Isotopic variability of cave bears (d15
N, d13
C) across Europe during
MIS 3
Magdalena Krajcarz a, *
, Martina Pacher b
, Maciej T. Krajcarz a
, Lana Laughlan c
,
Gernot Rabeder b
, Martin Sabol d
, Piotr Wojtal e
, Herve Bocherens f, g
a
Institute of Geological Sciences, Polish Academy of Sciences, Twarda 51/55, 00-818 Warszawa, Poland
b
Institute of Palaeontology, University of Vienna, Althanstr. 14, A-1090 Vienna, Austria
c
Museum Inatura, Jahngasse 9, 6850 Dornbirn, Austria
d
Department of Geology and Palaeontology, Faculty of Natural Sciences, Comenius University, Ilkovicova Str. 6, SK-84215 Bratislava, Slovak Republic
e
Institute of Systematics and Evolution of Animals, Polish Academy of Sciences, Sławkowska 17, 31-016 Krakow, Poland
f
Fachbereich Geowissenschaften, Pal€aobiologie (Biogeologie), Universit€at Tübingen, H€olderlinstrasse 12, D-72074 Tübingen, Germany
g
Senckenberg Center for Human Evolution and Palaeoenvironment (HEP), Universit€at Tübingen, H€olderlinstrasse 12, D-72074 Tübingen, Germany
a r t i c l e i n f o
Article history:
Received 17 March 2015
Received in revised form
14 October 2015
Accepted 15 October 2015
Available online 6 November 2015
Keywords:
Poland
Slovakia
Austria
Ursus spelaeus
Palaeoecology
Altitude
Carbon-13
Nitrogen-15
Diet
a b s t r a c t
Collagen, the organic fraction of bone, records the isotopic parameters of consumed food for carbon
(d13
C) and nitrogen (d15
N). This relationship of isotopic signature between diet and tissue is an important
tool for the study of dietary preferences of modern and fossil animal species. Since the ﬁrst information
on the isotopic signature of cave bear was reported, numerous data from Europe have become available.
The goal of this work is to track the geographical variation of cave bear collagen isotopic values in Europe
during Marine Isotopic Stage 3 (about 60,000e25,000 yr BP). In this study the results of new d13
C and
d15
N isotopic analyses of cave bear collagen from four Central-Eastern European sites are presented, as
well as a review of all published isotopic data for cave bears of the same period. The main conclusion is a
lack of geographical East-West pattern in the variations of d13
C and d15
N values of cave bear collagen.
Moreover, no relationship was found between cave bear taxonomy and isotopic composition. The cave
bears from Central-Eastern Europe exhibit d13
C and d15
N values near the average of the range of Central,
Western and Southern European cave bears. Despite the fact that most cave bear sites follow an altitudinal
gradient, separate groups of sites exhibit shift in absolute values of d13
C, what disturbs an
altitude-related isotopic pattern. The most distinct groups are: high Alpine sites situated over 1500 m a.s.l.
e in terms of d13
C; and two Romanian sites Pes¸ tera cu Oase and Urs¸ ilor e in case of d15
N. Although the
cave bear isotopic signature is driven by altitude, the altitudinal adjustment of isotopic data is not enough
to explain the isotopic dissimilarity of these cave bears. The unusually high d15
N signature of mentioned
Romanian sites is an isolated case in Europe. Cave bears from relatively closely situated Central-Eastern
European sites and other Romanian sites are more similar to Western European than to Romanian
populations in terms of isotopic composition, and probably ecology.
© 2015 Elsevier Ltd. All rights reserved.
1. Introduction
Bears from the spelaeus-group (Ursus ex gr. spelaeus Rosenmüller,
1794) represent large extinct ursids with the most abundant
fossil remains in Late Pleistocene caves in Europe. Cave bears
exhibit a large amount of variation in size and morphology (e.g.,
Kurten, 1976; Rabeder et al., 2000; Baryshnikov and Puzachenko,
2011). The large geographical range across Europe, combined
with morphological and genetic diversity of cave bears, opens the
possibility of some ecological ﬂexibility for this extinct taxon
(Bocherens, 2015). Based on morphological and genetic differences,
several types of cave bears have been recognized (e.g., Hofreiter
et al., 2002; Rabeder et al., 2008). In Europe, two genetic types
were common and coexisted between around 60,000 and 24,000
years ago, corresponding to Ursus spelaeus, present in the western
part of Europe, and Ursus ingressus, present in the eastern part of
continent and expanding westward (Stiller et al., 2014). The
occurrence of both types is documented in the western part of
Central Europe, sometimes in contemporaneous close sites, as in
* Corresponding author.
E-mail address: magdakraj@twarda.pan.pl (M. Krajcarz).
Contents lists available at ScienceDirect
Quaternary Science Reviews
journal homepage: www.elsevier.com/locate/quascirev
http://dx.doi.org/10.1016/j.quascirev.2015.10.028
0277-3791/© 2015 Elsevier Ltd. All rights reserved.
Quaternary Science Reviews 131 (2016) 51e72
the Totes Gebirge in Austria (Hofreiter et al., 2004a), while in other
cases both types occurred in chronological succession, as in the Ach
valley in southwestern Germany, where U. ingressus is present after
the extirpation of U. spelaeus (Hofreiter et al., 2007; Münzel et al.,
2011). The third genetic type recognized in Europe is Ursus ladinicus,
known from several cave sites in Prealpine and Alpine region
(Hofreiter et al., 2002). The cave bear remains from Caucasus
represent another genetic form Ursus kudarensis (Ursus “deningeri”
kudarensis), the sister group of the European cave bears, which split
from the rest of the cave bears between 800 and 270 ka ago
(Baryshnikov, 1998; Knapp et al., 2009). All these taxonomic types
are regarded as generally ecologically similar, what was deducted
from their isotopic signature (Bocherens, 2015), although some
minor differences between local populations are observed
(Bocherens et al., 2011b), and U. kudarensis and U. ladinicus forms
seem to be ecologically the most distinct (Bocherens, 2015).
Cave bear was the ﬁrst extinct species for which carbon and
nitrogen stable isotopes were used to document diet (Bocherens
et al., 1990, 1991). Early studies found nitrogen isotopic ratios as
low as those of herbivorous species, suggesting an essentially
vegetarian diet, and this view was conﬁrmed by most of later
studies (e.g., Bocherens et al., 1994, 1997, 2011a; Nelson et al., 1998;
Münzel et al., 2014; Bocherens, 2015). Other palaeodietary proxies
such as tooth microwear, ecomorphology and taphonomy have also
suggested a herbivorous diet for this species (e.g., Ehrenberg, 1927;
Kurten, 1976; Stiner et al., 1998; Van Heteren et al., 2009; Münzel
et al., 2014). In contrast, some morphometric analyses based on
the craniodental traits of cave bears suggested a similar level of
omnivory than modern omnivorous bears (Figueirido et al., 2009),
and some studies based on tooth microwear and taphonomy suggested
a more omnivorous diet (e.g., Quiles et al., 2006; Peigne
et al., 2009). The method of using the stable isotopes of carbon
and nitrogen to evaluate palaeodiets has been applied on cave bears
from dozens of cave sites, but so far mostly in Western Europe (e.g.,
Bocherens et al.,1994, 2006, 2011a,b; Blant et al., 2010; Pacher et al.,
2012; Bocherens, 2015), and fewer sites in Southern and SouthEastern
Europe have been studied (e.g., Fernandez-Mosquera
et al., 2001; Richards et al., 2008; Perez-Rama et al., 2011;
Horacek et al., 2012; Robu et al., 2013; Trinkaus and Richards,
2013). Interestingly, some cave bears from two South-Eastern European
sites (Urs¸ ilor and Pes¸ tera cu Oase, Romania) exhibited
higher concentrations of heavier nitrogen isotope than other cave
bear populations studied so far and therefore did not seem to
follow the general pattern of the cave bear isotopic signal (Richards
et al., 2008; Robu et al., 2013; Trinkaus and Richards, 2013). This
phenomenon led to the suggestion that these Romanian cave bears
may have occupied a higher trophic level than Western European
cave bears (Richards et al., 2008; Robu et al., 2013).
This situation raises a question about the geographical variation
of isotopic patterns in cave bear. Is the Romanian isotopic signature
a local phenomenon, or is it typical for East European cave bear
populations? Moreover, was this phenomenon caused by speciﬁc
local conditions, or did it follow a geographical trend connected
with the climatic divergence between oceanic conditions of
Western Europe and more continental climates of Eastern Europe?
However, since two other Romanian sites (Cioclovina and Muierii)
showed carbon and nitrogen isotopic signatures similar to most
other cave bears (Robu et al., 2013), a continental-scale climatic
hypothesis is less likely. This issue of cave bear palaeoecology is
important to better address the possible interactions between this
large mammal species and prehistoric human populations in
Europe (e.g., Münzel and Conard, 2004; Wojtal et al., 2015), as well
as to understand the causes of its ﬁnal demise prior to the onset of
the Late Glacial Maximum (Pacher and Stuart, 2009).
An in-depth discussion on the isotopic differences (and
ecological differences) between cave bears from Romanian sites
and other Eastern and Western European sites is problematic, as
isotopic data from cave bear collagen are still rare for Eastern
Europe. The lack of data from parts of Eastern Europe other than
Romania makes it difﬁcult to evaluate a geographical extent of the
phenomenon of high d15
N values of two cave bear assemblages.
The goal of this work is to complete the picture of cave bear
isotopic variability by adding new data from Central-Eastern
Europe (Austria, Poland and Slovakia), and then to evaluate the
distribution of C and N stable isotopes among European cave bears.
2. Principles of isotopic tracking of cave bear diet
2.1. Isotopes of C and N in terrestrial foodwebs
The organic fraction of bone, essentially made of collagen, records
the isotopic parameters of consumed food for carbon (d13
C)
and nitrogen (d15
N) (e.g., DeNiro and Epstein, 1978; Bryant and
Froelich, 1995; Cerling and Harris, 1999; Bocherens, 2000; LeeThorp
and Sponheimer, 2003; Passey et al., 2005; Koch, 2007).
The carbon isotopic composition in ecosystems depends on plant
communities and differs between dense and light forests, grasslands,
deserts and lichen-dominated environments (DeNiro and
Epstein, 1978; Bocherens, 2003). It is primarily related to the
photosynthetic pathways and secondarily to environmental conditions
that affect plant physiology, such as aridity, temperature,
CO2 partial pressure, salinity, light conditions, and altitude (e.g.,
K€orner et al., 1991; Fricke, 2007; Diefendorf et al., 2010; Zhu et al.,
2010). In addition, a canopy effect has an impact on the carbon
isotopic signatures in ecosystems. Indeed, under dense vegetation
cover, the isotopic signature of CO2 in air becomes 13
C-depleted,
because the lighter carbon, released by decaying organic material
and soil respiration, is recycled during photosynthesis, and because
of increased carbon fractionation during low-light photosynthesis
(Van der Merve and Medina, 1991; Fricke, 2007; Drucker et al.,
2008; Bonaﬁni et al., 2013). The primary isotopic signal of nitrogen
in ecosystems is essentially linked to microbial soil activity
(Mariotti et al., 1980; Amundson et al., 2003; Sah and Brumme,
2003). The lowest d15
N values are found in acidic soils (Mariotti
et al., 1980; Rodiere et al., 1996) and in pioneer vegetation
(Hobbie et al., 1998). Factors such as altitude (Mariotti et al., 1980),
cultivation (Riga et al., 1971), high grazing pressure (Schulze et al.,
1998), salinity (Page, 1995) and aridity (Ambrose and DeNiro, 1986;
Heaton et al., 1986; Bocherens et al., 2014a; Craine et al., 2015) also
modify the nitrogen isotopic signal in plants.
