www.geologicacarpathica.sk
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
, FEBRUARY 2012, 63, 1, 83—94 doi: 10.2478/v10096-012-0006-6
Late Pleistocene voles (Arvicolinae, Rodentia) from the
Baranica Cave (Serbia)
KATARINA BOGIĆEVIĆ
1
, DRAŽENKO NENADIĆ
2
and DUŠAN MIHAILOVIĆ
3
1
Department of Paleontology, University of Belgrade-Faculty of Mining and Geology, Kamenička Str. 6, P.O. Box 227, 11000 Belgrade,
Serbia; k.bogicevic@rgf.bg.ac.rs
2
Department of Historical and Dynamic Geology, University of Belgrade, Faculty of Mining and Geology, Kamenička Str. 6,
P.O. Box 227, 11000 Belgrade, Serbia
3
Department of Archaeology, Faculty of Philosophy, University of Belgrade, Čika Ljubina Str. 18—20, Belgrade, Serbia
(Manuscript received January 9, 2011; accepted in revised form June 9, 2011)
Abstract: Baranica is a cave system situated in the south-eastern part of Serbia, four kilometers south to Knjaževac, on
the right bank of the Trgoviški Timok. The investigations in Baranica were conducted from 1994 to 1997 by the Faculty
of Philosophy from Belgrade and the National Museum of Knjaževac. Four geological layers of Quaternary age were
recovered. The abundance of remains of both large and small mammals was noticed in the early phase of the research.
In this paper, the remains of eight vole species are described: Arvicola terrestris (Linnaeus, 1758), Chionomys nivalis
(Martins, 1842), Microtus (Microtus) arvalis (Pallas, 1778) and Microtus (Microtus) agrestis (Linnaeus, 1761), Micro-
tus (Stenocranius) gregalis (Pallas, 1779), Microtus (Terricola) subterraneus (de Sélys-Longchamps, 1836),
Clethrionomys glareolus (Schreber, 1780) and Lagurus lagurus (Pallas, 1773). Among them, steppe and open area
inhabitants prevail. Based on the evolutionary level and dimensions of the Arvicola terrestris molars, as well as the
overall characteristics of the fauna, it was concluded that the deposits were formed in the last glacial period of the Late
Pleistocene. These conclusions are rather consistent with the absolute dating of large mammal bones (23.520 ± 110 B.P.
for Layer 2 and 35.780 ± 320 B.P. for Layer 4).
Key words: Quaternary, Late Pleistocene, Balkans, Serbia, arvicolids, rodents.
Introduction
The study of the remains of Pleistocene arvicolids and other
rodents began in Serbia only recently. In the last twenty
years, some faunas from several localities of Late Pleisto-
cene age have been described. The best known of them
comes from the Smolućka Cave (Dimitrijević 1985, 1991).
Rodent remains of Late Pleistocene age have also been
found in some other caves: the Vrelska Cave (Marković &
Pavlović 1991; Pavlović & Marković 1991), the Petnička
Cave (Dimitrijević 1994, 1997a), the Vasiljska, Popšička
and Prekonoška Caves (Dimitrijević 1997a), Risovača
(Rakovec 1965; Dimitrijević 1997a; Jovanović & Simić
1998), Pećurski Kamen (Malez & Salković 1988; Medved
1994; Dimitrijević 1997a) and the Mirilovska Cave
(Dimitrijević & Jovanović 2002).
Among these sites, the Baranica Cave is by far the richest
locality considering the number of mammalian species that
have been found in it. It is of particular interest because it is
the first locality in Serbia which has yielded large mammal
fauna of the Last Glacial Maximum (Argant & Dimitrijević
2007).
This paper is the first in a series aimed at describing the re-
mains of small mammals from this cave. They will provide
data on the distribution, migration and evolution of small
mammals on the territory of Serbia and hence fill the gap
which exists in the knowledge of the fauna from this part of
the Balkans and Europe.
The site
The Carpatho-Balkan mountan range stretches in the N—S
and NW—SE direction in eastern Serbia. This mountain range
consists of a whole series of isolated massifs, made mainly of
carbonate rocks – limestones and dolomites of Jurassic and
Early Cretaceous age (Stevanović 1994). The limestone mas-
sifs are usually separated from each other by Neogene basins,
older formations and volcanic rocks (Zlokolica Mandić 1998).
The scientific research of karst in the eastern Serbia began
at the end of the nineteenth century (Cvijić 1887), but even
before that time, karst forms (especially caves) were de-
scribed by various scientists and travellers (Ćalić 2007). De-
spite this early interest, karst regions of eastern Serbia are
much less known than those in the adjacent areas – the Di-
narides and the Alpides (Stevanović & Filipović 1994).
Since in eastern Serbia isolated limestone masses of small-
er area are dominant, the shallow karst of the contact type
has been developed. For this reason, the caves that have been
formed are shorter than in other karst regions of the world
(Djurović 1998).
As opposed to the typical development of karst topogra-
phy in the Dinarides, the karst regions in the Carpatho-Bal-
kanides lack surficial karst landforms (polje, sinkholes), and
usually have vegetational cover (Stevanović & Filipović
1994). Nevertheless, the territory of eastern Serbia is rich in
different underground karst forms – Petrović (1976) de-
scribed more than 130 larger caves.
84
BOGIĆEVIĆ, NENADIĆ and MIHAILOVIĆ
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2012, 63, 1, 83—94
In the vicinity of Baranica many other caves and rock shel-
ters are found (Vasiljska pećina, Pećina iznad Vrela, Bolvan I,
II, III and IV, Gabrovnica, Kožuvarska pećina, etc.). In some
of them the remains of Pleistocene fauna and Paleolithic arte-
facts were discovered (Mihailović 2004), and in Gabrovnica
even prehistoric cave paintings – the first of their kind in Ser-
bia (Mihailović & Jovanović 1997).
Baranica is a composite cave system situated in the south-
eastern part of Serbia, four kilometers south to Knjaževac,
on the right bank of the Trgoviški Timok (Fig. 1). Its altitude
is approximately 400 meters, 10 meters above the riverbed
(Mihailović 2004). It was formed in the so-called Urgonian
rocks of the Early Cretaceous age, represented by limestones,
bioclastic limestones, sandstones and marlstones. These lime-
Fig. 1. Geographical position of the Baranica Cave.
stones are whitish to grey, with numerous remains of orbito-
linids, crinoids, gastropods, corals, etc. (Jankičević 1978).
In this work, only the fossils from Baranica I are described,
but it should be mentioned that there are also some other parts
of this cave system – Baranica II, and also Baranica III, in
which some large mammal remains have been found.
Baranica is a dry karst cave, without a stream. It has two
entrances – the larger one to the south and the smaller one
to the east. The cave consists of a small semicircular cham-
ber (6.80 4.0 m) connected by narrow passages to the north
and west with the rooms I and II (Fig. 2). Room I is situated
on the north and its approximate dimensions are 3 2.4 m.
Room II is bigger and filled with sediment which was depos-
ited through the opening in the western part of the room
Fig. 2. a – Map of Baranica I with the
position of the trenches after Sladić & Jo-
vanović (1996). b – Section of the Upper
Pleistocene deposits in the Baranica Cave.
The section has been compiled from the
data obtained in all three trenches.
85
LATE PLEISTOCENE VOLES FROM THE BARANICA CAVE (SERBIA)
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2012, 63, 1, 83—94
(Sladić & Jovanović 1996). The total extension of Room II
is not known yet.
During the 1994—1995 excavations, three small trenches
were opened – trench A (2.5 1.5 m), generally oriented
east—west; trench B in the entrance of Room II, and trench C
in the southern entrance of the cave (Fig. 2a).
The trenches were dug to depths of between 2 and
2.5 meters, but the bedrock has not yet been reached
(Mihailović 2004).
The cave floor is covered with clastic sediments. The clastic
material consists of weathering detritus and breakdown (frag-
ments of broken bedrock of various sizes). Both autochtho-
nous and allochthonous sediments are present in this cave,
although their relative proportions are not known. It seems,
however, that in the entrance area the autochthonous compo-
nent is predominant (as showed by the presence of abundant
fragments of Cretaceous limestones and fossils), while in
Room II most of the sediment material came from the outside,
through the opening in the ceiling (Sladić & Jovanović 1996).
