo-
cenu (10. – 6. milenijum pre n. e.). Unpublished PhD the-
sis. Department of Archaeology. Faculty of Philosophy.
University of Belgrade. Belgrade.
Ωivaljevi≤ I., Popovi≤ D., Snoj A. and Mari≤ S. submitted.
Ancient DNA analysis of cyprinid remains from the Meso-
lithic-Neolithic Danube Gorges reveals an extirpated fish
species Rutilus frisii (Nordmann, 1840).
Becoming sedentary| The seasonality of food resource exploitation in the Mesolithic-Neolithic Danube Gorges
119
TA
X
O
N
IN
V
. N
O
.
ELEM
EN
T
W
IN
TER
SPR
IN
G
SU
M
M
ER
A
U
TU
M
N
R
eference w
here the age
JA
N
FEB
M
A
R
A
PR
M
A
Y
JU
N
JU
L
A
U
G
SEP
O
C
T
N
O
V
D
EC
stage has been defined
V
L 29\1
skull fragm
ent w
ith antlers
Schm
idt 1972
\
unshed antler (9)*
Schm
idt 1972
\
shed antler (31)*
Schm
idt 1972
V
L 216\15
m
axilla sin. (D
3, 4)
C
arter 2006
V
L 145 x.5
P4 inf. dext.
C
arter 2006
V
L 20
4\5
D
3 sup.sin.
C
arter 2006
R
ed deer
V
L 20
9 x.1
P4 inf. dext.sin.
C
arter 2006
C
ervus elaphus
V
L 224\6
D
2 inf.dext.
C
arter 2006
V
L 237\4
I3, D
i2, D
i3 sin. &
 D
i2, D
i3, D
c dext.
H
aberm
ehl 1985
V
L 264 x.2
D
3 sup.dext.
C
arter 2006
V
L 286\3
M
2, M
3  inf. dext.
C
arter 2006
V
L 30
6\24
P2 inf.sin.
C
arter 2006
V
L 35\2
P4 inf.sin.
C
arter 2006
V
L 66 x.2
P4 inf. dext.
C
arter 2006
V
L 314 x.23
skull fragm
ent w
ith left antler
Schm
idt 1972
R
oe deer
\
unshed antler (3)*
Schm
idt 1972
C
apreolus capreolus
V
L 51\1
D
3\4 sup.sin.
C
arter 2006
V
L 285\7
D
2 sup.sin.
C
arter 2006
Fox V
ulpes vulpes
V
L 222\22-23
D
c sup.dext. &
 M
1 sup.dext. (germ
)
H
aberm
ehl 1985
V
L 133 x.2
m
ax. sin. (D
4)
C
arter, M
agnell 2007** 
W
ild boar
V
L 237\5
D
3 sup. dext.
C
arter, M
agnell 2007** 
Sus scrofa
V
L 282\3
D
2 sup.dext.
C
arter, M
agnell 2007** 
V
L 30
8 x.9
D
4 inf.sin.
C
arter, M
agnell 2007
V
L 242\1
M
2 inf.dext.
C
arter, M
agnell 2007
Tab. 1. Ten
tative kill season
 of m
am
m
als at V
lasac. D
ark shades represen
t the m
ost precise estim
ation
 of the season
 of capture, restricted to 1–2 m
on
ths. A
ll
elem
en
ts used for estim
ation
s of season
ality origin
ate from
 the faun
al assem
blage collected in
 2006–2009, apart from
 elem
en
ts m
arked w
ith an
 asterisk (*)
w
hich w
ere collected in
 1970–1971. **The agein
g of w
ild boar m
axillary teeth assum
in
g they follow
 the sam
e developm
en
t schedule as m
an
dibular teeth.
Vesna Dimitrijević, Ivana ?ivaljević and Sofija Stefanović
120
TA
X
O
N
IN
V
. N
O
.