The isotopic signal of primary sources is changed along with
trophic level in ecosystems, due to animal metabolism (e.g., DeNiro
and Epstein, 1981; Roth and Hobson, 2000; Bocherens, 2000, 2004;
Bocherens and Drucker, 2003; Seminoff et al., 2009). The d15
N
values increase on average 3e5‰ at each trophic step (Minagawa
and Wada, 1984; Schoeninger, 1985; Bocherens and Drucker,
2003), while d13
C values increase 0.8e1.3‰. The phenomenon of
carbon and nitrogen isotopic fractionation is a principle of isotopic
ecology and palaeoecology that is used to reconstruct a position of a
fossil animal in a trophic web, especially in the case of opportunistic
feeders such as bears (e.g., DeNiro and Epstein, 1978; Ambrose and
DeNiro,1986; Bocherens et al.,1994, 2011b, 2014a; Fizet et al.,1995;
Hilderbrand et al., 1996; Hobson et al., 2000; Vila Taboada et al.,
2001; Richards et al., 2008; García et al., 2009; Drucker et al.,
2011; Robu et al., 2013).
Nursing causes an additional effect on C and N isotopic
composition similar to a higher trophic level, since milk consumed
by young mammals during suckling will place them one trophic
level higher than adults of the same species (Fogel et al., 1989;
Bocherens, 2000; Jenkins et al., 2001; Dalerum et al., 2007).
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e7252
2.2. Isotopes of C and N in bear collagen
Bears from middle and high latitudes, including cave bears, are
unique among large mammals as they hibernate (e.g., Nelson et al.,
1975; Stiner et al., 1998), which has consequences on their isotopic
fractionation (e.g., Bocherens et al., 1994; Nelson et al., 1998; Liden
and Angerbjørn, 1999; Fernandez-Mosquera et al., 2001;
Bocherens, 2004; Perez-Rama et al., 2011). The winter sleep has a
signiﬁcant impact on their physiology, as it causes changes in the
metabolism of amino acids. During winter sleep the animals do not
defecate, and the nitrogen-containing waste is reabsorbed (Nelson
et al., 1975). The reuse of the urea component increases the trophic
level of amino acids and leads to an increase in the d15
N
signature. During hibernation bears do noteat, but use fat stores. The
typical relationship between d13
C and different tissues established
for modern animal follows the pattern of highest d13
C in hair, then
decreasing in brain, muscle and liver, with most depleted d13
C in fat
(Tieszen et al.,1983; De Smet et al., 2004). Therefore the bear tissues
formed during hibernation may record more negative values of d13
C
(Hilderbrand et al., 1996; Bocherens et al., 1997). However, it is
claimed that the effect of hibernation affects mostly bones of young
individuals, while bones of adults seem to record a diet-related
isotopic signature (Bocherens, 2004; Perez-Rama et al., 2011).
Therefore palaeodietary studies using C and N isotopes in ancient
bears should use bones of adult individuals (Bocherens, 2004).
3. Material and methods
In the present study, new isotopic data from cave bears from
Central-Eastern Europe were obtained, and compared to previously
published data for cave bears from the rest of Europe.
3.1. Sites from Central-Eastern Europe
The studied faunal material comes from four Central-Eastern
European caves located in Austria, Poland and Slovakia (Fig. 1).
3.1.1. Nietoperzowa Cave
Nietoperzowa Cave is located in the Krakow-Cze˛stochowa
Upland, in Be˛dkowska Valley, Poland (N 5011038.000 E 1946029.000).
The entrance elevation is located at an altitude of 438 m a.s.l. The
archaeological exploration started at the beginning of 20th century
and was terminated in 1969, with the most extensive excavation
during 1953e1969. It is one of the most important cave sites in
Poland with cultural layers of the Jerzmanowician, and sediments
treated as the stratotype proﬁle of late Middle and Late Pleistocene
cave sediments (Ro_zycki, 1967; Madeyska, 1982; Mojski, 1985;
Lindner, 1992). The stratigraphy was based on geological, palaeontological
and archaeological evidence (Chmielewski, 1961, 1975;
Kowalski, 1961; Madeyska-Niklewska, 1969; Wysoczanski-Minkowicz,
1969; Madeyska, 1981; Krajcarz and Madeyska, 2010). Palaeontological
studies have been published by Wojtal (2007). Bone
assemblage comprises over 11,000 of remains, representing mostly
carnivorans (U. ex gr. spelaeus, Canis lupus, Meles meles, Vulpes
lagopus, V. vulpes, Gulo gulo, Felis silvestris, Panthera spelaea) and in
small number ungulates (Mammuthus primigenius, Equus sp., Coelodonta
antiquitatis, Sus scrofa, Capreolus capreolus, Cervus elaphus,
Megaloceros giganteus, Rangifer tarandus, Bos/Bison sp., Ovibos
moschatus). The material is clearly dominated by cave bear bones
(more than 90% of the large mammal assemblage), some bones bear
traces of cut marks made by Palaeolithic hunters. The studied
material comes from the following layers: 1 (mixed sediments),
3e8 (MIS 3), 9 (MIS 4), 10e16 (MIS 5e6) (Table 1).
3.1.2. Perspektywiczna Cave
Perspektywiczna Cave is situated in the Krakow-Cze˛stochowa
Upland, in Udorka Valley, Poland (N 5026034.000 E 1946001.000),
about 28 km north of Nietoperzowa Cave. The site, with the
entrance located at 345 m a.s.l., was discovered in 2012. During the
excavation in 2013 the sediments with abundant fossil faunal remains
were discovered. All of the studied fossil remains come from
one layer; the AMS 14
C dating of the bones yielded ages of around
40,000 years BP, i.e. representing MIS 3 (Table 1). From this layer
about 750 mammalian bones were excavated, consisting of several
carnivorans species: U. ex gr. spelaeus, V. vulpes, M. meles, F. silvestris,
P. spelaea, Crocuta crocuta spelaea; and ungulate species: C. antiquitatis,
C. capreolus, M. giganteus, R. tarandus, Bos/Bison sp.
Fig. 1. Location of the newly studied sites (red spots, the more detail map to the right) and other European sites analysed in this paper (green spots e sites with unusually high d15
N
values according to Richards et al., 2008 and Robu et al., 2013; yellow spots e other sites). (For interpretation of the references to color in this ﬁgure legend, the reader is referred to
the web version of this article.)
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e72 53
Table 1
List of cave bear fossil samples from Austrian, Polish and Slovakian sites and of isotopic results and direct radiocarbon dating of cave bear bones (* e bone equivalent, bone e data for whole bone or tooth, coll e data for extracted
collagen).
Lab. no. Collection Site MIS Bone/tooth %Nbone %Cbone %Ccoll %Ncoll C/Ncoll d13
C d15
N 14
C age
determination
(calibrated age
acc. To IntCal2013,
95.4%)
IN-29 ISEA Nietoperzowa Cave 3 ? Metapodium 1.7 6.4 41.0 14.7 3.2 À21.2 2.7
IN-55 ISEA Nietoperzowa Cave 3 Phalanx 1 0.5 6.9 32.8 11.4 3.4 À21.8 4.2
IN-56 ISEA Nietoperzowa Cave 3 Phalanx 1 1.1 6.2 38.1 13.3 3.3 À21.2 2.1 26,800 ± 300 BP
Poz-62582
(31,330e30,493 calBP)
IN-25 ISEA Nietoperzowa Cave 3 Metatarsus V 0.7 4.6 35.4 12.6 3.3 À21.2 1.4
IN-24 ISEA Nietoperzowa Cave 3 Metacarpus III 0.6 4.9 36.6 13.0 3.3 À21.2 2.7 28,840 ± 190 BP
Poz-56744
(33,587e32,418 calBP)
29,800 ± 300
BP Poz-62581
(34,515e33,441 calBP)
IN-26 ISEA Nietoperzowa Cave 3 Phalanx 1 2.6 9.5 43.8 15.6 3.3 À21.1 3.8 26,190 ± 250
BP Poz-63860
(30,942e29,782
calBP)
IN-31 ISEA Nietoperzowa Cave 3 Phalanx 1 1.7 7.1 39.5 14.1 3.3 À21.1 2.1 29,970 ± 350 BP
Poz-63861
(34,703e33,533
calBP)
IN-54 ISEA Nietoperzowa Cave 3 Metatarsus III 3.3 10.8 40.6 14.4 3.3 À21.3 3.0 32,550 ± 600
BP Poz-56639
(36,482e34,486
calBP)
31,500 ± 500 BP
Poz-63862
(38,370e35,355
calBP)
IN-15 ISEA Nietoperzowa Cave 3 Metatarsus III 0.5 4.9 31.1 11.0 3.3 À21.3 2.5
IN-01 ISEA Nietoperzowa Cave 3 Canine 3.2 10.6 43.4 15.5 3.3 À22.5 * À22.1 7.0 * 5.1
IN-02 ISEA Nietoperzowa Cave 3 Mandibula 0.4 3.9 10.6 3.7 3.4 À22.2 3.3
IN-09 ISEA Nietoperzowa Cave 3 Tibia 0.5 4.2 20.7 7.2 3.3 À21.8 2.4
IN-32 ISEA Nietoperzowa Cave 3 Metatarsus III 0.4 3.9 6.5 1.6 4.6 À24.4 2.5
IN-33 ISEA Nietoperzowa Cave 3 Metatarsus V 0.3 3.4 e e e e e
IN-21 ISEA Nietoperzowa Cave 4 Metapodium 1.0 5.4 31.2 11.0 3.3 À21.2 3.6
IN-49 ISEA Nietoperzowa Cave 5 Metatarsus I 0.7 4.4 28.8 9.8 3.4 À21.5 3.9
IN-04 ISEA Nietoperzowa Cave 5 Mandibula 0.2 3.1 e e e e e
IN-28 ISEA Nietoperzowa Cave 5 Phalanx I 0.3 3.2 e e e e e
IN-03 ISEA Nietoperzowa Cave 5 Metatarsus IV 2.1 7.6 41.9 14.8 3.3 À20.5 2.5
IN-46 ISEA Nietoperzowa Cave 5 (5e?) Metatarsus IV 3.5 11.3 42.0 15.0 3.3 À21.3 1.7
IN-60 ISEA Nietoperzowa Cave 5 (5e?) Metatarsus I 2.4 8.1 37.6 13.2 3.3 À21.2 2.8
IN-22 ISEA Nietoperzowa Cave 5 (5e?) Metatarsus III 3.0 9.7 41.1 14.6 3.3 À21.0 2.1
IN-30 ISEA Nietoperzowa Cave 6 Phalanx 1 2.7 9.1 41.0 14.7 3.3 À23.2 2.8
IN-23 ISEA Nietoperzowa Cave 6? Metacarpus V 0.4 3.2 7.4 2.6 3.3 À22.1 1.7
IN-19 ISEA Nietoperzowa Cave 6? Metacarpus IV 0.2 2.9 e e e e e
IN-20 ISEA Nietoperzowa Cave 6? Cranium 0.1 1.8 e e e e e
IP-05 IGS Perspektywiczna Cave 3 Mandibula 2.5 9.4 41.1 14.8 3.2 À21.2 1.9
IP-06 IGS Perspektywiczna Cave 3 Mandibula 3.4 11.1 43.8 15.9 3.2 À21.5 5.1 40,200 ± 1200
BP Poz-61114
calBP)
M.Krajcarzetal./QuaternaryScienceReviews131(2016)51e7254
IP-09 IGS Perspektywiczna Cave 3 Metacarpus IV 4.0 12.5 39.5 14.2 3.2 À21.4 3.3 39,500 ± 1000 BP Poz-63863
(45,235e42,025 calBP)
41,600 ± 1400 BP Poz-61115
(48,500e42,932 calBP)
IP-13 IGS Perspektywiczna Cave 3 Mandible 1.9 7.6 30.7 10.4 3.4 À21.3 4.1
IP-14 IGS Perspektywiczna Cave 3 Phalanx 3 3.5 11.7 41.6 14.5 3.3 À21.2 3.0
IP-15 IGS Perspektywiczna Cave 3 Vertebra 3.5 11.4 43.3 15.2 3.3 À21.1 2.3
IP-16 IGS Perspektywiczna Cave 3 Fibula 2.6 9.4 39.7 13.7 3.4 À21.9 2.7