Also, there are no typical stream deposits, such as gravels
and sands, in this cave. In the interior, the amount of rock de-
bris is small, and sediments are more compact, while in the en-
trance sediments are less compact and contain more rock
fragments and blocks.
The excavations revealed the profile of the sediments com-
posed of four sub-horizontal geological layers, some of which
have been disturbed by post-depositional processes, including
modern human activities (treasure hunters digging). The main
criteria for distinguishing the layers were the colours of the
sediments and amount of rock fragments. The sequence of de-
posits, from the top to the bottom, is as follows (Fig. 2b):
1. Surficial clay of Holocene age (approximately 30 cm
thick). Fragments of prehistoric and Roman pottery were
found in this layer, which indicate the Holocene age.
2. Yellowish clay (80 cm thick) – with coarse rock frag-
ments and the remains of large and small mammals. Two hori-
zons were distinguished in this unit, 2a and 2b, which differ
mainly in colour (2b being somewhat darker). In this layer nu-
merous remains of large mammals were found – bison,
horse, rhinoceros and other species. Some Upper Paleolithic
artefacts (two flint blades and a small endscraper) and a fire-
place were also found in the upper part of this layer, indicating
that Paleolithic man made short visits to this cave (Mihailović
et al. 1997).
3. Ash grey homogeneous silt (20 cm thick) – with rare
small mammal remains. It underlies Layer 2 in the northern
part of trench A. The upper part (3a) contains rare small
mammal remains, while the lower one (3b) is paleontologi-
cally sterile.
4. Dark brown clayey silt (more than 120 cm thick) – with
coarse rock fragments. In this sediment, two horizons can also
be differentiated:
4a – compact sediment with rock fragments;
4b – homogeneous sediment without rock fragments.
In this layer, remains of both large and small mammals are
scarcer than in Layer 2. Some Paleolithic artefacts were also
found (Mihailović et al. 1997).
The investigations in Baranica I began in 1994 and have
been conducted jointly by the Faculty of Philosophy from Bel-
grade and the National Museum of Knjaževac. They were part
of a research project on the prehistory of the Knjaževac area,
which includes trial excavations in several caves in the vicini-
ty of this site.
From 1995 to 1997, some Upper Paleolithic artefacts were
found in Baranica I (Mihailović et al. 1997) and some large
mammal remains in both Baranica I and II (Dimitrijević
1997b, 1998). Samples for analysis of arvicolid (and other
small mammal) remains were taken only from Baranica I.
The abundance of remains of both large and small mammals
was already noticed in the early phase of the research. In addi-
tion to the mammal remains, this cave also yielded some re-
mains of other animals and plants (birds, reptiles, amphibians,
fish, gastropods and seeds; Bogićević 2008). Some pollen
grains were also extracted from hyena’s coprolites (Argant &
Dimitrijević 2007). General remarks on the fossil fauna from
Baranica with the preliminary faunal list were published in
several papers (Dimitrijević 1997b, 1998, 2004). Although
some species and groups have been described more thorough-
ly (subterranean voles in Brunet-Lecomte et al. 2001, horses
in Forsten & Dimitrijević 2004, carnivores in Salčin 1996),
the fauna from Baranica (especially small mammals) is still
poorly known. Only two unpublished master theses (Krantić
1997; Jovanović 2005), and one conference abstract
(Bogićević 2004) contain a partial analysis of the rodent fauna
from Baranica.
Material and methods
During the 1995 excavation, some sediment samples were
taken at 22 sampling points in Baranica I: six in Layer 2, one
in Layer 3 and 15 in Layer 4. (Layer 3 is represented by only
one sample, because it was thought to be without fossils.) All
samples were screen-washed on three screens of 2, 1 and
0.5 mm mesh. The sample size was not large, approximately
50 kg of sediments in total. Only a few teeth were found in the
sample from Layer 3, but Layers 2 and 4 have proved to be
rather rich in vertebrate remains. More than four hundred ar-
vicoline teeth were identified to the species level and are in-
cluded in this study.
The material is fairly well preserved, with almost no traces
of dissolution. All skeletal elements are present, but the fau-
nal list is based exclusively on teeth (mainly M
1
). All the
teeth that could be determined to the level of species are list-
ed, but only M
1
were measured. The dimensions are given in
mm. The method of measurements is described in Nada-
chowski (1982). The SDQ index (the enamel band
(Schmelzband) differentiation quotient) was measured by
the method originally proposed by Heinrich (1978). For the
morphological elements of arvicolid teeth, the terminology
of van der Meulen (1973) and Nadachowski (1982) is em-
ployed. The names of the morphotypes of M
1
are after Nada-
chowski (1982, 1984a, 1985). The teeth were drawn under a
binocular microscope.
The minimum numbers of individuals (MNI) were calculat-
ed according to the number of M
1
in most cases, but also
based on the numbers of some other teeth when they were
characteristic enough for species determination. These num-
86
BOGIĆEVIĆ, NENADIĆ and MIHAILOVIĆ
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2012, 63, 1, 83—94
bers were used for calculating the percentage of species in
the arvicoline fauna.
The material described in this paper is stored at the Depart-
ment of Paleontology of Belgrade University, under the inven-
tory numbers BAR 2/6—15, BAR 3/2—5 and BAR 4/12—25.
The data on the Pleistocene distribution of particular arvicolid
species in Europe were taken mainly from Kowalski (2001).
All of the material described in this paper comes from
Baranica I, hence, for the sake of simplicity, the name “Bara-
nica” will be consistently used instead of “Baranica I”.
Systematic paleontology
Order: Rodentia Bowdich, 1821
Family: Cricetidae Fischer, 1817
Subfamily: Arvicolinae Gray, 1821
Arvicola
Lacép
e
de, 1799
Arvicola terrestris (Linnaeus, 1758)
(Fig. 3)
M a t e r i a l a n d d i m e n s i o n s: Layer 2: M
1
sin.,
M
1
dext., M
2
sin., M
2
dext., M
3
sin., M
3
dext., M
1
dext. (L =
4.1 mm; A = 1.72 mm), M
2
sin., M
2
dext. BAR 2/6.
Layer 3: M
1
sin., M
2
sin. BAR 3/2.
Layer 4: 2 M
1
dext., M
2
sin., 2 M
3
sin., M
3
dext., 2 M
1
sin.
(L = 4.03; 4.07 mm; A = 1.62; 1.69), 2 M
1
dext. (L = 4.07; -;
A = 1.69; 1.48 mm), 2 M
2
sin., M
2
dext., M
3
dext. BAR
4/12—16.
D e s c r i p t i o n: The molars are large, rootless, with abun-
dant crown cement in re-entrant angles (Fig. 3). The enamel is
thicker on the concave side of the triangles (“positive” or Arvi-
cola-differentiation). On the first lower molars, the anterior
cap is broadly confluent with T4 and T5. All the third upper
molars show a more complicated shape – so-called “exitus”
morphotypes (Nadachowski 1982) – with three dentine trian-
gles, more or less closed.
C o m m e n t s: The enamel thickness quotient (SDQ index
after Heinrich 1978) of M
1
is 90.11 in Layer 2 (only one
tooth measured) and 92.48, with a range of 89.37—94.66
(n = 3) in Layer 4. Lower values of the index (less than 100),
such as these, are typical for geologically younger (Late
Glacial and Holocene) and recent populations of Arvicola
terrestris (Heinrich 1978, 1987). The dimensions of M
1
(average length greater than 4 mm) are also in agreement
with the proposed young age of the populations from Baranica
(the older populations had a somewhat smaller M
1
; Maul et
al. 2000).
Arvicola terrestris was among the most common rodent
species in the Late Pleistocene of Europe. Populations of this
species have been found (although usually represented by a
small number of specimens) throughout Europe, both in
warm and cold periods of the last glacial (Kowalski 2001;
Cuenca-Bescós et al. 2008).
It was also found in the Pleistocene deposits of several
caves in Serbia – the Smolućka and Vrelska Cave (Dimitri-
jević 1997a) and in Montenegro – Crvena Stijena (Malez
1975) and Mališina Stijena (Bogićević & Dimitrijević 2004).
Layer 2
Layer 4
n min. max. median n min. max. mean SD
L
3 2.88 3.3 2.97 16 2.46 3.5 2.84 0.25
a
5 1.32 1.7 1.38 20 1.05 1.85 1.37 0.18
a/L*100 3 46
52 46
16 42
53
48
2.98
B
1
/W
1
7 4
23 11
18 3
21
8.37 5.22
Table 1: Dimensions of M
1
of Chionomys nivalis.