ELEM
EN
T
W
IN
TER
SPR
IN
G
SU
M
M
ER
A
U
TU
M
N
Reference w
here the age
JA
N
FEB
M
A
R
A
PR
M
A
Y
JU
N
JU
L
A
U
G
SEP
O
C
T
N
O
V
D
EC
stage has been defined
\
unshed antler (8)
Schm
idt 1972
\
shed antler (8)
Schm
idt 1972
LV
 261\1-2
skull w
ith shed antlers
Schm
idt 1972
R
ed deer
LV
 10
74\5
m
axilla sin. (D
4, M
1)
H
aberm
ehl 1985
C
ervus elaphus
LV
 1296\1
m
axilla dext. (D
3-D
4, M
1) &
 sin. (D
4, M
1)
H
aberm
ehl 1985
LV
 273\4
m
axilla dext. (D
3, P2-M
3) &
 sin. (D
3, P2- M
2)
B
row
n, C
hapm
an 1991
LV
 bb-30
\2
m
andibula sin. (D
2-D
3, alv. D
4, M
1 bud)
C
arter 2006
LV
 bb-14\13
m
andibula sin. (D
2-D
4)
B
row
n, C
hapm
an 1991
R
oe deer
\
unshed antler
Schm
idt 1972
C
apreolus capreolus
\
shed antler
Schm
idt 1972
Tab. 2. Ten
tative kill season
 of m
am
m
als at Lepen
ski V
ir. D
ark shades represen
t the m
ost precise estim
ation
 of the season
 of capture, restricted to 1–2 m
on
ths.
TA
X
O
N
IN
V
. N
O
.
ELEM
EN
T
W
IN
TER
SPR
IN
G
SU
M
M
ER
A
U
TU
M
N
R
eference w
here the age
JA
N
FEB
M
A
R
A
PR
M
A
Y
JU
N
JU
L
A
U
G
SEP
O
C
T
N
O
V
D
EC
stage has been defined
R
ed deer
PA
D
 8.70
\279\1
unshed antler
Schm
idt 1972
C
ervus elaphus
\
shed antler (5)
Schm
idt 1972
R
oe deer
PA
D
 7.70
\140
\1
unshed antler
Schm
idt 1972
C
apreolus capreolus
PA
D
 8.69\80
\1 
shed antler
Schm
idt 1972
PA
D
 28.8\69\1
m
andibula sin. (alv. D
2-D
3, D
4, M
1)
C
arter 2006
Tab. 3. Ten
tative kill season
 of m
am
m
als at Padin
a.
Becoming sedentary| The seasonality of food resource exploitation in the Mesolithic-Neolithic Danube Gorges
121
TA
X
O
N
V
L 20
0
6–20
0
9
LV
 1968–1970
PA
 1968–1970
W
IN
TER
SPR
IN
G
SU
M
M
ER
A
U
TU
M
N
N
ISP
%
N
ISP
%
N
ISP
%
JA
N
FEB
M
A
R
A
PR
M
A
Y
JU
N
JU
L
A
U
G
SEP
O
C
T
N
O
V
D
EC
R
ussian sturgeon A
cipenser gueldenstaedtii
9
0
,1
54
6,4
13
0
,6
Ship sturgeon A
cipenser nudiventris
\
\
4
0
,5
\
\
Sterlet A
cipenser ruthenus
\
\
9
1,1
1
0
,0
Stellate sturgeon A
cipenser stellatus
5
0
,1
7
0
,8
16
0
,7
B
eluga H
uso huso
27
0
,3
83
9,9
72
3,1
B
lack Sea shad A
losa im
m
aculata
92
1,1
1
0
,1
\
\
C
om
m
on carp C
yprinus carpio
80
3
9,5
92
10
,9
20
8
8,9
B
ream
 A
bram
is bram
a
22
0
,3
\
\
\
\
D
anube bleak A
lburnus chalcoides
2
0
,0
\
\
\
\
A
sp A
spius aspius
8
0
,1
2
0
,2
\
\
B
elica Leucaspius delineatus
1
0
,0
\
\
\
\
C
hub Leuciscus cephalus
\
\
3
0
,4
\
\
Ide Leuciscus idus
4
0
,0
2
0
,2
\
\
R
azor fish Pelecus cultratus
21
0
,2
1
0
,1
\
\
V
yrezub R
utilus frisii
7320
86,4
257
30
,5
162
6,9
Pigo R