IP-17 IGS Perspektywiczna Cave 3 Fibula 3.8 12.2 42.8 15.0 3.3 À21.7 4.1
IP-18 IGS Perspektywiczna Cave 3 Metacarpus I 3.2 10.9 40.7 14.3 3.3 À21.2 2.4
IP-19 IGS Perspektywiczna Cave 3 Metatarsus IV e e 29.0 9.8 3.5 À21.0 1.6
MJ-01 SM Medvedia Cave 3 Metatarsus V 4.2 13.2 42.9 15.7 3.2 À20.7 1.4
MJ-02 SM Medvedia Cave 3 Metatarsus III 3.5 12.9 43.2 15.7 3.2 À21.0 1.9
MJ-03 SM Medvedia Cave 3 Metacarpus V 3.0 11.7 41.7 14.9 3.3 À21.0 3.1
MJ-04 SM Medvedia Cave 3 Metacarpus III 4.0 12.8 43.3 15.5 3.3 À20.9 1.9
MJ-05 SM Medvedia Cave 3 Metacarpus V 2.2 10.9 38.6 13.9 3.2 À21.0 2.3
MJ-06 SM Medvedia Cave 3 Metacarpus III 3.7 12.0 43.3 15.7 3.2 À21.0 2.2
MJ-07 SM Medvedia Cave 3 Metacarpus II 3.0 9.9 42.9 15.4 3.3 À21.3 1.5
MJ-08 SM Medvedia Cave 3 Metatarsus IV 3.7 12.1 44.7 15.8 3.3 À21.1 1.5
MJ-09 SM Medvedia Cave 3 Metacarpus II 3.8 12.9 39.6 14.2 3.2 À21.0 1.9
MJ-10 SM Medvedia Cave 3 Metacarpus IV 3.1 10.9 43.1 15.3 3.3 À20.7 3.7
MJ-11 SM Medvedia Cave 3 Metatarsus V 3.2 10.7 42.6 15.0 3.3 À20.9 2.8
MJ-12 SM Medvedia Cave 3 Metatarsus IV 3.6 11.4 43.9 15.1 3.4 À20.8 0.8
MJ-13 SM Medvedia Cave 3 Metatarsus II 3.5 12.3 43.4 15.5 3.3 À21.4 1.8
MJ-14 SM Medvedia Cave 3 Metatarsus I 3.7 12.8 43.6 15.6 3.3 À21.0 2.3
MJ-15 SM Medvedia Cave 3 Metatarsus I 2.9 10.5 44.0 15.2 3.4 À20.8 1.1
MJ-16 SM Medvedia Cave 3 Phalanx 1 3.9 12.3 41.8 14.9 3.3 À20.7 0.8
MJ-17 SM Medvedia Cave 3 Phalanx 1 2.7 10.4 43.8 15.3 3.3 À21.2 1.8
MJ-18 SM Medvedia Cave 3 Phalanx 2 2.5 9.5 44.3 15.5 3.3 À21.0 1.7
MJ-19 SM Medvedia Cave 3 Phalanx 2 3.4 12.4 45.2 15.5 3.4 À20.8 1.3
MJ-20 SM Medvedia Cave 3 Metatarsus III 0.8 4.4 42.9 14.7 3.4 À20.9 1.6
LPZ-17 SMB, IP Winden 3 Metapodial e e 33.1 11.9 3.3 À21.8 3.9
LPZ-18 SMB, IP Winden 3 Metapodial e e 26.6 9.7 3.2 À22.0 3.3
LPZ-19 SMB, IP Winden 3 Metapodial e e 32.5 11.5 3.3 À21.5 1.9
LPZ-20 SMB, IP Winden 3 Metapodial e e 38.8 13.7 3.3 À22.0 4.3
LPZ-21 SMB, IP Winden 3 Metapodial e e 39.3 13.9 3.3 À21.5 3.5
LPZ-22 SMB, IP Winden 3 Metapodial e e 33.1 11.8 3.3 À22.0 3.3
LPZ-23 SMB, IP Winden 3 Metapodial e e 36.8 13.2 3.3 À21.7 2.8
LPZ-24 SMB, IP Winden 3 Metapodial e e 29.6 10.5 3.3 À21.7 1.7
LPZ-25 SMB, IP Winden 3 Metapodial e e 35.7 12.7 3.3 À21.9 3.8
LPZ-26 SMB, IP Winden 3 Metapodial e e 28.4 10.1 3.3 À21.9 3.4
LPZ-27 SMB, IP Winden 3 Metapodial e e 24.5 8.8 3.3 À21.8 3.8
LPZ-28 SMB, IP Winden 3 Metapodial e e 27.6 10.9 2.9 À21.5 2.1
ISEA ¼ Institute of Systematics and Evolution of Animals of the Polish Academy of Sciences, Krakow (Poland).
IGS ¼ Institute of Geological Sciences of the Polish Academy of Sciences, Warszawa (Poland).
SM ¼ Spis Museum in Spisska Nova Ves (Slovakia).
SMB ¼ State Museum of Burgenland in Eisenstadt; IP ¼ Institute of Palaeontology, University of Vienna (Austria).
M.Krajcarzetal./QuaternaryScienceReviews131(2016)51e7255
3.1.3. Medvedia Cave
Medvedia Cave is located on the south-eastern extremity of the
Glac Plateau within the territory of the Slovenský Raj Mountains,
eastern Slovakia (N 4854056.500 E 2023054.100), with the entrance
at an altitude of 905 m a.s.l. The cave was discovered in 1952 and
the ﬁrst palaeontological investigations were started by Fejfar
(1953) and Janacik and Schmidt (1965). The analysed cave bear
remains (U. ingressus) come from one fossiliferous layer of the cave
part “Cintorín jaskynných medvedov” (“Cave Bears Cemetery”).
This layer has been dated to the MIS 3 period, between 40,000 and
50,000 years BP, as conﬁrmed by AMS radiocarbon dating (Sabol
et al., 2008). Preliminary palaeontological studies have been presented
by Sabol et al. (2008). Apart from remains of cave bears
(forming the majority of all found mammalian fossils), the fossil
record contains also ﬁndings of C. lupus, P. spelaea, G. gulo, and Lepus
sp. Osteological remains of U. arctos, V. vulpes, Caprinae gen. et spec.
indet. as well as birds (Galliformes? gen. et spec. indet.) from Holocene
layers complements the whole faunal assemblage from the
site.
3.1.4. Winden
Winden cave (Widener B€arenh€ohle) is located in the mountains
of Leithagebirge, North Burgenland, about 3 km north of the village
Winden am See, Austria. The cave is situated at 190 m a.s.l. This site
has two entrances and is about 45 m long, 20 m broad and only
1.60 m high. The ﬁrst collection of fossils took place in the cave in
1923. Under the direction of Ehrenberg the palaeontological excavations
were carried out from the year 1929 till 1931. The recovered
fossil material is archived at the Institute of Palaeontology and in
the State Museum of Burgenland in Eisenstadt (D€oppes and
Rabeder, 1997). The sediments were strongly disturbed by fox
dens (from V. vulpes) and therefore the differences between the
layers could not be distinguished. The fossil remains of the large
mammals, consisting of U. ingressus, U. arctos priscus, C. crocuta
spelaea und C. lupus, date back to the Middle Würmian warm period
(MIS 3). However, in the cave they are compounded with the microvertebrate
fossils of the Late Glacial period and the Early Holocene.
One cave bear bone from this site was AMS radiocarbon dated to
38,500 þ450/À430 years BP (Pacher, 2003). For cave bears from
this site genetic studies are available (Hofreiter et al., 2004b;
Rabeder et al., 2008).
3.2. Fossil material
The material includes 35 remains of cave bear from the Polish
sites, 23 remains from the Slovakian site (Table 1), and 12 remains
from the Austrian site. In addition, several bones and tooth of
carnivores and ungulates from the same sites were sampled
(Table 2) to achieve a wider context of ecosystems where the bears
lived.
The sampled remains were bones with exception of two cases of
teeth: the cave bear canine from Nietoperzowa Cave no. IN-01 and
the wolf canine no. MJ-23 from Medvedia Cave. The sample no. IN-
01 was taken from a tooth still attached to its mandible that was
sampled as IN-02. All studied remains represent only adult specimens.
The samples of cave bear bones chosen from Nietoperzowa
and Perspektywiczna Caves were AMS radiocarbon dated (Poznan
Radiocarbon Laboratory, Poland) to conﬁrm the geological age of
the bone assemblages (Table 1).
The palaeogenetic analysis conﬁrmed that bears from Medvedia
Cave and Winden belong to U. ingressus genetic type (Hofreiter
et al., 2004b; Rabeder et al., 2008; Sabol et al., 2008). The
detailed taxonomy of the cave bears from Nietoperzowa and Perspektywiczna
Caves has not yet been investigated. Until now, the
palaeogenetic analysis of cave bears from other sites in Poland
yielded the U. ingressus type (Baca et al., 2012, 2014; Popovic et al.,
2015) for each analysed site. Also only U. ingressus type has been
identiﬁed in the neighbouring territories (Stiller et al., 2014).
Therefore it is very likely that the cave bears from Nietoperzowa
and Perspektywiczna Caves also represent the same taxonomic
unit.
3.3. Isotopic values from Western, Central and Southern Europe
To correlate the results for Central-Eastern European cave bears
with other parts of Europe, we used the previously published results
for Western, Central and Southern Europe. Only sites within
the chronological framework of MIS 3 (about 60,000e25,000 yr BP)
that yielded isotopic results for at least three specimens were
chosen. This comprises 26 sites from different regions of Europe
(Fig.1) within altitudinal range from 125 to 2800 m a.s.l. A complete
list of locations, together with average d13
C and d15
N values of cave
bear collagen and standard deviation is given in Table 3 along with
references. The complete list of isotopic values in specimens from
presented sites is shown in the Appendix. Additional sites with
single isotopic analysis of cave bear remains are also presented in
Appendix.
3.4. Methods of isotopic analysis
For each specimen, a small fragment was carefully sawn with a
dremmel rotating tool equipped with a circular diamond-coated
blade, ultrasonicated in acetone and water, rinsed with distilled
water, dried and crushed to a powder of <0.7 mm grain size
(Bocherens et al.,1997). Then an aliquot of around 5 mg was used to
measure the nitrogen content (%N) of the whole bone, in order to
screen out samples with excessive collagen loss (Bocherens et al.,
2005). For instance, fresh bones contain 4% nitrogen while
ancient bones with less than 0.4% nitrogen usually fail to yield good
collagen (Bocherens et al., 2005). The measurements were performed
using a Vario EL III elemental analyser using Sulfanilic acid
from Merck as internal standard. The mean standard errors were
better than 0.05% for %N.
The collagen was puriﬁed according to a well-established protocol
(Bocherens et al., 1997). The elemental and isotopic measurements
were performed at the Geochemical unit of the
Department of Geosciences at the University of Tübingen (Germany),
using an elemental analyser NC 2500 connected to a
Thermo Quest Delta þ XL mass spectrometer. The elemental ratios
C/N were calculated as atomic ratios. The isotopic ratios are
expressed using the “d” (delta) value as follows:
d13
C ¼ [(13
C/12
C)sample/(13
C/12
C)standard À 1] Â 1000‰, and
d15
N ¼ [(15
N/14
N)sample/(15
N/14
N)standard À 1] Â 1000‰. The international
standards are V-PDB for d13
C values, and atmospheric nitrogen
(AIR) for d15
N values. Measurements were normalized to
d13
C values of USGS 24 (d13
C ¼ À16.0‰) and to d15
N values of IAEA
305 A (d15
N ¼ 39.8‰). Analytical error, based on within-run
replicate measurement of laboratory standards (albumen, modern
collagen, USGS 24, IAEA 305 A), was ±0.1‰ for d13
C values and
±0.2‰ for d15
N values.