Fig. 3. Arvicola terrestris (Linnaeus, 1758). a – M
1
dext. (BAR 4/14),
b – M
3
sin. (morphotype C), c – M
3
dext. (morphotype D), d – M
3
sin. (morphotype E). (Names of morphotypes after Nadachowski
1984a.)
Fig. 4. Chionomys nivalis (Martins, 1842). a – M
1
dext. (morpho-
type B – “gud”), b – M
1
dext. (morphotype C), c – M
1
dext.
(morphotype D
1
), d – M
1
sin. (morphotype D
2
). (Names of mor-
photypes after Nadachowski 1984a.)
Chionomys Miller, 1902
Chionomys nivalis (Martins, 1842)
(Fig. 4)
M a t e r i a l: Layer 2: 5 M
1
sin., 2 M
1
dext. BAR 2/7.
Layer 4: 12 M
1
sin., 11 M
1
dext. BAR 4/17. (For dimen-
sions, see Table 1.)
D e s c r i p t i o n: The dentine triangles are massive. The tri-
angles T1—T4 on M
1
are closed, while T5 can sometimes be
confluent with the anterior lobe. However, most of the M
1
from Baranica (73.3 %) have five closed triangles and the an-
terior part of the tooth is in the shape of an arrow or a spear
(this is a progressive or “nivalid” morphotype; Fig. 4c—d). The
second most frequent is the transitional morphotype (“nivalid-
ratticeps” after Nadachowski 1984a; Fig. 4b) with 23.3 %,
è
87
LATE PLEISTOCENE VOLES FROM THE BARANICA CAVE (SERBIA)
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2012, 63, 1, 83—94
while only one tooth showed a so-called “gud” morphotype,
with broadly confluent T5 and T6 and a well developed BRA4
and BSA4 (Fig. 4a). No tooth showed a structure characteris-
tic for Microtus oeconomus (Pallas, 1776) (“ratticeps“
morphotype after Nadachowski 1984a).
C o m m e n t s: The small values of the B
1
/W
1
ratio clearly
indicate that the teeth from Baranica belong to Chionomys niva-
lis and not to Microtus oeconomus. In the latter species, this ra-
tio is much larger; meaning that the triangle T5 is less separated
from the anterior loop. The values of this ratio measured in C. ni-
valis from Bacho Kiro, Bulgaria (Nadachowski 1984a) varied
from 1 to 24 (3—23 in Baranica), while the mean values in differ-
ent layers are 7—11; in Baranica 11 in Layer 2 and 8 in Layer 4.
In recent populations of this species in Serbia, the morpho-
type frequency distribution is very similar to that observed in
the Pleistocene deposits of Baranica. For example, in a re-
cent population from the Stara Planina Mountain (an area of
similar altitude and geographically very close to Baranica),
66.8 % of the teeth showed the “nivalid” morphotype,
16.6 % “nivalid-ratticeps”, while the “gud” morphotype was
not present at all (Kryštufek 1990).
This species inhabited mountainous areas of Europe in the
Middle and Late Pleistocene (Kowalski 2001). During the
latter period, the snow vole was widespread mostly in the
northern part of the Balkan Peninsula (Terzea 1972; Rabeder
2004; Toškan & Kryštufek 2007; Toškan 2009), but their re-
mains have also been found in some caves and shelters of the
southern Balkans – in Montenegro (Crvena Stijena, Mališina
Stijena, Trebački Krš – Malez 1975; Dimitrijević 1999;
Bogićević & Dimitrijević 2004) and Serbia (the Vrelska and
Smolućka Cave) (Dimitrijević 1997a). These facts are not in
accordance with the conclusions of some authors (Terzea
1972; Toškan & Kryštufek 2007) that during the Late Pleis-
tocene, this species was distributed only to the north of the
Danube and Sava Rivers. It is assumed that Chionomys niva-
lis migrated southwards during the Holocene and replaced
the species Dinaromys bogdanovi (Martino, 1922) that had
similar ecological preferences (Kryštufek 2004).
Microtus Schrank, 1798
Microtus (Microtus) arvalis (Pallas, 1778) and
M. (Microtus) agrestis (Linnaeus, 1761)
(Fig. 5)
M a t e r i a l: Layer 2: 21 M
1
sin., 17 M
1
dext., M
2
dext. BAR
2/8—9.
Layer 3: 2 M
1
sin., 3 M
1
dext. BAR 3/3.
Layer 4: 61 M
1
sin., 60 M
1
dext., 4 M
2
sin., 4 M
2
dext. BAR
4/18—20. (For dimensions: see Table 2.)
D e s c r i p t i o n: The typical “arvalid” morphotypes
(Fig. 5a—b; the names of morphotypes after Nadachowski
Table 2: Dimensions of M
1
of Microtus arvalis/agrestis.
Layer 2
Layer 3
Layer 4
n min. max. mean SD n min. max. median n min. max. mean SD
L
22 2.45 2.97 2.8
0.15
3 2.9 3.14 3
64 2.41 3.31 2.87
0.22
a
22 1.31 1.66 1.53
0.09
3 1.59 1.66 1.62
64 1.24 1.90 1.53
0.14
a/L*100
22 51.8 56.7 54.6
1.14
3 51.6 55.3 54.8
64 50.5 57.4 53.3
1.43
(LT4/LT5)*100
22 56.2 82.6 71.84
6.41
3 65.6 71.4 68.7
64 51.60 84.00 67.50
6.91
Fig. 5. Microtus (Microtus) arvalis (Pallas, 1778) & M. (M.) agrestis (Linnaeus, 1761). a – M
1
sin. (morphotype C), b – M
1
dext. (mor-
photype D), c – M
1
sin. (morphotype F), d – M
1
dext. (morphotype G – “maskii”), e – M
1
sin. (morphotype I – “extratriangulatus“
after Nadachowski 1985), f – M
2
dext. (with exsul-loop). (Names of morphotypes after Nadachowski 1984a).
88
BOGIĆEVIĆ, NENADIĆ and MIHAILOVIĆ
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2012, 63, 1, 83—94
1982, 1985) are the most abundant in all layers (89 % in
Layer 2; 100 % in Layer 3; 87 % in Layer 4). The other mor-
photypes are much less common: “maskii” (Fig. 5d) –
5.4 % in Layer 2; 4 % in Layer 4; the morphotype with an
additional lingual syncline – LRA6 (Fig. 5c) – 5.4 % in
Layer 2; 5 % in Layer 4; and the “extratriangulatus” mor-
photype (Fig. 5e) – 4 % in Layer 4. The average length of
M
1
is 2.8 mm in Layer 2 and 2.87 mm in Layer 4.
Only 9 out of 166 M
2
that were found in Baranica have a
well-developed additional triangle (1 in Layer 2 and 8 in
Layer 4).
C o m m e n t s: It is almost impossible to distinguish these
two species on the basis of isolated teeth. The only unambigu-
ous difference between them is the presence of an additional
medial triangle on the posterior side of M
2
(Fig. 5f), which is
well developed only in M. agrestis (Zimmermann 1956).
The M
1
of M. arvalis are somewhat smaller and in recent
specimens, the length ranges from 2.4 to 2.9 mm, while in M.
agrestis, this value varies from 2.75 to 3.4 mm; with mean
values about 2.6 and 2.9 mm, respectively (Nadachowski
1982, 1984a; Gromov & Polyakov 1992). In Layer 2, there
were no specimens longer than 3 mm and the percentage of
small specimens was larger than in Layer 4. However, the
finding of a M
2
with a well-developed additional triangle
proves the presence of the species M. agrestis in Layer 2. On
the other hand, large teeth are quite numerous in Layer 4
(about 31.8 %).
It is sometimes considered that these species can be distin-
guished on the basis of the symmetry of the lingual and labial
dentine triangles. These triangles are more symmetrical in M.
arvalis, and rather asymmetrical in M. agrestis (Chaline
1972). The measure of this symmetry/asymmetry is an index
that is obtained by dividing the lengths of the triangles T4 and
T5 (Nadachowski 1984b). The best results in distinguishing
between the two species could be obtained by plotting the
length of M
1
against the LT4/LT5 index (Nadachowski
1984b). Values of this index lower than 65 (typical for M.
agrestis) were documented in 13.6 % of the teeth from Lay-
er 2 and in 32.8 % from Layer 4.