utilus pigus
2
0
,0
1
0
,1
\
\
R
oach R
utilus rutilus
1
0
,0
\
\
\
\
R
udd Scardinius erythrophthalm
us
7
0
,1
\
\
\
\
V
im
ba bream
 V
im
ba vim
ba
4
0
,0
1
0
,1
\
\
W
els catfish Silurus glanis
61
0
,7
168
20
,0
180
6
77,1
Pike Esox lucius
44
0
,5
1
0
,1
\
\
H
uchen H
ucho hucho
29
0
,3
154
18,3
65
2,8
B
lack Sea trout Salm
o trutta labrax
\
\
1
0
,1
\
\
Pikepearch Sander lucioperca
8
0
,1
1
0
,1
\
\
TO
TA
L
8470
10
0
842
10
0
2343
10
0
Tab. 4. Fish taxa iden
tified to the species level from
 V
lasac, Lepen
ski V
ir, an
d Padin
a expressed in
 N
ISP frequen
cies (after Ωivaljevi≤ in prep.) an
d their sea-
son
al availability on
 the basis of m
igration
/spaw
n
in
g season
s, feedin
g season
s an
d optim
al w
ater tem
perature. Light shades – availability; dark shades – peak
availability (after D
inu 2010.Tab. 2; Bartosiew
icz, Bonsall 2004.Tab. 7; Bartosiew
icz et al.2008.Tab. 6; Petrovi≤ 1998 [1941]; R
isti≤ 1977; Sim
onovi≤ 2001; K
ot-
telat, Freyhof 2007).
Vesna Dimitrijević, Ivana ?ivaljević and Sofija Stefanović
122
N
ISP
W
IN
TER
SPR
IN
G
SU
M
M
ER
A
U
TU
M
N
TA
X
O
N
V
LA
SA
C
PA
D
IN
A
L. V
IR
1970
–1971
20
0
6–20
0
9
1968–1970
1968–1970
JA
N
FEB
M
A
R
A
PR
M
A
Y
JU
N
JU
L
A
U
G
SEP
O
C
T
N
O
V
D
EC
B
lack-throated diver G
avia arctica
\
\
1
\
W
hite pelican Pelecanus cf. onocrotalus
3
\
1
\
C
orm
orant Phalacrocorax carbo
3
\
1
\
M
ute sw
an C
ygnus olor
\
\
1
\
G
reat w
hite egret Egretta alba
2
\
\
\
M
allard A
nas platyrhynchos
2
\
\
1
Teal A
nas crecca
2
\
\
\
G
rey-lag goose A
nser anser
\
\
2
\
Ferruginous (|) duck A
ythya nyroca
(|)
1
\
\
\
Shell-duck Tadorna tadorna
\
\
1
\
B
lack kite M
ilvus m
igrans
5 (4)
\
\
\
Im
perial eagle A
quila heliaca
6 (16)
2
\
\
W
hite-tailed eagle H
aliaeëtus albicilla
27 (0
)
\
8
\
Taw
ny ow
l Strix aluco
2 (1)
1
\
\
Jay G
arrulus glandarius
1
\
\
\
M
agpie Pica pica
1
\
\
\
R
aven C
orvus corax
4
1
2
\
R
ook C
orvus frugilegus
\
\
1
\
Fieldfare Turdus pilaris
\
2
\
\
TO
TA
L
59 (21)
6
18
1
Tab. 5. B
ird taxa iden
tified to the species level from
 V
lasac (n
um
bers in
 paren
thesis represen
t the revision
 of the preserved bird faun
al assem
blage an
alyzed
by Bökönyi 1978, after D
im
itrijevi≤
et al. subm
itted), Padin
a (after Clason 1980), an
d Lepen
ski Vir (after Bori≤, D
im
itrijevi≤ 2006) expressed in
 N
ISP frequen
-
cies,an
d their season
al availability on
 the basis of residen
cy an
d m
igration
 pattern
s (after Puzovi≤ 2000; R
a∏ajski, K
iss 2004; G
ruba≠
et al. 2013; Clem
ents
et al. 2014).