The reliability of the isotopic signatures of the extracted
collagen was addressed using their chemical composition (%C, %N,
and C/N ratios). These values must be similar to those of collagen
extracted from fresh bone to be considered reliable for isotopic
measurements and radiocarbon dating. Several studies have shown
that collagen with atomic C/N ratios lower than 2.9 or higher than
3.6 is altered or contaminated and should be discarded, as well as
extracts with %N <5% (e.g., DeNiro, 1985; Ambrose, 1990).
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e7256
3.5. Statistical processing
For statistical analyses only samples dated to MIS 3 were chosen,
to be sure that we compare data derived from the cave bears that
lived during one period. Therefore the results obtained for layers 1
and 9e14 from Nietoperzowa Cave were excluded from the calculations.
For each site the average (arithmetic mean) of d13
C and d15
N
were calculated and all further processing was applied to average
values.
The PAST software (version 2.17b, Hammer et al., 2001) was used
for statistical analysis. To document possible similarities between
the isotopic composition of cave bear populations from European
sites we applied the multivar cluster analysis of average d13
C and
d15
N data from each site (separately altitudinal adjusted data and
raw data) using the Euclidean distance and paired group algorithm,
recommended for environmental data processing (Hammer,
1999e2014). The number of bootstrap replicates Boot. N ¼ 1000
was given. To track the possible relationships between the isotopic
compositions of cave bear remains from different sites and their
geographic distribution, the parametric correlation was used with
Pearson's r coefﬁcient. The signiﬁcance of correlation showing that
the columns are uncorrelated was computed using a two-tailed t
test with n-2 degrees of freedom.
3.6. Adjustment of isotopic data e bone equivalent
Comparing the isotopic values of d13
C and d15
N extracted from
different tissues raises some methodological problems (Bocherens,
2015). The collagen in mammalian bones is constantly remodelled
during the life span and shows average diet isotopic composition of
several years before death (e.g., Hindelang et al., 2002; Huja and
Beck, 2007). The dentine collagen is formed during the tooth
growth, including the moment of milk diet, and that is why the
tooth dentine reveals higher d15
N values than bone (Bocherens
et al., 1994). Before comparing isotopic values of teeth and bones,
an adjustment should be performed to compensate the differences.
In this study, the following adjustment of d13
C and d15
N between
dentine and bone collagen was applied: d13
Cbone ¼ d13
Cteeth þ 0.4‰
and d15
Nbone ¼ d15
Nteeth À 1.9‰ for cave bear (Bocherens, 2015),
and d13
Cbone ¼ d13
Cteeth À 0.2‰ and d15
Nbone ¼ d15
Nteeth À 1.9‰ for
wolf (Bocherens et al., 2011a).
3.7. Adjustment of isotopic data e altitudinal and latitudinal
gradient of isotopes
The isotopic values from different European cave bear sites
come from a wide range of altitudes, and altitude may be a strong
driving factor of variations of isotopic composition. To compare the
isotopic values from sites located at different altitudes and to avoid
considering the differences linked to elevation as if they were differences
in palaeoecology, an adjustment of isotopic values needs
to be applied.
Values of d13
C and d15
N in soils and plants are known to follow
altitudinal gradient, which is a function of among others: air
pressure, air temperature and precipitation changing along
topography (Schuur and Matson, 2001; Amundson et al., 2003; Zhu
et al., 2010). Since the d15
N gradient depends primary on physical
parameters, we assume that the recent value can be used for MIS 3
period, as to our knowledge there is no reason for different altitudinal
gradient of air pressure or temperature to be different during
different periods of the Quaternary. The d13
C gradient derives from
multiple physical, biological and ecological factors, therefore recent
values may differ from those for MIS 3. We therefore used recent
gradient for d13
C values with caution, as we are aware that the real
value may be different for MIS 3, as discussed in more detail in the
discussion section.
In this study we tested the usefulness of the altitudinal correction
using the altitudinal gradient reported for both plants and
herbivorous animals by M€annel et al. (2007): d15
N-adjalt
¼ d15
N þ (0.0011 $ altitude), and d13
C-adjalt
¼ d13
C À (0.0011 $ altitude), where altitude is given in meters.
The correction removes the effect of altitude and allows to equalizing
all data to the same level (i.e., 0 m a.s.l.) to make them comparable.
In the following text the d13
C-adj-alt and d15
N-adj-alt
mean the d13
C and d15
N values adjusted according to altitude. We
also attempted to search for internal altitudinal gradient exhibited
by our dataset.
In contrast with the altitudinal gradient, a latitudinal gradient is
less unequivocal. On the basis of data presented by K€orner et al.
(1991) for plant communities from tropical, temperate and subarctic
zones, a latitudinal gradient 0.0255‰/ from Equator to
North pole for d13
C may be calculated. Du et al. (2015) found a
0.21‰/ gradient for d13
C in modern tree leaves in China, but also
showed strong correlation between d13
C values and air temperature
and precipitation in the same samples, which means that the
observed gradient may be more related to non-latitudinal factors,
such as regional precipitation pattern. Bakwin et al. (1998) reported
decreasing latitudinal gradient of air d13
C values in a 35Ne55N
latitude range, and Suits et al. (2005) simulated a similar gradient in
terrestrial biosphere, at least for the 40Ne50N latitude range, on
the basis of meteorological data. Miller et al. (2003) found no latitudinal
gradient of d13
C values. Craine et al. (2015) showed that
latitudinal gradient of d15
N does not occur in soils. Therefore we
decided against using a latitudinal adjustment of data a priori, but
we explored in this study the possible occurrence of latitudinal
gradient in MIS 3 cave bears.
4. Results
4.1. Preservation and quality of collagen
For the new sites, the N content in whole bones varied among
the analysed samples. The lowest values were obtained from the
samples from Nietoperzowa Cave, with 5 samples (IN-04, IN-19, IN-
20, IN-28, IN-33) excluded from further analysis at this stage
because of their extremely low N content (<0.3%), making the
collagen extraction difﬁcult. All remains from Winden, Perspektywiczna
and Medvedia Caves exhibited sufﬁcient N content in whole
bone.
The extracted collagen of samples from Nietoperzowa Cave
exhibited atomic C/N ratio between 3.2 and 3.4, which is the
standard for collagen (DeNiro, 1985), except for one sample no. IN-
32. The %Ccoll in samples ranged from 6.5 to 43.8%, and the %Ncoll
ranged from 1.6 to 15.6%. Values lower than 15.3% for %C and lower
than 5.5% for %N are beyond the expected range for well-preserved
collagen (acc. to Ambrose, 1990) and such samples (nos. IN-2, IN-
23) were also excluded from further analysis.
In the cases of Winden, Perspektywiczna and Medvedia Caves,
the C/N ratios, %N and %C conﬁrmed good preservation of collagen.
4.2. Isotopic results from new Central-Eastern European sites
The d13
C values ranged from À23.1‰ to À20.5‰ (Nietoperzowa
Cave), from À21.9‰ to 21.0‰ (Perspektywiczna Cave),
from À21.4‰ to À20.7‰ (Medvedia Cave) and from À22.0
to À21.5‰ (Winden) (Table 1). Samples from Perspektywiczna and
Medvedia Caves and Winden showed a narrow range of d13
C values,
while the difference between samples from Nietoperzowa Cave
reaches 2.7‰. However, this difference is visible only in samples
from older sediments, dated to MIS 5e6 (sd ¼ 0.92). Samples from
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e72 57
the MIS 3 period exhibited the maximum difference in d13
C of
0.7‰ (sd ¼ 0.27), similar to the differences observed in Perspektywiczna
Cave (0.9‰, sd ¼ 0.28), Medvedia Cave (0.7‰,
sd ¼ 0.18), and Winden (0.5‰, sd ¼ 0.18) for the same geological
age.
The d15
N values ranged from 1.4‰ to 5.1‰ (Nietoperzowa Cave),
from 1.6‰ to 5.1‰ (Perspektywiczna Cave), 0.8‰e3.7‰ (Medvedia
Cave) and from 1.7 to 4.3‰ (Winden) (Table 1). The d15
N values
were more variable than the d13
C values, even among samples from
one site. In Nietoperzowa Cave the difference in range of d15
N between
the older and younger periods were not as distinct as in case
of d13
C, however there were differences in the average values.
The additional isotopic results obtained for other species are
presented in Table 2. The d13
C values measured for carnivores
ranged from À20.4‰ to À17.7‰ and the d15
N values ranged from
7.4‰ to 10.3‰. The d13
C and d15
N values measured for reindeer
ranged respectively from À19.2‰ to À18.5‰ and from 2.7‰ to
4.5‰. For other herbivores the d13
C and d15
N values ranged
from À20.5‰ to 19.9‰, and from 3.3‰ to 7.7‰.
The pattern of d13
C and d15
N values in cave bears from Poland,
Slovakia and eastern Austria is similar to that previously found in
sites in Western and Central-Western Europe (Bocherens et al.,
2006, 2011a, 2011b; Blant et al., 2010; Perez-Rama et al., 2011;
Horacek et al., 2012; Pacher et al., 2012). Low values of both d13
C
and d15
N place Central-Eastern European bears in a position within
the d13
Ced15
N isospace characteristic for herbivores (Fig. 2). The
position of cave bears in the Central-Eastern European palaeoecosystem
during MIS 3 is clearly the same as it was in Western
Europe, for example in Ardennes and Swabian Jura (Bocherens
et al., 1997, 2011a), Pyrenees and southern France (Bocherens
et al., 1991, 1994) as well as the Swiss Alps (Blant et al., 2010;
Pacher et al., 2012). The range of isotopic fractionation between
the collagen of prey and predator (þ0.8‰eþ1.3‰ for d13
C and
from þ3‰ to þ5‰ for d15
N, according to Bocherens and Drucker,
2003; Bocherens et al., 2011a), allows the reconstruction of the
average prey isotopic signatures for predators. In case of studied
remains there is no evidence for predation of large carnivores on
cave bear in Central-Eastern Europe, as the predators feeding on
cave bear would show lower d13
C than those measured. Individual
specialization in the hunting of cave bear is known for some cave
lions from Belgium and the Swabian Jura based on isotopic studies
(Bocherens et al., 2011a), and for cave hyenas based on taphonomic
investigation of German caves (Diedrich, 2009). In the present
study, none of the analysed predators exhibit low d13
C values that
could indicate an individual prey preference on cave bears in
southern Poland or Slovakia.
4.3. Isotopic signature of European cave bear in geographical layout
The average values of newly analysed sites together with other
European sites (Tables 3 and 4) plotted against the geographical
measurements of the sites, i.e. altitude, latitude and longitude, are
presented on Fig. 3. A clear trend may be observed only in the case
of d15
N plotted against altitude (R2
¼ 0.3603) and indicates a
decrease of d15
N with altitude. This observation stays in accordance
with data known for modern soils, plants and animals (Garten,
1993; Garten and Migroet, 1994; Amundson et al., 2003; Craine
and Lee, 2003; Sah and Brumme, 2003; M€annel et al., 2007). The
linear correlation between d15
N and altitude is strong (correlation
coefﬁcient À0.60026) and is statistically signiﬁcant
(p ¼ 0.00045376) (Table 5). There is no other correlation between
raw average isotopic values and geographical measurements
(Table 5). Moreover, there is no mutual correlation between
geographical measurements of the sites, meaning that there are no
random relationships between the geographical parameters of the
sites and therefore that the geographical locations of the analysed
set of sites has no impact on the statistical processing.
When plotted against altitude, the average d13
C values do not
follow any trend known from the literature and observed in modern
plants and herbivores (Arroyo et al., 1990; K€orner et al., 1988,
1991; M€annel et al., 2007; Zhu et al., 2010). Contrary to the expected
increasing trend, the cave bear data shows a weak
(R2
¼ 0.0801) decrease with altitude (Fig. 3). The known relation
between d13
C and latitude, described for plant communities by
K€orner et al. (1991), is also not visible in the case of cave bears.
4.4. Altitudinal gradient
The strong correlation between d15
N values and altitude (Section
4.3) suggests that an altitudinal adjustment is necessary if one
wants to compare isotopic cave bear data from European sites.