Considering all these data, it could be concluded that both
species were present in the material. However, M. agrestis is
relatively abundant in Layer 4, while it almost disappears in
Layer 2.
It could also be observed that the complexity of the antero-
conid (A/L ratio) gradually increases. In Layer 4, this ratio
was 53.3; in Layer 3 (measured on only 3 specimens) 53.9,
and in Layer 2–54.6. The value in Layer 4 is similar to that in
the Italian locality Castelcivita, with an estimated age of ap-
proximately 30,000—40,000 years (Maul et al. 1998), and in
Croatian localities (Vindija – Layer G, Marlera I, Mujina
Cave), with an age of approximately 27,000—45,000 years
(Mauch Lenardić 2007). The ratio in Layer 2 is very close to
the ratio in recent populations of these species (Maul et al.
1998). Small values of A/L ( < 52) are typical for the species
M. ‘arvalinus’, which is the probable ancestor of M. arvalis
(Maul & Parfitt 2009).
Only three M
1
with the morphological characteristics of
these species were found in Layer 3, but all of them are rather
large, so they probably belonged to M. agrestis. This is inter-
esting, because in most Balkan localities of the Late Pleistocene
age, M. arvalis is much more common. In the Bacho Kiro
Cave, M. agrestis was more numerous than M. arvalis only in
two periods: at the beginning of the last glacial and in the mid-
dle of this period (Nadachowski 1984a). After the latter maxi-
mum (at the beginning of the “Pleniglacial II”), this species
totally disappeared from the studied area. At present, M.
agrestis occurs neither in Bulgaria (Nadachowski 1984a), nor
in Serbia south of the Sava and Danube (Kryštufek et al.
1989). In Baranica, the “agrestis-peak” in Layer 3 is followed
by an almost total disappearance of this species in Layer 2.
Since these caves are situated in the same geographical area
(the Stara Planina Mountain Range), the comparison is justi-
fied. Hence, for Layer 2, an age slightly older than 20,000 years
B.P. could be proposed (in that period, M. agrestis disappeared
from the area of the Bacho Kiro Cave) and for Layer 3, older
than 30,000 years (second “agrestis-peak” in Bacho Kiro).
Microtus arvalis and M. agrestis are among the most com-
mon of all Pleistocene rodents in Serbia and Montenegro.
Their remains have been found in the Smolućka, Vrelska,
Vasiljska and Petnička Cave in Serbia (Dimitrijević 1997a), as
well as in Mališina Stijena (Bogićević & Dimitrijević 2004)
and Trebački krš (Dimitrijević 1999) in Montenegro.
Microtus (Stenocranius) gregalis (Pallas, 1779)
(Fig. 6)
M a t e r i a l: Layer 2: 4 M
1
sin., 3 M
1
dext. BAR 2/10. (For
dimensions: see Table 3.)
D e s c r i p t i o n: The first lower molar of this species has a
poorly developed (“gregalo-arvalid”) or completely absent
BSA4 (“gregalid” morphotype). It differs from the similar
(and probably ancestral) species M. gregaloides (Hinton,
Fig. 6. Microtus (Stenocranius) gregalis (Pallas, 1779). a – M
1
sin., b – M
1
dext. (a, b – morphotype A – “gregalid”), c – M
1
sin., d – M
1
dext. (c, d – morphotype B – “gregalo-arvalid”).
(Names of morphotypes after Nadachowski 1984a.)
Table 3: Dimensions of M
1
of Microtus gregalis (Layer 2).
Layer 2
n min.
max.
mean
SD
L
6 2.43 3.00 2.78 0.21
a
6 1.23 1.53 1.45 0.11
a/L*100
6
51
54
52.5
1.05
89
LATE PLEISTOCENE VOLES FROM THE BARANICA CAVE (SERBIA)
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2012, 63, 1, 83—94
Layer 2
Layer 4
n min. max. mean SD n min. max. mean SD
L
9 2.25 2.58 2.46 0.11 18 2.31 2.70 2.53 0.11
a
9 1.17 1.35 1.26 0.06 18 1.17 1.47 1.31 0.09
a/L*100 9 49
52
51.2 0.97 18 49
54
51.6
1.69
Table 4: Dimensions of M
1
of Microtus (Terricola) subterraneus.
Fig. 7. Microtus (Terricola) subterraneus (de Sélys-Longchamps,
1836). a – M
1
dext. (morphotype A), b – M
1
sin. (morphotype
B – “maskii”), c – M
1
dext. (morphotype C), d – M
1
dext. (mor-
photype with broad connection between T1 and T2). (Names of
morphotypes after Nadachowski 1984a.)
1923) in having a longer anteroconid complex ( > 52) and T5
separated from T6 (Maul & Parfitt 2009).
C o m m e n t s: The transitional “gregalo-arvalid” morpho-
type is sometimes, although very rarely, found in M. arvalis
and a little more often in M. agrestis, but the “pure” “gregalid”
morphotype is characteristic for M. gregalis (Nadachowski
1982). In the Late Pleistocene populations of M. arvalis/
agrestis from Serbia, the transitional morphotype has only
been found in a few localities (Smolućka and Hadži Prodano-
va Cave) and always with a very low frequency (Dimitrijević
1991; Bogićević 2008).
The oldest record of this species comes from the Early
Pleistocene deposits of Western and Central Europe (Maul
1990). Microtus gregalis is a steppe and tundra dweller and
it was very abundant and widespread in Europe during the
cold episodes of the Late Pleistocene (Nadachowski 1982),
even in Spain (Pokines 1998; Sesé 2005; Cuenca-Bescós et
al. 2008). It has also been found in some other localities in
the Balkans (Kozarnika and Cave 16 in Bulgaria – Popov
2000; Popov & Marinska 2007) but this is the first time it
has been found in Serbia.
Microtus (Terricola) subterraneus
(de Sélys-Longschamps, 1836)
(Fig. 7)
M a t e r i a l: Layer 2: 3 M
1
sin., 7 M
1
dext. BAR 2/11.
Layer 3: 3 M
1
dext. BAR 3/4.
Layer 4: 12 M
1
sin., 11 M
1
dext. BAR 4/21. (For dimen-
sions: see Table 4.)
D e s c r i p t i o n: The first lower molar of Microtus (Terri-
cola) subterraneus (Fig. 7) is characterized by the presence
of a so-called “Pitymys-rhombus” (broadly connected trian-
gles T4 and T5) and by its small dimensions. The relative
length of the anteroconid (A/L ratio) is about 51, in both
Layers 2 and 4.
C o m m e n t s: The morphology of the M
1
is similar to those
of the recent species Microtus subterraneus and Microtus
multiplex (Fatio, 1905), but according to the length of M
1
(mean value 2.46 mm in Layer 2 and 2.53 mm in Layer 4),
the material from Baranica is more similar to the smaller
species – Microtus subterraneus.
A detailed morphometrical analysis based on recent and
fossil material from Serbia and Montenegro (Brunet-
Lecomte et al. 2001) showed that the populations from Ba-
ranica have an intermediate taxonomic position between M.
(T.) grafi Brunet-Lecomte, Nadachowski & Chaline, 1992
and M. (T.) brauneri (Martino, 1926). The systematic posi-
tion of M. (T.) grafi is not yet clarified, hence it could be
treated either as a subspecies of M. (T.) subterraneus (more
probable) or as a distinct, but very closely related species
(Brunet-Lecomte et al. 2001).
This species was found at several Late Pleistocene locali-
ties in Serbia – the Smolućka and Vrelska Cave (Dimitrijević
1997a), the Vasiljska Cave and Pećurski kamen (Brunet-
Lecomte et al. 2001). The only known Pleistocene locality of
this species in Montenegro is Trebački krš (Dimitrijević
1999). It was also found in Holocene deposits of the Vruća
Cave in Montenegro (Brunet-Lecomte et al. 2001).
Clethrionomys Tilesius, 1850
Clethrionomys glareolus (Schreber, 1780)
(Fig. 8)
M a t e r i a l: Layer 2: M
1
sin., 2 M
1
dext., 3 M
2
sin., M
3
dext., 3 M
1
sin., 2 M
1
dext., 2 M
2
sin. BAR 2/12—13.
Layer 4: 7 M
1
sin., 4 M
1
dext., M
2
sin., 6 M
2
dext., 5 M
3
sin., 8 M
3
dext., 4 M
1
sin., 4 M
1
dext., 4 M
2
sin., M
2
dext.,
M
3
sin., 2 M
3
dext. BAR 4/22—23. (For dimensions: see
Table 5.)
D e s c r i p t i o n: Not only the first lower molars, but also
the other teeth of this species could be distinguished by the
presence of roots and a characteristic rounded shape of the
dentine triangles. All the molars have two roots. The enamel
is thinner on the convex sides of the triangles and thicker on
the concave ones. Only in one M
1
were T1 and T2 completely
closed. In all the other specimens, there was a more or less
broad connection between these triangles (Fig. 8).
C o m m e n t s: All the M
1
from Baranica have two roots, a
characteristic which is typical (after Radulescu & Samson
1992) of populations from the last glacial period and the
Holocene, while some of the earlier representatives of the
species could have three roots. The dimensions of the M
1
from Baranica (Table 5) are very similar to those of Cave 16
in Bulgaria (Popov 2000).
Several localities of Late Pleistocene age in Serbia yielded
remains of this species: the Smolućka, Vrelska, Vasiljska,
Petnička, Hadži Prodanova and Canetova Cave (Dimitrijević
1997a; Bogićević 2008). In Montenegro this species was
found only in Trebački krš (Dimitrijević 1999).
90
BOGIĆEVIĆ, NENADIĆ and MIHAILOVIĆ
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2012, 63, 1, 83—94
Lagurus Gloger, 1841
Lagurus lagurus (Pallas, 1773)
(Fig. 9)
M a t e r i a l: Layer 2: 7 M
1
sin., 7 M
1
dext., 5 M
2
sin., 3 M
2
dext., 4 M
3
sin., 2 M
3
dext., 4 M
1
sin., M
1
dext., 2 M
2
sin., M
2
dext., 5 M
3
sin., 5 M
3
dext. BAR 2/14—15.
Layer 3: M
1
sin., 2 M
1
dext., M
1
sin., M
1
dext., M
3
dext.
BAR 3/5.
Layer 4: 3 M
1
sin., M
1
dext., 2 M
2
sin., M
2
dext., M
3
sin.,
M
3
dext., mand. sin. (M
1
—M
2
), M
1
sin., 4 M
1
dext., M
2
sin., 4
M
2
dext., 4 M
3
sin., M
3
dext. BAR 4/24—25. (For dimensions:
see Table 6.)
D e s c r i p t i o n: The molars of Lagurus lagurus are charac-
terized by a complete lack of cement in the synclines. The lin-
gual and labial triangles are of approximately the same size
(Fig. 9). There is a small interruption of the enamel on BSA4
of the M
1
. The upper teeth of this species have a “laguroid
protuberance” (small additional lingual triangle).
C o m m e n t s: The “lagurid” morphotype (Fig. 9b; morpho-
type C after Nadachowski 1984a) dominates – it was devel-
oped in 77 % of the M
1
, while only one tooth shows a
primitive “transienid” morphotype (Fig. 9a). It is also typical
for the Late Pleistocene populations of this species from Bul-
garia – it has been found in about 70 % M
1
in the Cave 16
(Popov 2000) and 84 % in Bacho Kiro (Nadachowski 1984a).
Remains of this steppe species were found in Serbia only in
the Pleistocene deposits of the Vrelska Cave (Pavlović &
Marković 1991).
Discussion
Paleoenvironment
A more detailed interpretation of paleocological condi-
tions will be given in a subsequent paper, when the analysis
of the whole material from the Baranica Cave has been com-
pleted. In this paper, only some preliminary conclusions,
drawn on the basis of the presence of arvicoline species and
the previous knowledge of fauna and flora (large mammals,
pollen), will be presented.
Remains of large mammals have been found in both
Baranica I and II. The composition of fauna in these two caves
is very similar. It could not be determined in which layers the
remains of particular species were found (a complete descrip-
tion of the large mammal fauna has not yet been published),
but, according to some earlier reports (Mihailović et al. 1997),
it could be concluded that most of them come from Layer 2.
The large mammal fauna contains some species indicative of
cold climate, such as Gulo gulo (Linnaeus, 1758) and Coelo-
donta antiquitatis (Blumenbach, 1799), while species indica-
tive of warm climate, such as roe deer and wild boar are
completely absent (Argant & Dimitrijević 2007).
Several pollen grains found in a hyena’s coprolite from
Baranica indicate an open landscape with the presence of
steppe taxa (Argant & Dimitrijević 2007).
Since extant arvicolines have a distribution that is mainly
controlled by the climate, a method has been developed to
use the number of arvicoline species for an estimation of
some climatic parameters, notably temperature – a greater
number of species is typical for places with a colder climate
(Montuire 1996; Montuire et al. 1997). This method was ap-
plied to the faunas from some Hungarian and Central Euro-
pean localities of Pleistocene age. The number of arvicoline
species in Baranica (8 species in Layer 2 and 7 in Layer 4)
Table 5: Dimensions of M
1
of Clethrionomys glareolus.
Layer 2
Layer 4
n min. max. median n min. max. mean SD
L
3 2.25 2.49 2.4
6 2.13 2.58 2.36 0.15
a
3 0.87 1.02 0.99
6 0.84 1.08 0.98 0.09
a/L*100 3 39
42
40
6 39
44
41.67 1.75
Fig. 8. Clethrionomys glareolus (Schreber, 1780). a – M
1
sin.
(morphotype A), b – M
1
dext. (morphotype B), c – M
1
dext.
(morphotype B/C), d – M
1
sin. (morphotype C – “maskii”).
(Names of morphotypes after Nadachowski 1984a.)
Table 6: Dimensions of M
1
of Lagurus lagurus.
Layer 2
Layer 4
n min.
max. n min.
max.
L
2
2.46 2.55
2
2.46 2.67
a
2
1.26 1.38
2
1.32 1.47
a/L*100
2
51
54
2
54
55
Fig. 9. Lagurus lagurus (Pallas, 1773). a – M
1
sin. (morphotype
B), b – M
1
dext. (morphotype C), c – M
1
sin. (morphotype D).
(Names of morphotypes after Nadachowski 1984a.)
91
LATE PLEISTOCENE VOLES FROM THE BARANICA CAVE (SERBIA)
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2012, 63, 1, 83—94
corresponds to some “cold” localities of Central Europe,
with mean annual temperatures of —2.4 °C and 0.4 °C, re-
spectively (Montuire 1996). According to this method, the
mean temperatures for July was 14.7 °C in Layer 2 and
16.3 °C in Layer 4, while the mean temperatures for Decem-
ber were —19.9 °C and —15.8 °C, respectively. Although we
think that these results should be accepted with caution, they
seem to indicate that the deposits were formed under rather
cold conditions.
It should be mentioned that the estimates for Layer 2 are
similar to the temperatures, obtained by some other methods,
during the Last Glacial Maximum in this part of Europe
(“vole thermometer” method – Kordos 1987; BIOME3 ter-
restrial biosphere model – Guiot et al. 2000; species distri-
bution modelling – Fl
o
jgaard et al. 2009, etc.).
Among all the micromammal remains at the Baranica
Cave most of them belong to arvicolines (Bogićević et al.
2011). The most abundant among the vole species are
Microtus arvalis and M. agrestis (Table 7). They comprise
43.7 % of the arvicoline remains in Layer 2, 33.3 % in Layer 3
and even 61.6 % in Layer 4. Lagurus lagurus and Microtus
subterraneus are also very common. The compositions of
the arvicoline fauna in Layers 2 and 4 are generally very
similar, however, some differences should be emphasized,
such as the somewhat lower percentage of Lagurus lagurus
in Layer 4 and the presence of a boreal species – Microtus
gregalis – in Layer 2.
In all three layers, inhabitants of open grasslands and rela-
tively dry areas are predominant (Microtus arvalis and M.
agrestis, Lagurus lagurus). Since Lagurus lagurus, Microtus
gregalis and Microtus agrestis no longer live in the vicinity
of the cave, it can be assumed that the climatic conditions
were much more dry, cold and steppe-like than today. The
climate was particularly cold during the formation of Lay-
er 2. Such conclusions fit well with the results of palynologi-
cal analysis (Argant & Dimitrijević 2007).