When applied, the adjustment leads to an uniﬁcation of d15
N values
(Fig. 4), with the slope of d15
N/altitude trend changing
from À0.0015 in raw data to À0.0004 in adjusted data (more horizontal
trend on Fig. 4 in comparison with that on Fig. 3). The
altitudinal adjusted d15
N data do not correlate with altitude
(Table 5), what may indicate that the used adjustment is properly
working. However, two Romanian sites (Pes¸ tera cu Oase and
Urs¸ ilor) that were outstanding from raw data, still remain outliers
even when the altitudinal adjustment is applied.
As one may expect from the lack of correlation between d13
C and
altitude (Table 5), the altitudinal adjustment will not work properly
in the case of d13
C. The slope of d13
C/altitude trend is changing
from À0.0002 in raw data (Fig. 3) to À0.0012 in adjusted data
(Fig. 4), which leads to a widening of variability and stands in
contradiction with the concept of data adjusting. The expected
situation after applying altitudinal adjustment is that the
arrangement of points on d13
C-alt-adj/altitude graph follows more
or less a horizontal line, which is not the case with our data. Here,
we observed stronger correlation between altitude and d13
C after
altitudinal adjustment (Table 5), what means the used adjustment
is not leading to expected effect.
4.5. Latitudinal and longitudinal gradient
A weak correlation between raw d13
C values and latitude (Section
4.3) suggests that the latitudinal gradient observed in modern
plant communities and air (K€orner et al., 1991; Bakwin et al., 1998;
Du et al., 2015) might be absent in MIS 3 cave bear populations.
Only a weak decreasing trend in d13
C/latitude plot is observed
(Fig. 3) and is the reverse of the ones reported by K€orner et al.
(1991) or Du et al. (2015), but has similar monotonicity as presented
by Bakwin et al. (1998). A latitudinal gradient for d15
N is also
not visible in our data (Fig. 5). Altitudinal adjusted d15
N data exhibit
a weak decreasing trend with latitude.
The lack of correlation between d13
C and d15
N values and
longitude (Section 4.3) also suggests that longitudinal gradient in
cave bears did not exist. Applying the altitudinal adjustment causes
even lowering of d13
C/longitude relation, and in case of both d13
C
and d15
N the R2
values are very low, regardless if altitudinal
adjusting is applied or not (Fig. 5).
5. Discussion
5.1. Cave bear isotopes and altitude
5.1.1. Altitudinal gradient of d15
N
The values of d15
N is soils and plants depend mainly on aridity
and temperature (Schuur and Matson, 2001; Amundson et al.,
2003; Pardo and Nadelhoffer, 2010; Craine et al., 2015), but also
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e7258
on mineralization of nitrogen in soil (Garten and Miegroet, 1994),
which may be dependent on mineralization rate but also on age
and maturity of soil. Usually d15
N decrease along altitude in
different environments worldwide (Garten, 1993; Amundson et al.,
2003; Craine and Lee, 2003; Sah and Brumme, 2003; M€annel et al.,
2007), however it may be disturbed by local precipitation distribution
and interspeciﬁc response to N supply (Garten, 1993). Altitudinal
gradient in modern ecosystems is known to be variable,
ranging from À0.0007‰/m for domestic cattle (M€annel et al.,
2007), to À0.0012‰/m for African high mountain plants
(Amundson et al., 2003) and À0.0017‰/m for domestic sheep
(M€annel et al., 2007), and to below À0.002‰/m in soils and plants
(Amundson et al., 2003; Craine and Lee, 2003; Sah and Brumme,
2003). For some environments the relatively low altitudinal
gradient was reported, for example À0.0025‰/m for Kyle Canyon,
USA (Amundson et al., 2003), and even À0.0046 to À0.0078‰/m
for grasslands of New Zealand (Craine and Lee, 2003). This probably
reﬂects some interspecies differences (Garten, 1993; Miller and
Bowman, 2002). We should be aware that an absolute value for
the d15
N altitudinal gradient cannot be given with the current state
of knowledge.
We found the decreasing trend of d15
N signature of cave bear
collagen along altitude, which reﬂects the altitudinal gradient
known for modern plants and animals (M€annel et al., 2007). The
strong correlation between d15
N and altitude (Section 4.3) indicates
that altitudinal adjustment of d15
N data is valid and necessary if we
want to ﬁnd ecological differences between sites located at
different elevations. All sites except two of them, namely the
Romanian sites Pes¸ tera cu Oase and Urs¸ ilor, follow the decreasing
trend (Fig. 3). The value of gradient for all sites (À0.0015‰/m) and
for the sites excluding the two outstanding ones (À0.0013‰/m) is
similar to the altitudinal gradient observed in modern grassland
plants and grazers in the Alps (À0.0011‰/m, M€annel et al., 2007). It
is also similar to the gradient detected for a smaller set of cave bear
sites by Bocherens (2015), being À0.0017‰/m. This similarity
supports the use of an altitudinal adjustment based on the gradient
observed by M€annel et al. (2007). However, using of internal gradients
deriving from our results seems to be more accurate for cave
bears.
Factors responsible for the altitudinal gradient of d15
N are
physical parameters, i.e. altitudinal gradients of precipitation and
air temperature (Schuur and Matson, 2001; Amundson et al., 2003).
As absolute values of both precipitation and temperature were for
sure differentiated between regions and periods of Pleistocene,
there are no premises that altitudinal gradients of precipitation and
temperature changed signiﬁcantly through time.
5.1.2. Altitudinal gradient of d13
C
The discussion on altitudinal gradient of d13
C is more complicated
than in the case of d15
N. The altitudinal gradient for d13
C has
been measured for different modern ecosystems and geographical
location, and it is known that the d13
C values tend to increase with
altitude. The rate oscillates between 0.0006‰/m measured for
temperate grassland (K€orner et al., 1991) and high-alpine vegetation
of Andes (Arroyo et al., 1990), and 0.0018‰/m measured for
open habitats of Swiss Alps (Zhu et al., 2010) and 0.0019‰/m reported
for subarctic grassland (K€orner et al., 1991). The average
values for European grassland is 0.0011‰/m according to M€annel
et al. (2007); 0.0012‰/m for high mountain ecosystems according
to K€orner et al. (1988); and 0.0014‰/m for European mountains
according to Zhu et al. (2010). Data given by Kohn (2010) allow
calculating a gradient of 0.0003‰/m as a global average for the
great database of different plant taxa from different environments.
The altitudinal gradient observed in plants is followed by gradient
in herbivore tissues (M€annel et al., 2007), which oscillates between
0.0009, 0.001 and 0.0015‰/m in different species. However, a
reverse altitudinal trend may occur in some types of environments,
for instance in semi-arid areas of southern United States (Van de
Water et al., 2002).
The main driver of the d13
C altitudinal gradient is air pressure
and partial pressure of CO2 in atmosphere (K€orner et al., 1991; Zhu
et al., 2010), but additional factors such as air temperature, precipitation
and hydrological regime also play a role and may disturb
the gradient (Van de Water et al., 2002). However, Schuur and
Matson (2001) did not found a relationship between precipitation
and d13
C in plants in mountain forest. In addition to the physical
drivers, biological and ecological factors also regulate the d13
C
signature in plants, such as the density of leave cover, known as
canopy effect (Van der Merve and Medina, 1991; Fricke, 2007;
Drucker et al., 2008; Bonaﬁni et al., 2013) as well as leaf thickness,
leaf N content, stomatal conductance and stomatal density
(K€orner et al., 1989; Vitousek et al., 1990; Meinzer et al., 1992;
Morecroft and Woodward, 1990; Cordell et al., 1999). These factors
are difﬁcult to recognize in the case of palaeoecosystems, since
the layout of vegetation zones with different taxa and canopy effect,
such as more forested and more open environments, is dependent
on altitude, soil type, regional distribution of biomes, and is
climate-sensitive.
In case of ancient ecosystems, including Pleistocene ones, the
factors responsible for the d13
C altitudinal gradient might occur in
different conﬁgurations and changed through times. It should be
noticed that MIS 3 period, which is the chronological range for our
research material, was not a time of constant climate, but of
numerous climatic ﬂuctuations (Houmark-Nielsen, 2009; Fletcher
et al., 2010; Helmens, 2014). It means that vegetation cover
changed multiple times, what is also conﬁrmed by pollen data (e.g.,
Komar et al., 2009) and climate modelling (Alfano et al., 2003;
Huntley et al., 2003; Willis and van Andel, 2004; Van Meerbeeck
et al., 2011). The partial pressure of CO2 also changed many times
through MIS 3 period (Sigman and Boyle, 2000).
Taken together, the European cave bear d13
C data do not follow
the expected altitudinal gradient (see also Section 4.3 and
Bocherens, 2015) as a whole set. However, if we try to search for
altitudinal trends close to the expected one on a d13
C/altitude
graph, we can point out several groups of points, each of them
following the increasing trend known for modern plants and animals
(Fig. 6). The subdivision of dataset into groups is arbitrary, and
it is based on the tracing of sets of points that are arranged along
increasing trend lines similar to those described in recent ecological
contexts. Assignment of some points to groups may be problematic,
for example in case of Kudaro and A Ceza caves. The most abundant
group (black diamonds on Fig. 6) shows an increasing trend
(gradient 0.0006‰/m), close to the altitudinal gradients 0.0006‰/
m and 0.0009‰/m observed in modern plants and animals (Arroyo
et al., 1990; K€orner et al., 1991, 1998; Hultine and Marshall, 2000;
M€annel et al., 2007) and corresponds to sites from different altitudes
(from 125 to 1600 m a.s.l.) and from different parts of Europe
(Iberian Peninsula, Western, Central and South-Eastern Europe,
Caucasus and all new sites from Central-Eastern Europe). Five sites
clearly stand out from this group: Conturines, Balme a Collomb,
B€arenloch, Drachenloch and Ramesch. All of them are sites situated
in the Alps at high altitudes. The last four show a trend parallel to
the main group (blue circles on Fig. 6). The slope of the altitudinal
gradient of this group (0.0003‰/m) is lower than that of main
group, however the dataset is not numerous. Despite this, the trend
is similar to the altitudinal trend in plants (calculated from the huge
worldwide dataset shown by Kohn, 2010). Conturines alone forms
its own set. Five other sites (red squares on Fig. 6: Aldene, Chauvet,
Divje Babe, Mialet and Font-de-Gaume with Olaskoa) are less
distinctly separated from the main group, but may also be treated
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e72 59
as a separate set. They also show an increasing trend (altitudinal
gradient 0.0008‰/m). These sites are situated in Mediterranean/
Atlantic region, although not all sites from this region tend to follow
the trend of this group (for example Iberian sites or Preletang are
closer to the main group). The difference in d13
C values might be
due to differences in water stress effect on the plants, linked to
aridity.
Such situation may suggest that altitudinal gradient in MIS 3
cave bears existed. However, the whole population was divided into
subpopulations, each of them followed a similar gradient but with
different absolute d13
C values. The differences in d13
C values in cave
bears suggest different d13
C values at the basis of the ecosystems,
i.e. in plants. We can propose two hypotheses explaining this
observation:
1) The bears were feeding on the plants that followed altitudinal
gradient, but were foraging and hibernating at different altitudes.
This hypothesis explains the lower values in some specimens
or populations, with assumption that bear hibernate at
higher elevation, for example to avoid predators (Bocherens,
2015), but were moving to lower elevation to ﬁnd more abundant
plant food. However, in the case of our data this explanation
is insufﬁcient, since differences should be recorded both in
d13
C and d15
N values. The cave bear exhibiting relatively low
d13
C values (for example blue circle group or Conturines on
Fig. 6) in comparison with the main group, show the same d15
N
pattern as the main group. In other words, several trend lines in
d13
C along altitude suggest that particular populations might
have been feeding at different altitude than hibernating, while
one trend line in d15
N along altitude indicates that each population
had been feeding and hibernating at the same altitude.