As is the case with large mammals, forms that indicate
woodland conditions are rare.
Age of the fauna
As early as during the preliminary investigations
(Mihailović et al. 1997), it was established that the deposits
of Baranica are of Late Pleistocene age. A preliminary study
of the large mammal fauna and cultural stage of the archaeo-
logical artefacts indicated an age between 15,000 and 40,000
years for Layer 2 (Mihailović et al. 1997).
There were no extinct forms among the arvicolines. Nev-
ertheless, the fauna differs considerably from the recent one
by containing species that are no longer present in the extant
fauna of Serbia (Lagurus lagurus, Microtus gregalis), or in
the vicinity of the cave (Microtus agrestis).
We could conclude from the composition and ecological
characteristics of the fauna (the presence of boreal species of
both large and small mammals), that both Layers 2 and 4
were formed under conditions of a cold and dry climate,
characteristic of a glacial period. Layer 2 was formed during
a very cold period, which could correspond to the Last Glacial
Maximum, about 20,000 years ago.
The evolutionary stage of the Arvicola enamel (the mean
value of the SDQ index is well under 100), rather complicat-
ed anteroconids in Microtus arvalis and M. agrestis and the
predominance of progressive morphotypes in Lagurus lagu-
rus in Layer 4 are also typical for the Last Glacial. The SDQ
index (92.48) in this Layer is slightly higher than in the Cen-
tral European localities of Kemathenhöhle (with an estimat-
ed age of about 31,400 years B.P.; value of SDQ index –
89.23), Peskö (estimated age – 34,600 B.P.; SDQ index –
89.31) and Istállóskö (estimated age – 36,400 B.P.; SDQ
index – 89.54); and much lower than in Burgtonna (esti-
mated age – 80,000 B.P.; SDQ index – 99.65) (Heinrich
1987). Hence, it can be supposed that the age might be older
than in the former localities, but much younger than in the
latter one. However, since the extant populations of the spe-
cies in southern Europe were shown to have higher values of
the SDQ index than the northern ones (Röttger 1987), Lay-
er 4 might be of similar or slightly greater age than the
above-mentioned localities (Kemathenhöhle, Peskö, and
Istállóskö).
The greater size of M
1
of Arvicola terrestris is also in
agreement with the proposed age. A gradual increase in
length of this tooth was already well documented in many
European populations (Maul et al. 2000). According to
Heinrich (1987), an average length of M
1
greater than 4 mm
is characteristic for populations from the Late Weichselian.
Lately, some bone remains from Baranica have been abso-
lutely dated at the Oxford laboratory by AMS method: a sec-
ond phalanx of a giant deer (BAR 97/19/16; OxA-13827)
from Layer 2 in Baranica I has been dated to 23,520±110 B.P.
(
13
C —19.415 ‰), while a third molar of a cave bear (BAR
97/80/1; OxA-13828) from Layer 4 has been dated
to 35,780 ± 320 B.P. (
13
C —20.980 ‰) (Pacher & Stuart
2008; Dimitrijević in prep.). These data are rather consistent
with the age estimated on the basis of arvicoline remains.
Layer 2
Layer 3
Layer 4
Species
NISP % MNI % NISP % MNI % NISP % MNI %
Arvicola terrestris
9 6.8 1 2.1
2
12.5
1
11.1
14 5.4 2 2
Chionomys nivalis
7 5.3 5 10.4
–
–
–
–
23 8.8 12 12.1
Microtus arvalis & M. agrestis 39 29.6 21 43.7
5
31.2
3
33.3
129 49.4 61 61.6
Microtus gregalis
7 5.3 4 8.3
–
–
–
–
–
–
–
–
Microtus subterraneus
10 7.6 7 14.6
3
18.8
3
33.3
23 8.8 12 12.1
Clethrionomys glareolus
14 10.6 3 6.3
–
–
–
–
47 18
8 8.1
Lagurus lagurus
46 34.8 7 14.6
6
37.5
2
22.2
25 9.6 4 4
Table 7: Abundance of individual taxa of arvicolids.
ø
92
BOGIĆEVIĆ, NENADIĆ and MIHAILOVIĆ
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2012, 63, 1, 83—94
Conclusions
In this paper, the remains of eight species of arvicolines
from Layers 2, 3 and 4 have been analysed. On the basis of
their study, it was concluded that the associations from Lay-
ers 2 and 4 (Layer 3 is very poor in fossils) are very similar in
composition and that they were formed during cold episodes
of the last glacial, under conditions of an open environment
and dry climate.
The method for temperature estimation (described by Mon-
tuire 1996; Montuire et al. 1997) has also been applied, and al-
though the results thus obtained should be accepted with
caution, they seem to indicate that the deposits were formed
under rather cold conditions.
By measuring of the SDQ index on the Arvicola terrestris
teeth, it has been established that the values of this index in
Layers 2 and 4 from Baranica are similar to those from the last
glacial localities in Central Europe, with absolute ages of
about 30,000—35,000 B.P. The absolute dating gives the age
of 23,520 ± 110 B.P. for Layer 2 and 35,780 ± 320 B.P. for
Layer 4, which is rather consistent with the estimates obtained
by measuring of the SDQ index.
These results should be seen as preliminary. Other faunal el-
ements also need to be studied and so we will get a more com-
plete picture of the composition of the fauna and
paleoecological conditions in the Late Pleistocene in this re-
gion. In this way, correlations with the faunas from neighbour-
ing countries will be enabled, as well as the study of the
evolution and migration of small mammals in this interesting
and dynamic part of Europe.
Acknowledgments: We wish to express our gratitude to
Vesna Dimitrijević (Department of Archaeology, Faculty of
Philosophy, Belgrade) who gave us the sediment samples
from Baranica I and some useful information on large mam-
mals. We are indebted to Jadranka Mauch Lenardić (Institute
for Quaternary Paleontology and Geology, Croatian Academy
of Sciences and Arts, Zagreb) and Zoran Stevanović (Depart-
ment of Hydrogeology, University of Belgrade-Faculty of
Mining and Geology) for their assistance in obtaining the
necessary literature. Comments and suggestions by two re-
viewers, Lutz Maul (Forschungsinstitut Senckenberg,
Forschungsstation für Quartärpaläontologie, Weimar) and
Borut Toškan (Inštitut za arheologijo ZRC SAZU, Ljubljana)
and by the Editor are much appreciated. This work was sup-
ported by grants from the Ministry of Science and Techno-
logical Development of the Republic of Serbia under
Projects No. 176015 (for K. B. and D. N.) and 177023 (for
D. N. and D. M.).
References
Argant J. & Dimitrijević V. 2007: Pollen analyses of Pleistocene
hyaena coprolites from Montenegro and Serbia. Geol. An. Bal-
kan. Poluostr. 68, 73—80.
Bogićević K. 2004: An Upper Pleistocene arvicolid fauna from Ba-
ranica cave near Knjaževac (Eastern Serbia). 18th International
Senckenberg Conference. 6th International Palaeontological
Colloquium in Weimar. 25—30 April 2004, Weimar, Germany,
Conference Vol., 78.
Bogićević K. 2008: Pleistocene rodents (Rodentia) of Serbia. Unpubl.
Doctoral Dissertation, University of Belgrade, Belgrade, 1—202
(in Serbian).
Bogićević K. & Dimitrijević V. 2004: Quaternary fauna from Mališi-
na Stijena near Pljevlja (Montenegro). Zbornik radova Odbora
za kras i speleologiju 8, 119—133.
Bogićević K., Nenadić D., Mihailović D., Lazarević Z. & Milivojević
J. 2011: Late Pleistocene rodents (Mammalia: Rodentia) from
the Baranica Cave near Knjaževac (eastern Serbia): systematics
and palaeoecology. Riv. Ital. Paleont. Stratigr. 117, 2, 331—346.
Brunet-Lecomte P., Montuire S. & Dimitrijevic V. 2001: The Pleis-
tocene subterranean voles Terricola (Rodentia) of Serbia and
Montenegro. Paläont. Z. 75, 189—196.
Chaline J. 1972: Les ronguers du Pléistoc
e
ne Moyén et Supérieur de
France. Cahiers de Paléontologie, CNRS, Paris, 1—410.
Cuenca-Bescós G., Straus L.G., González Morales M.R. & García Pi-
mienta J.C. 2008: Paleoclimate and landscape at the Late Qua-
ternary in Cantabria: the small mammals from El Mirón Cave
(Ramales de la Victoria).
Rev. Esp. Paleontol. 23, 1, 91—126
(in Spanish with English summary).
Cvijić J. 1887: Contribution to our geographical terminology. Pros-
vetni Glasnik, December 1887, 903—916 (in Serbian).
Ćalić J. 2007: Karst research in Serbia before the time of Jovan
Cvijić. Acta Carsologica 36, 2, 315—319.
Dimitrijević V. 1985: Quaternary fauna from the Smolućka Cave.
Novopazarski Zbornik 9, 19—20 (in Serbian).
Dimitrijević V. 1991: Quaternary mammals of the Smolućka cave in
southwest Serbia. Palaeont. Jugoslavica 41, 1—88.
Dimitrijević V. 1994: Palaeontological research in Petnička cave –
excavations in 1992. Zbornik radova Odbora za kras i spele-
ologiju 5, 99—106 (in Serbian).
Dimitrijević V. 1997a: Upper Pleistocene mammals from cave depos-
its of Serbia. Geol. An. Balkan. Poluostr. 61, 179—370.
Dimitrijević V. 1997b: Pleistocene mammal fauna of eastern Serbia.
In: Lazić M. (Ed.): Arheologija istočne Srbije. Filozofski
Fakultet – Centar za Arheološka Istraživanja, Beograd – Donji
Milanovac 18, 45—55 (in Serbian).
Dimitrijević V. 1998: Fossil remains of vertebrates in Serbian Cave
Deposits. In: Djurović P. (Ed.): Speleological atlas of Serbia.
Geografski institut SAN “Jovan Cvijić”, Posebna izdanja, Bel-
grade, 52, 51—57.
Dimitrijević V. 1999: Vertebrate fauna from the Epigravettian site
of Trebački Krš near Berane. In: Bailey G., Adam E.,
Panagopoulou E., Perles C. & Zachos K. (Eds.): The Palae-
olithic Archaeology of Greece and Adjacent Areas. Proceed-
ings of the ICOPAG Conference, British School at Athens
Studies. September 1994, 3, Ioannina, Greece, 357—360.
Dimitrijević V. 2004: Cave hyaena’s prey selection in the Upper
Pleistocene fauna from the Baranica Cave near Knjaževac
(southeast Serbia). 18th International Senckenberg Conference.
6th International Palaeontological Colloquium in Weimar.
25—30 April 2004, Weimar, Germany, Conference Vol., 100.
Dimitrijević V. & Jovanović K. 2002: Quaternary mammals from the
Mirilovska Cave near Ćuprija (eastern Serbia). Zbornik Radova
Odbora za Kras i Speleologiju 7, 113—124.
Djurović P. 1998: Morphological and hydrological features of caves
and pits. In: Djurović P. (Ed.): Speleological atlas of Serbia.
Geografski Institut SAN “Jovan Cvijić”, Posebna izdanja, Bel-
grade 52, 39—44.
Fl
o
jgaard C., Normand S., Skov F. & Svenning J.-C. 2009: Ice age
distributions of European small mammals: insights from species
distribution modelling. J. Biogeography 36, 6, 1152—1163.
Forsten A. & Dimitrijević V. 2004: Pleistocene horses (genus Equus)
in the Central Balkans. Geol. An. Balkan. Poluostr. 65, 55—75.
ø
è
93
LATE PLEISTOCENE VOLES FROM THE BARANICA CAVE (SERBIA)
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2012, 63, 1, 83—94
Gromov I.M. & Polyakov I.Y. 1992: Fauna of the USSR: Mam-
mals. Vol. III, No. 8. Voles (Microtinae). Smithsonian Institu-
tion Libraries & National Science Foundation, Washington,
D.C., 1—725.
Guiot J., Torre F., Jolly D., Peyron O., Boreux J.J. & Cheddadi R.
2000: Inverse vegetation modeling by Monte Carlo sampling to
reconstruct palaeoclimates under changed precipitation season-
ality and CO
2
conditions: application to glacial climate in Medi-
terranean region. Ecological Modelling 127, 119—140.
Heinrich W.-D. 1978: Zur biometrischen Erfassung eines Evolution-
strends bei Arvicola (Rodentia, Mammalia) aus dem Pleistozän
Thüringiens. Säugetierkundliche Informationen 2, 3—21.
Heinrich W.-D. 1987: Neue Ergebnisse zur Evolution und Biostratig-
raphie von Arvicola (Rodentia, Mammalia) im Quartär Europas.
Z. Geol. Wissenschaft. 15, 389—406.
Jankičević J. 1978: Barrémien et Aptien des parties moyennes des
Carpatho-Balkanides dans la Serbie orientale au point de vue du
développement d’Urgonien. Geol. An. Balkan. Poluostr. 42,
103—194 (in Serbian, with a French summary).
Jovanović K. 2005: Late Pleistocene rodents (Rodentia) from the Ba-
ranica Cave near Knjaževac (eastern Serbia). Unpubl. Master
Thesis, University of Belgrade, Belgrade, 1—98 (in Serbian).
Jovanović M. & Simić V. 1998: Fossil rodents (Rodentia) from the
Risovača Cave near Arandjelovac (Serbia). Zapisnici Srpskog
Geološkog Društva za 1992—1997 godinu, 191—197 (in Serbian).
Kordos L. 1987: Climatostratigraphy of Upper Pleistocene Verte-
brates and the Conditions of Loess Formation in Hungary. Geo-
Journal 15, 2, 163—166.
Kowalski K. 2001: Pleistocene rodents of Europe. Folia Quaternaria
72, 1—389.
Krantić J. 1997: Pleistocene rodents (Rodentia, Mammalia) from the
cave deposits of the Baranica Cave near Knjaževac. Unpubl. Un-
dergraduate Thesis, University of Belgrade, Belgrade, 1—29 (in
Serbian).
Kryštufek B. 1990: Geographic variation in Microtus nivalis (Mar-
tins, 1842) from Austria and Yugoslavia. Bonner Zoo. Beitr. 41,
121—139.
Kryštufek B. 2004: A quantitative assessment of Balkan mammal di-
versity. In: Griffiths I.H., Kryštufek B. & Reed J.M. (Eds.): Bal-
kan biodiversity, pattern and process in the European Hotspot.
Kluwer Acad. Publ., Dordrecht, 79—133.
Kryštufek B., Tvrtković N. & Mikuska J. 1989: Distribution of the
field vole (Microtus agrestis) in Yugoslavia. Věstník Českoslo-
venské Společnosti Zoologické 53, 195—199.
Malez M. 1975: Quaternary fauna of Crvena Stijena. [Kvartarna fau-
na Crvene Stijene.] In: Basler Dj. (Ed.): Crvena stijena. Zbornik
radova. Zajednica Kulturnih Ustanova, Nikšić, 147—169.
Malez M. & Salković S. 1988: Geological and palaeontological char-
acteristics of the Quaternary deposits in the Cave over the
Hajdučki Izvor beside Čitluk near Sokobanja (SR Serbia). Naš
Krš 14, 89—99 (in Croatian).
Marković Z. & Pavlović G. 1991: The first results of the investiga-
tions of the fauna from the Vrelska Cave (Bela Palanka, Serbia).
Geol. An. Balkan. Poluostr. 55, 221—230 (in Serbian).
Mauch Lenardić J. 2007: Comparative metric analysis of Late Pleis-
tocene Microtus ex gr. arvalis/agrestis (Arvicolidae, Rodentia,
Mammalia) teeth from some Croatian localities. Cour. Forsch.-
Inst. Senckenberg 259, 149—154.
Maul L. 1990: Überblick über die unterpleistozänen Kleinsäuger-
faunen Europas. Quartärpaläontologie 8, 153—191.
Maul L. & Parfitt S. 2010: Micromammals from the 1995 Mam-
moth Excavation at West Runton, Norfolk, UK: Morphometric
data, biostratigraphy and taxonomic reappraisal. Quart. Int.
228, 91—115.
Maul L., Masini F., Abbazzi L. & Turner A. 1998: The use of differ-
ent morphometric data for absolute age calibration of some
South and Middle European arvicolid populations. Palaeontogr.
Italica 85, 111—151.