This contradiction falsiﬁes this ﬁrst hypothesis in case of MIS 3
cave bears treated globally. However, this explanation may be
correct in the case of some sites, for example Conturines (see
also Bocherens, 2015). This site exhibits both d13
C values below
the main trend and d15
N values above the main trend. This
suggests that feeding and hibernating at different altitudes may
be at least partially responsible for the relatively low d13
C values.
This hypothesis seems reasonable if we take into consideration
the unusually highly elevated hibernating place, as Conturines
cave is situated at 2800 m a.s.l.
2) The different bear populations were feeding on plants with
different d13
C signatures, related to factors other than altitudinal
gradient. This situation is possible when:
a. Animals are feeding on different types of plants. It is widely
known that different photosynthetic groups of plants, such as
C3, C4, CAM or lichens, exhibit different d13
C signature (e.g.,
DeNiro and Epstein, 1978; Bocherens, 2003), and also
different genera growing at the same altitude may exhibit
different d13
C values and different altitudinal gradient (Zhu
et al., 2010);
b. Animals inhabited environments with different canopy effects,
i.e. with different density of vegetation. It is known that
specimens of one plant taxon may exhibit different d13
C
signature if living in different environments (van der Merwe
and Medina, 1991; Bonaﬁni et al., 2013);
c. Observed populations are chronologically distant, i.e. they
might have been living during periods with different layout
of physical parameters, such as air temperature, precipitation,
CO2 partial pressure. During the long time span represented
by the analysed sites, covering most of the MIS 3, numerous
environmental shifts occurred (e.g., Sigman and Boyle, 2000;
Fletcher et al., 2010; Helmens, 2014). Therefore cave bears
populations with different altitudinal trends may represent
different ecosystems or periods with different physical parameters,
which were changing in time.
All these situations a, b and c are connected to the occurrence of
different environmental conditions (Fig. 7). This may be related to
either altitudinal vegetation zones, different biomes or different
pattern of physical parameters not related to ecological differences,
such as CO2 partial pressure. These differences followed environmental
variations either in geographical regions or chronological
periods. All these situations are expected to give similar isotopic
pattern and may not be differentiated with isotopic methods.
Looking more regionally at our data we could try to ﬁnd isotopic
differences between elevation zones inside one region (what might
indicate the impact of altitude related vegetation zones). The best
region for comparing bears from different altitudes among our
research area are the Alps, thanks to the great altitudinal differences
occurring there. To the Alpine region we can include the
following sites: Balme a Collomb, B€arenloch, Conturines, Drachenloch,
Gamssulzen, Preletang, Ramesch and Tischoferh€ohle.
Sites located at elevations below and above ca. 1500 m a.s.l. follow
different trend lines and may be easily separated into separate
groups (Fig. 8). The sites from lower elevation exhibit higher d13
C
values, which may be related to more open habitats. Higher
elevated sites represent relatively more C3 plants in diet or stronger
canopy effect. Such situation seems to be unusual in modern Alps,
Table 2
List of additional species fossil samples from Polish and Slovakian sites and of isotopic results (* e bone equivalent, coll e data for extracted collagen, collection symbols are the
same as those used at Table 1).
Lab. no. Collection Site Species Bone/tooth %Ccoll %Ncoll C/Ncoll d13
C d15
N
IP-07 IGS Perspektywiczna cave Crocuta crocuta spelaea Cranium 43.2 15.6 3.2 À19.2 9.4
IP-21 IGS Perspektywiczna cave Crocuta crocuta spelaea Femur 39.9 14.1 3.3 À19.3 10.3
IP-22 IGS Perspektywiczna cave Crocuta crocuta spelaea Ulna 34.1 11.9 3.3 À19.5 9.1
IP-23 IGS Perspektywiczna cave Panthera spelaea Metacarpus IV 42.7 15.1 3.3 À17.7 7.4
IP-08 IGS Perspektywiczna cave Rangifer tarandus Metacarpus 44.0 15.9 3.2 À18.7 3.6
IP-10 IGS Perspektywiczna cave Rangifer tarandus Phalanx 2 37.0 13.4 3.2 À19.2 2.8
IP-24 IGS Perspektywiczna cave Rangifer tarandus Phalanx II 42.2 14.9 3.3 À19.0 2.7
IP-25 IGS Perspektywiczna cave Rangifer tarandus Phalanx I 39.6 13.9 3.3 À18.7 2.9
IP-26 IGS Perspektywiczna cave Rangifer tarandus Metacarpus 42.6 14.9 3.3 À18.5 4.5
IP-11 IGS Perspektywiczna cave Megaloceros giganteus Metacarpus 32.4 11.1 3.4 À19.9 4.2
IP-12 IGS Perspektywiczna cave Bos/Bison Mandibula 41.3 14.8 3.3 À20.1 4.6
IP-27 IGS Perspektywiczna Cave Bos/Bison Scapula 44.4 15.6 3.3 À20.4 7.7
IP-20 IGS Perspektywiczna Cave Coelodonta antiquitatis Metapodial 38.1 13.3 3.3 À20.5 3.3
MJ-21 SM Medvedia cave Panthera spelaea Phalanx 39.9 13.8 3.4 À19.2 8.4
MJ-22 SM Medvedia cave Canis lupus Vertebra 40.5 14.2 3.3 À19.3 8.5
MJ-23 SM Medvedia cave Canis lupus Canine 38.1 13.5 3.3 À20.2 * À20.4 10.6 * 8.7
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e7260
Table 3
List of Western and Southern European cave bear sites dated to MIS 3 and average isotopic values. Adjusted d13
C-adj-alt and d15
N-adj-alt with use of altitudinal gradients given by M€annel et al. (2007).
Site Site
symbol
Region Altitude
m a.s.l.
Latitude

N
Longitude

E
Palaeogenetics n d13
C (av ± sd) d15
N (av ± sd) d13
C-adj-alt (av) d15
N adj-alt (av) Reference for isotopic data
A Ceza AC Spain 1004 42,691108 À7,100708 e 10 À21.7 ± 0.2 3.2 ± 1.2 À22.8 4.3 Perez-Rama et al. (2011)
Aldene (Grotte d'Aldene) Al France 270 44,117908 2,674142 e 5 À20.9 ± 0.4 3.8 ± 1.4 À21.1 4.1 Bocherens et al. (1994);
Bocherens (2015)
Balme a Collomb BC French Alps 1700 45,689632 5,816232 spelaeus
(Orlando et al., 2002)
6 À21.4 ± 0.7 1.5 ± 0.5 À23.3 3.4 Bocherens et al. (2011b)
B€arenloch B€a Switzerland 1645 47,068951 7,925607 - 7 À21.7 ± 0.9 0.2 ± 1.3 À23.6 2.0 Blant et al. (2010)
Chauvet Ch France 240 44,839072 3,81672 spelaeus
(Baca et al., 2012)
26 À20.6 ± 0.3 4.0 ± 0.9 À20.9 4.2 Bocherens et al. (2006);
Bon et al. (2011)
Cioclovina Ci Romania 770 45,949955 23,042794 e 30 À21.2 ± 0.6 3.0 ± 1.2 À22.0 3.9 Robu et al. (2013)
Conturines Co Italy 2800 46,994069 11,92463 ladinicus
(Hofreiter
et al., 2002)
13 À22.3 ± 0.6 1.6 ± 1.3 À25.4 4.7 Horacek et al. (2012)
Cova Eiros Ei Spain 780 42,763562 À7,20534 spelaeus
(Stiller et al.,
2010)
36 À21.2 ± 0.5 4.5 ± 0.8 À22.0 5.4 Perez-Rama et al. (2011)
Cova Li~nares Li Spain 1115 42,695681 À7,074094 spelaeus
(Loreille
et al., 2001)
23 À21.0 ± 0.3 1.9 ± 0.5 À22.2 3.1 Vila Taboada (1998);
Vila Tabaoda et al. (1999);
Perez-Rama et al. (2011)
Divje Babe DB Slovenia 450 46,497209 13,902169 ingressus
(Baca et al., 2012)
14 À20.5 ± 0.4 1.9 ± 0.7 À21.0 2.4 Nelson and Ku (1997);
Nelson et al. (1998)
Drachenloch Dr Switzerland 2475 47,233323 9,44172 e 3 À21.3 ± 0.4 0.3 ± 0.6 À24.0 3.1 Pacher et al. (2012)
Font-de-Gaume
III þ Olaskoa
FG France 140 45,73566 0,828439 spelaeus
(Stiller et al., 2014)
4 þ 1 À20.7 ± 0.4 2.8 ± 0.6 À20.8 3.0 Bocherens et al. (2011b)
Gamssulzen Ga Austria 1300 47,717845 14,305917 ingressus
(Hofreiter et al., 2004b)
9 À20.8 ± 0.2 1.1 ± 0.4 À22.3 2.6 Bocherens et al. (2011b),
Hofreiter et al. (2004b)
Geißenkl€osterle Ge Germany 580 48,409605 9,770841 spelaeus þ ingressus
(Baca et al., 2012)
20 À20.9 ± 0.3 2.8 ± 0.9 À21.5 3.5 Münzel et al. (2011)
Goyet Go Belgium 130 51,095544 4,959298 spelaeus (Stiller et al.,
2014)
23 À21.5 ± 0.4 4.1 ± 1.0 À21.7 4.3 Bocherens et al. (2011a)
Grotte du
Renne (Arcy-sur-Cure)
Re France 125 47,591539 3,766046 ladinicus Frischauf et al.,
2010)
14 À21.3 ± 0.4 3.4 ± 0.6 À21.4 3.5 Bocherens (2015)
Hohle Fels HF Germany 534 48,389546 9,755048 spelaeus þ ingressus
(Baca et al., 2012)
22 À20.9 ± 0.3 3.4 ± 1.0 À21.5 4.0 Münzel et al. (2011)
Kudaro Ku Caucasus 1600 43,451671 43,6092 kudarensis (Baca et al.,
2012)
7 À20.2 ± 0.4 2.4 ± 0.8 À22.0 4.2 Bocherens et al. (2014)
Mialet Mi France 300 44,49528 3,904611 e 5 À20.2 ± 0.5 2.8 ± 1.6 À20.5 3.2 Bocherens et al. (1994)
Muierii (Pes¸ tera Muierilor) Mu Romania 585 45,535933 23,745919 e 4 À21.0 ± 0.1 4.0 ± 0.3 À21.6 4.6 Robu et al. (2013)
Pes¸ tera cu Oase Oa Romania 600 45,474335 21,812326 ingressus (Richards
et al., 2008)
74 À21.4 ± 0.4 7.9 ± 1.4 À22.1 8.5 Robu et al. (2013);
Trinkaus and Richards
(2013)
Preletang Pr France 1225 45,086853 5,428964 ladinicus (Orlando
et al., 2002; Frischauf
et al., 2010)
7 À21.0 ± 0.8 1.9 ± 0.9 À22.3 3.3 Bocherens (2015)
Ramesch Ra Austria 1960 47,699363 14,242746 spelaeus (Hofreiter
et al., 2004b)
11 À21.7 ± 0.4 0.6 ± 0.7 À23.9 2.8 Bocherens et al.
(2011b), Hofreiter et al.
(2004b)
Scladina Sc Belgium 138 50,50906 5,023374 spelaeus (Orlando
et al., 2002)
7 À22.1 ± 0.2 4.9 ± 1.2 À22.2 5.1 Bocherens et al. (1997)
Tischoferh€ohle Ti Austria 594 47,5925 12,196278 e 9 À21.0 ± 0.3 2.9 ± 0.8 À21.6 3.5 Sp€otl et al. (2014)
Urs¸ ilor Ur Romania 482 46,889057 22,515451 e 28 À21.7 ± 0.5 7.4 ± 1.9 À22.2 7.9 Robu et al. (2013)
M.Krajcarzetal./QuaternaryScienceReviews131(2016)51e7261
where forested area is situated at lower elevations, and grasslands
at higher position. However, in more arid regions such as the
southern United States where lower parts of mountains are occupied
by open environment of shrublands and higher elevations are
forested, the d13
C values in higher elevated plants are lower (Van de
Water et al., 2002).