Maul L., Rekovets L., Heinrich H.-D., Keller T. & Storch G. 2000:
Arvicola mosbachensis (Schmidtgen, 1911) of Mosbach 2: a ba-
sic sample for the early evolution of the genus and a reference
for further biostratigraphical studies. Senckenberg. Lethaea 80,
129—147.
Medved M. 1994: Late Pleistocene fauna of the Pećurski kamen
Cave. Unpubl. Undergraduate Thesis, University of Belgrade,
Belgrade, 1—37 (in Serbian).
Meulen van der A.A.J. 1973: Middle Pleistocene smaller mammals
from the Monte Peglia (Orvieto, Italy) with special reference to
the Phylogeny of Microtus (Arvicolidae, Rodentia). Quaternaria
14, 1—144.
Mihailović D. 2004: Investigations of the cave archaeological sites in
the Timok and Nišava basins. Zbornik radova Odbora za kras i
speleologiju 8, 135—144 (in Serbian, with English summary).
Mihailović D. & Jovanović S. 1997: Prehistoric paintings in the Cave
Gabrovnica near Kalna (East Serbia). Starinar 48, 135—144
(in Serbian, with English summary).
Mihailović D., Djuričić Lj. & Kaludjerović Z. 1997: Investigations of
the Palaeolithic on the territory of eastern Serbia. In: Lazić M.
(Ed.): Arheologija Istočne Srbije. Filozofski Fakultet – Centar
za arheološka istraživanja, Beograd, 33—42 (in Serbian, with
English summary).
Montuire S. 1996: Rodents and climate. II. Quantitative climatic esti-
mates for Plio-Pleistocene faunas from Central Europe. Acta
Zoologica Cracoviensia 39, 1, 373—379.
Montuire S., Michaux J., Legendre S. & Aguilar J.P. 1997: Rodents
and climate 1. A model for estimating past temperatures using
arvicolids (Mammalia: Rodentia). Palaeogeogr. Palaeoclimatol.
Palaeoecol. 128, 187—206.
Nadachowski A. 1982: Late Quaternary rodents of Poland with spe-
cial reference to morphotype dentition analysis of voles. Państw.
Wydaw. Naukowe, Warszawa—Kraków, 1—109.
Nadachowski A. 1984a: Morphometric variability of dentition of the
Late Pleistocene voles (Arvicolidae, Rodentia) from Bacho Kiro
Cave (Bulgaria). Acta Zoologica Cracoviensia 27, 149—176.
Nadachowski A. 1984b: Taxonomic value of anteroconid measure-
ments of m1 in common vole and field voles. Acta Theriologica
29, 123—127.
Nadachowski A. 1985: Biharian voles (Arvicolidae, Rodentia, Mam-
malia) from Kozi Grzbiet (Central Poland). Acta Zoologica Cra-
coviensia 29, 13—28.
Pacher M. & Stuart A.J. 2008: Extinction chronology and palaeo-
biology of the cave bear (Ursus spelaeus). Boreas 38, 189—206.
Pavlović G. & Marković Z. 1991: The first finding of Lagurus lagu-
rus (Pallas) (Rodentia, Mammalia) in Yugoslavia. Geol. An.
Balkan. Poluostr. 55, 197—206 (in Serbian).
Petrović J. 1976: Pits and caves of SR Serbia. Vojnoizdavački Centar,
Beograd, 1—511 (in Serbian).
Pokines J.T. 1998: The Paleoecology of Lower Magdalenian Cant-
abrian Spain. BAR Int. Ser. 713, 1—189.
Popov V.V. 2000: The small mammals (Mammalia: Insectivora,
Lagomorpha, Rodentia) from Cave 16 (North Bulgaria) and the
paleoenvironmental changes during the Late Pleistocene. In:
Ginter B., Kozlovski J.K., Guadelli J.-L. & Laville H. (Eds.):
Temnata Cave. Excavations in Karlukovo Karst Area, Bulgaria
2, 1. Jagellonian University Press, Kraków, 159—240.
Popov V.V. & Marinska M. 2007: An almost one million year long
(Early to Late Pleistocene) small mammal succession from the
archaeological layers of Kozarnika Cave in Northern Bulgaria.
Cour. Forsch.-Inst. Senckenberg 259, 79—92.
Rabeder G. 2004: Micro-mammals from Pleistocene sediments of
Potočka zijalka (Slovenia). In: Pacher M., Pohar V. & Rabeder
G. (Eds.): Potočka Zijalka: palaeontological and archaeological
94
BOGIĆEVIĆ, NENADIĆ and MIHAILOVIĆ
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA
GEOLOGICA CARPATHICA, 2012, 63, 1, 83—94
results of the campaigns 1997—2000. Mitt. Kommission Quartär-
forschung Österr. Akad. Wissenschaft. 13, Verlag Österr. Akad.
Wissenschaft., Wien, 55—65.
Radulescu C. & Samson P. 1992: Small mammals of the penultimate
glacial cycle (Saale/Riss) discovered in two caves from North-
western Oltenia, Romania. Theoretical and Applied Karstology
5, 203—211.
Rakovec I. 1965: Pleistocene mammal fauna from the Risovača Cave
near Arandjelovac. Razprave IV. razreda SAZU 8, 223—317 (in
Slovenian, with an extended summary in English).
Röttger U. 1987: Schmelzbandbreiten an Molaren von Schermäusen
(Arvicola Lacepede, 1799). Bonner Zoo. Beitr. 38, 95—105.
Salčin E. 1996: Quaternary carnivores (Mammalia, Carnivora) from
the Baranica Cave near Knjaževac. Unpubl. Undergraduate The-
sis, University of Belgrade, Belgrade, 1—35 (in Serbian).
Sesé C. 2005: Contribution of small mammals to the knowledge of
paleoenvironment of Late Pleistocene in Cantabria Region:
new data and synthesis. [Aportación de los micromamíferos al
conocimiento paleoambiental del Pleistoceno Superior en la
Región Cantábrica: Nuevos datos y síntesis.] Museo y Centro
de Investigación de Altamira Monográfias 20, 167—200 (in
Spanish).
Sladić M. & Jovanović S. 1996: Sondage recognition of the cave sites
in the area of Knjaževac. Glasnik Srpskog Arheološkog Društva
11, 228—232 (in Serbian).
Stevanović Z. 1994: Karst ground waters of Carpatho-Balkanides in
Eastern Serbia. In: Stevanović Z. & Filipović B. (Eds.): Ground
waters in carbonate rocks of the Carpathian-Balkan mountain
range. Carpathian-Balkan Geol. Assoc., Belgrade, 203—237.
Stevanović Z. & Filipović B. 1994: Hydrogeology of carbonate
rocks of Carpatho-Balkanides. In: Stevanović Z. & Filipović
B. (Eds.): Ground waters in carbonate rocks of the Carpathian-
Balkan mountain range. Carpathian-Balkan Geol. Assoc.,
Belgrade, 35—112.
Terzea E. 1972: Remarques sur la morphologie dentaire et la répariti-
tion de Microtus nivalis Martins dans le Pléistoc
e
ne de Rou-
manie. Travaux de l’Institut de Spéleologie “Émile Racovitza”
11, 271—298.
Toškan B. 2009: Small terrestrial mammals (Soricomorpha, Chi-
roptera, Rodentia) from the Early Holocene of Mala Triglavca
(SW Slovenia). Acta Carsologica 38, 1, 117—133.
Toškan B. & Kryštufek B. 2007: Small terrestrial mammals (Erina-
ceomorpha, Soricomorpha, Chiroptera, Rodentia) from Divje
babe I. In: Turk I. (Ed.): Divje babe I. Paleolitsko najdišče mlaj-
šega pleistocena v Sloveniji. I del: Geologija in paleontologija.
Opera Instituti Archaeologici Sloveniae, 13, Založba ZRC,
Ljubljana, 193—219.
Zimmermann K. 1956: Zur Evolution der Molarenstruktur der Erd-
maus, Microtus agrestis (L.). Zoo. Jb., Abteilung für Systematik,
Ökologie und Geographie der Tiere 84, 270—274.
Zlokolica Mandić M. 1998: Structural-tectonic elements as a factor in
cave development. In: Djurović P. (Ed.): Speleological Atlas of
Serbia. Geografski Institut SAN “Jovan Cvijić”, Posebna izdanja,
Belgrade 52, 15—21.
è