However, we may try to look at the same situation visible on
Fig. 8 in relationship with chronology. Cave bear assemblages from
high Alpine sites mostly yielded open radiocarbon dates
(Appendix), a fact that suggests a chronology corresponding to an
older part of MIS 3. The lower Alpine sites provided younger
radiocarbon ages, about 31e40 thousand years BP for Tischoferh€ohle
and 37e47 thousand years BP for Gamssulzen (Appendix),
i.e. indicating a younger part of MIS 3. The chronological distance
may reﬂect occurrence of different ecosystems or periods with
different conditions, such as CO2 pressure, which may be a reason
for isotopic dissimilarity in case of these sites. Despite this, each
group follows an altitudinal gradient, but with different d13
C
baselines.
5.1.3. To use or not use d13
C altitudinal adjustment for cave bears?
The d13
C altitudinal trends exhibited by several groups of cave
bear sites (Fig. 6) are in our opinion a strong hint in favour of using
some altitudinal adjustment. If an altitudinal adjustment was not
applied, then the differences in d13
C values related to altitude that
occur inside groups would be included in the variation patterns. As
a consequence, identifying sites where cave bears were ecologically
different might become difﬁcult. For example, the site-to-site differences
exceed in numerous cases the intra-site variability for the
main group of cave bears deﬁned here (black diamonds on Fig. 6). In
such a situation the comparison of isotopic signatures between
sites for palaeoecological or physiological studies would show
differences, which in fact are an effect of elevation differences,
rather than being related to ecological or physiological differences.
Therefore using an adjustment for altitude is recommended to sort
out difference patterns linked to ecological differences among cave
bears.
The remaining question is which value of altitudinal gradient is
appropriate to be used in altitudinal adjustment in the case of cave
bears. The internal trends exhibited by groups on Fig. 6 are characterised
by less steep slopes than observed by M€annel et al. (2007)
(0.0003, 0.0006 and 0.0008‰/m against 0.0011‰/m). On the basis
of our data we can accept internal altitudinal gradient and use it for
altitudinal adjustment. On the Fig. 9 we compared the effects of
altitudinal adjustment made with use of M€annel et al. (2007)
gradient and cave bear internal gradient. Here we used a
0.0006‰/m gradient, which is the gradient exhibited by the most
numerous group (black diamonds on Fig. 6). The adjustment is used
with equation: d13
C-adj-alt ¼ d13
C À (0.0006 $ altitude), where
altitude is given in meters.
We also tried to use separate gradients for particular groups
(0.0003 for blue circles, 0.0006 for black diamonds and 0.0008‰/m
for red squares), as shown in Appendix. However, in our opinion
gradients derived from less numerous groups are less reliable since
these groups are not statistically representative. In addition, using
different adjustments for different groups may eliminate the nonaltitude-related
diversity, which is an unwanted effect. For
example, the main group would incorporate the ‘high Alpine’ group
because of stronger lowering of the main group due to its steeper
gradient. Such a homogenisation of two groups would eliminate
the differences not related to altitude, as sites from ‘high Alpine’
group (such as B€arenloch and Balme a Collomb) have signiﬁcantly
lower d13
C signature than sites from the main group that are
located at similar altitude (such as Gamssulzen and Kudaro).
5.2. Cave bear isotopes and latitude
We did not ﬁnd evidence for signiﬁcant impact of latitude on the
cave bear isotopic signature. This may be because the cave bear
sites in Europe are distributed in a relatively narrow latitudinal
range. We can expect that in the case of European cave bear sites
altitude may play a more important role than latitude in the impact
on isotopic values. The maximum altitudinal distance among analysed
sites is 2675 m (Grotte du Renne as the lowest site e
125 m a.s.l., Conturines as the highest e 2800 m a.s.l.), which corresponds
to a 2.94‰ difference in d13
C if using 0.0011‰/m altitudinal
gradient according to M€annel et al. (2007); while the
maximum latitudinal distance is only 8.4 (A Ceza as most southern
site e 42.691N, Goyet as the most northern one e 51.096N),
which represents only the 0.21‰ difference in d13
C if using
0.0255‰/ latitudinal gradient according to data by K€orner et al.
(1991), which is less than the intra-site variability for almost each
analysed site, or 1.76‰ if using 0.21‰/ latitudinal gradient according
to data by Du et al. (2015). After altitudinal adjustment of
d13
C data the latitudinal gradient is even less visible (À0.0209‰/,
R2
¼ 0.0022, Fig. 5).
If we tried to analyse the impact of latitudinal adjustment on
d13
C average values, we can observe mainly a shift of absolute
values, but the relations between sites remain almost unchanged. It
may be observed for example in the impact of latitudinal adjustment
according to latitudinal gradient by K€orner et al. (1991), on
d13
C/altitude relation (Fig. 10). This indicates that irregularities
observed in altitudinal pattern of d13
C values are not an effect of
latitudinal impact. This also indicates that the impact of latitude
may be ignored in comparison of d13
C signature of cave bear
collagen across Europe.
5.3. Cave bear isotopes and longitude
Climatic differences between Western and Eastern Europe are
visible today, and were also marked during MIS 3 (Kjellstr€om et al.,
2010), the most important differences being in the mean temperature
of winter and precipitation, as derived from pollen data and
climate modelling (Van Meerbeeck et al., 2011). The climatic shift
from West to East was also followed by differences in vegetation
(Alfano et al., 2003; Huntley et al., 2003; Willis and van Andel,
2004; Komar et al., 2009). Despite this, the isotopic values of cave
bears from the West to the East of Europe do not seem to follow
Fig. 2. Bone collagen d13
C/d15
N diagram of cave bear and other fossil animals from MIS
3 deposits of Central-Eastern European sites.
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e7262
longitudinal gradient. This suggests that the cave bear isotopic
ecology was independent from climatic WeE shift, for example
because cave bear fed on the same plants that occurred in different
biomes. It should be noticed, however, that MIS 3 was not a time of
constant climate, but of numerous climatic ﬂuctuations of
millennial duration (Fletcher et al., 2010; Helmens, 2014), when
minor events of glacial advance occurred (Houmark-Nielsen, 2009).
We cannot exclude the scenario in which cave bear populations
followed the certain climatic-vegetation zones that moved across
Europe during the climatic turnovers; and particular populations,
Table 4
Average values and standard deviation of d13
C and d15
N in analysed sites from Central-Eastern Europe (new data), with d13
C and d15
N adjusted to altitude with use of altitudinal
gradient given by M€annel et al. (2007).
Site Site
symbol
Region Altitude m
a.s.l.
Latitude

N
Longitude

E
Palaeogenetics n d13
C
(av ± sd)
d15
N
(av ± sd)
d13
C-adj-alt
(av)
d15
N-adj-alt
(av)
Winden MIS 3 Wi Austria 190 47,975515 16,748062 ingressus (Hofreiter et al., 2004b;
Rabeder et al., 2008)
12 À21.8 ± 0.2 3.2 ± 0.8 À22.0 3.4
Nietoperzowa Ni Poland 438 50,194317 19,774488
Nietoperzowa (all) e 18 À21.4 ± 0.6 2.9 ± 0.9 À21.9 3.3
Nietoperzowa
(MIS 3)
ingressus ? 10 À21.4 ± 0.4 2.9 ± 1.1 À21.9 3.4
Nietoperzowa
(MIS 4)
? 1 À21.2 3.6 À21.7 4.1
Nietoperzowa
(MIS 5e6)
? 6 À21.5 ± 0.9 2.6 ± 0.8 À21.9 3.1
Perspektywiczna
(MIS 3)
Pe Poland 345 50,442602 19,766957 ingressus ? 10 À21.4 ± 0.3 3.1 ± 1.1 À21.7 3.4
Medvedia MIS 3 Me Slovakia 905 48,915863 20,398344 ingressus (Sabol et al., 2008) 20 À21.0 ± 0.2 1.9 ± 0.7 À22.0 2.9
Fig. 3. Raw isotopic values (diamonds e average for sites, bars e standard deviations) plotted against geographical factors: altitude, latitude, longitude of sites.
Table 5
Linear correlation matrix of mean values of d13
C and d15
N in bone collagen of cave bears from MIS 3 European sites and geographical localization of these sites. Correlation
coefﬁcients over 0.5 or below À0.5 are bolded.
d13
C d13
C-adj-alt d15
N d15
N-adj-alt Altitude Latitude Longitude
d13
C 0 1.7657E-05 0.63733 0.15308 0.12971 0.044626 0.50887
d13
C-adj-alt 0.69851 0 0.026768 0.97556 9.4848E-11 0.8068 0.91364
d15
N À0.08971 0.40413 0 2.2414E-11 0.00045376 0.89377 0.61818
d15
N-adj-alt À0.26744 0.0058425 0.89593 0 0.33432 0.49164 0.34812
Altitude À0.28298 ¡0.88401 ¡0.60026 À0.18253 0 0.34461 0.564
Latitude À0.36926 À0.046607 À0.025458 À0.13057 À0.17875 0 0.17141
Longitude 0.12546 À0.020678 0.094823 0.17747 0.10967 0.2564 0
Correlation values are given in the lower triangle of the matrix, and the two-tailed probabilities that the columns are uncorrelated are given in the upper.
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e72 63
distinct in terms of both geography and time, lived in similar
habitats. The isotopic record of d13
C/d18
O in cave bear enamel,
which does not change signiﬁcantly between layers accumulated
during warm and cold periods in long proﬁles of Bisnik Cave and
Grotte d’Aldene (Bocherens et al., 1991; Krajcarz and Krajcarz,
2014), suggests that cave bear was present in particular areas
only when environmental conditions were suitable. Our data may
support this lack of ecological variability in cave bear. Whatever the
reason, it caused the homogenization of cave bear isotopic
composition in Europe.
5.4. Isotopic similarity between cave bears from European sites
We did not ﬁnd any signiﬁcant relation between isotopic
signature of cave bears and latitude or longitude. It suggests that
cave bear population during MIS 3 period was not ecologically (or
strictly speaking, dietary) altered from South to North, nor from
West to East. However, cave bear isotopic data gathered from
numerous European sites exhibit inter-site variation.
5.4.1. Geographical variability of cave bear isotopic signature
To track the similarities/dissimilarities between cave bear sites,
a cluster analysis was applied. Both carbon and nitrogen isotopic
values averaged for sites were used. There are several possibilities
in data selection: to compare raw d13
C and d15
N values, or d13
C-adjalt
and d15
N-adj-alt with altitudinal adjustment applied according
to different altitudinal gradients. It is also possible to combine
different types of data. However, not all of them seem to be justiﬁed.
The raw d13
C and d15
N values allow us to recognize the general
variability, including altitude-dependant differences (Fig. 11a).
Because the altitudinal gradient occurs, the most reasonable way
for palaeoecological application is to use only altitudinally adjusted
data. Being aware that choosing the absolute values of altitudinal
gradient for extant species and past environments is imperfect on
the basis of current state of research, we present two variants.
Fig. 11b and c show clustering of data adjusted using altitudinal
gradient of M€annel et al. (2007), and adjusted using internal gradients
of cave bears (À0.0013‰/m for d15
N and 0.0006‰/m for
d13
C, see also discussion in Section 5.1.2).
Independently of data selection, two Romanian sites (Pes¸ tera cu
Fig. 4. Raw average isotopic values (open diamonds) and altitudinal adjusted (dark diamonds) plotted against altitude. The expected trend after adjustment should be horizontal in
case of data following altitudinal gradient.
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e7264
Fig. 5. Raw average isotopic values (open diamonds) and altitudinal adjusted (dark diamonds) plotted against latitude (upper diagrams) and longitude (lower diagrams) of sites.
Fig. 6. d13
C (raw data) plotted against altitude. Knowing the expected altitudinal gradient (grey lines e different trends according to different authors; note that only trend directions
are shown here, not absolute values), we may try to divide data into several groups (marked by squares, diamonds, circles and star), each of them showing a trend similar to
expected altitudinal gradient.
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e72 65
Oase and Urs¸ ilor) always form an outstanding cluster. These sites
are famous for their unusually high d15
N values (Richards et al.,
2008; Robu et al., 2013; Bocherens, 2015). Reconsidering Romanian
data in terms of altitudinal gradient of isotopes (Fig. 11b,c),
these sites still remain the most isotopically distinct in Europe.
Addition of new sites from Central-Eastern Europe, which are
relatively geographically close to the Romanian ones, did not reveal
similarities between them and Pes¸ tera cu Oase and Urs¸ ilor.
All other sites are grouped together, but can also be divided into
subclusters. The arrangement of these subclusters changes along
different adjusting treatment. Some general patterns however
appear. A distinct group of bears in terms of stable isotopes are
those from the high Alpine region (Fig. 11aec). Remains of bears
found in Alpine caves situated over 1500 m a.s.l. (Conturines,
Ramesch, Drachenloch, B€arenloch, Balme a Collomb) are characterized
by low to extremely low d15
N values and average to low d13
C
values. The most elevated cave bear site, Conturines (2800 m a.s.l.),
represents the most extreme case of Alpine isotopic signature, with
the lowest d13
C values in Europe. The second distinct group is
formed by Scladina, Cova Eiros and A Ceza, all situated in Western
Europe. These sites are characterized by relatively high d15
N values
and extremely low d13
C values. They tend to cluster together if
altitudinal adjustment applied (Fig. 11b and c), which means that
isotopic signature in this case is strongly altitude-related, although
Scladina is distinct from other sites even if raw data are analysed.
All other sites form one wide cluster (Fig. 11aec). The statistical
similarity of these sites becomes slightly rearranged along different
adjusting treatment, however the isotopic similarity remains. Cave
bears of these sites lived in lowland and upland regions, ranging
from the westernmost to easternmost fringes of Europe (Fig. 11:
maps). The sites corresponding to this group are located at different
altitudes, from 125 to 1600 m a.s.l.
The group of Mediterranean/Atlantic sites that differs in d13
C
values (red squares on Figs. 6 and 9) is less distinct if both carbon
and nitrogen values are analysed together (Fig. 11). If altitudinal
adjustment is applied, three of these sites (Divje Babe, Mialet and
Font-de-Gaume with Olaskoa) tend to cluster together, however
still remain similar to most of European sites.
5.4.2. Isotopic variability and cave bear taxonomy
The isotopic variability of cave bears does not seem to be related
to taxonomic position, although such relation was observed on the
local scale, between particular pairs of sites (Bocherens et al.,
2011b; Bocherens, 2015). The various genetic forms, named U.
kudarensis, U. spelaeus, U. ingressus and U. ladinicus, are scattered
among different clusters. Cave bear from Conturines, included to
U. ladinicus species (Hofreiter et al., 2002), seems to represent the
isotopically distinct form (Fig. 11). However, other sites with cave
bear identiﬁed as U. ladinicus (Grotte du Renne and Preletang, according
to Orlando et al., 2002; Frischauf et al., 2010) do not tend to
cluster together with Conturines, which does not allow us to treat
U. ladinicus as an ecologically distinct taxon. The most genetically
distinct form, U. kudarensis from Caucasus (Baca et al., 2012), is
characterised by relatively high d15
N values (Bocherens, 2015), but
altitudinal correction reveals the isotopic similarity of this bear to
both U. spelaeus and U. ingressus from different parts of Europe
(compare Fig. 11b and c and Table 3).
Fig. 7. Scheme showing the possible mechanisms responsible for different values of
d13
C in different populations: a e all animals inhabit one vegetation zone and follow
the same altitudinal gradient; b e animals from regions occupied by different biomes
or with different conditions exhibit different d13
C base; c e animals from elevations
occupied by different altitudinal vegetation zones exhibit different d13
C base; in situations
b and c the altitudinal gradient is the same in each population, but the absolute
values are shifted.
Fig. 8. d13
C average values of Alpine sites plotted against altitude. Two subpopulations
may be found, corresponding to lower and higher elevations. Conturines may represent
the high Alpine group (explanation in text).
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e7266
5.4.3. Central-Eastern European cave bears and unusual Romanian
isotopic pattern
All newly studied Central-Eastern European sites (Medvedia
Cave, Nietoperzowa Cave, Perspektywiczna Cave and Winden) are
clustering together independently of the adjusting treatment. After
altitudinal adjustment these site become even more similar
(Fig. 11b and c ), which indicates very similar palaeoecology of bear
populations from these sites, and most of observed differences
(Fig. 11a) are caused by elevation. Central-Eastern European sites
exhibit isotopic similarity to most sites of Central, Western and
Southern Europe and Caucasus, but are clearly distinct from both
mentioned Romanian sites and group of high Alpine sites.
The newly added Central-Eastern European sites ﬁlled the gap in
a map of Europe, and are situated relatively closely to two unusual
Romanian sites, Pes¸ tera cu Oase and Urs¸ ilor. However, isotopic
signatures of cave bears from Austria, Poland and Slovakia are
clearly far from Romanian cave bears with unusually high d15
N
values. New data had emphasised the isotopic uniqueness of
Romanian sites.
The attempt to explain the isotopic isolation of Romanian sites is
beyond the scope of this paper. The hypothesis that this cave bear
population had a more omnivorous diet, including meat and/or
ﬁsh, was presented by Richards et al. (2008). The Romanian cave
bear d13
C and d15
N signatures were presented as overlapping on
those of modern omnivorous brown bears (Robu et al., 2013).
Bocherens et al. (2014b) also discussed this problem and showed
that the most herbivorous modern brown bears overlap with the
Romanian cave bears, while more carnivorous modern brown bears
are clearly different from the Romanian cave bears. In addition,
these Romanian cave bears overlap only with two types of Pleistocene
herbivores: the woolly mammoths and the fallow deer
(Bocherens, 2015). Another possibility is that these bears still had a
herbivorous diet, but that they relied on a different type of plant
food than the other cave bear populations, more similar to that
consumed by mammoth or fallow deer (Bocherens, 2015). Instead
of a carnivorous diet, these cave bears with high d15
N may have
recorded in their bone collagen the isotopic signature of hibernation
(Grandal d'Anglade and Fernandez Mosquera, 2008). The
Romanian cave bears with unusual nitrogen isotopic signature
seem to follow the general pattern presented by FernandezMosquera
et al. (2001), according to which the higher d15
N values
may have been connected with longer time of hibernation during
cool periods of Pleistocene. However, since other cave bear populations
that lived during the same time interval (MIS 3) do not
Fig. 9. Altitudinal adjusted d13
C average values of sites plotted against altitude. Adjustments
use altitudinal gradients: a e 0.0011‰/m reported for modern Alpine plants
and grazers by M€annel et al. (2007); b e internal gradient 0.0006‰/m calculated for
main group of cave bear sites (black diamonds at Fig. 6). Symbols are the same as used
in Fig. 6.
Fig. 10. Impact of latitudinal adjustment on the altitudinal gradient exhibited by d13
C
average values of sites. Open symbols e raw values; black symbols e values adjusted
with use of 0.0255‰/ latitudinal gradient according to K€orner et al. (1991).
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e72 67
Fig. 11. Results of cluster analysis of the isotopic similarity of MIS 3 cave bears sites with regards to average values of collagen carbon and nitrogen isotopic signature, supplemented with carbon/nitrogen average ratios plots and map of
the distribution of clusters in Europe: a e d13
C and d15
N raw values (coph. corr. ¼ 0.8785); b e d13
C-adj-alt and d15
N-adj-alt adjusted with use of altitudinal gradients 0.0011‰/m for d13
C and À0.0011‰/m for d15
N according to M€annel
et al. (2007) (coph. corr. ¼ 0.9211); c e d13
C-adj-alt and d15
N-adj-alt adjusted with use of internal altitudinal gradients 0.0006‰/m for d13
C and À0.0013‰/m for d15
N (coph. corr. ¼ 0.9267).
M.Krajcarzetal./QuaternaryScienceReviews131(2016)51e7268
exhibit such signature, a question arises: why would some bears
have stronger impact of physiology than others? It is possible that
not only climate controlled the duration of hibernation. The duration
of winter sleep also depended on biological factors, as it is in
case of modern bears (Seryodkin et al., 2003; Manchi and Swenson,
2005; Baldwin and Bender, 2011).
In the light of analysis of geographic variability of cave bear
isotopic signature presented in this study, the locations of Romanian
sites with high d15
N values sites are not connected to latitude
or longitude. Although the cave bear isotopic signature is driven by
altitude, the altitudinal adjustment of isotopic data is not enough to
explain the isotopic dissimilarity of these cave bears. Further investigations
are required to understand the reason for such isotopic
isolation. Especially the analysis of d13
C and d15
N values of more
sites located to the south of Romanian ones are necessary, to widen
the regional context of these unusual sites.
6. Conclusion
In this study the results of new isotopic analysis of cave bear
collagen from four Central-Eastern European sites were presented
and the spatial variation of cave bear collagen isotopic values in
Europe was discussed.
This study demonstrates that there is no geographical East-West
and South-North pattern of d13
C and d15
N values of cave bear
collagen in assemblages dated to MIS 3. The distribution of isotopic
signature does not vary with relation to longitude nor latitude. This
means that the shift from oceanic climates in Western Europe to
more continental climates in the eastern part of the continent was
not a main agent affecting the cave bear ecology during MIS 3. The
cave bears from Central-Eastern Europe exhibit the d13
C and d15
N
values near the middle of the range typical for most of the cave
bears from Central, Western and Southern Europe. This isotopic
homogeneity may either reﬂect the climatic and vegetational uniﬁcation
of ecosystem, which stays in contradiction to palaeoclimatic
data, or may be a characteristic of cave bear, for example
due to low ecological ﬂexibility. This phenomenon needs further
study and has to be also investigated in the case of other European
mammalian species during MIS 3.
Most of the cave bear sites in Europe follow an altitude-related
isotopic pattern. Altitudinal gradient À0.0013‰/m of d15
N values is
followed by most of sites except of two Romanian sites Pes¸ tera cu
Oase and Urs¸ ilor, exhibiting unusually high d15
N signature. In case
of d13
C values the expected increasing altitudinal gradient is visible
in several groups of sites, but these sets exhibit the group-to-group
shift of absolute values. The most abundant group of sites reveals
0.0006‰/m internal gradient of d13
C values. The other groups of
sites also exhibit increased altitudinal gradient, however with absolute
values shifted. The most distinct of such groups are high
Alpine sites situated over 1500 m a.s.l.
The unusual nitrogen isotopic values of Romanian cave bears
from Urs¸ ilor and Pes¸ tera cu Oase, observed by Robu et al. (2013) and
Richards et al. (2008), are isolated cases in Europe. These
outstanding isotopic signatures are probably local phenomenon, as
the locations of these sites are not connected to any basic
geographical factor such as altitude, latitude or longitude. Also the
distance from the sea coast may not be a responsible factor. The
new isotopic data for Central-Eastern European cave bears
emphasized the unusual isotopic values of these two Romanian
cave bear populations, for which further investigations are required
to understand the reason for such values, since omnivory is not the
sufﬁcient explanation (Bocherens et al., 2014b; Bocherens, 2015).
Data on isotopic variability of European cave bears presented in
this study may serve as a background for further research on
palaeoecology and palaeogeography of this taxon.
Acknowledgements
National Science Centre, Poland, supported the research into:
the remains from Nietoperzowa Cave (grant number 2012/05/B/
HS3/03751), and radiocarbon dating of cave bears from Perspektywiczna
Cave (grant number 2011/01/N/HS3/01299). The
research has been funded also by the Slovak grant agencies under
contracts APVV-0280-07 and Vega 1/0095/14.
Authors are grateful to two anonymous Reviewers for their
constructive comments on the earlier draft of this manuscript.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.quascirev.2015.10.028.
References
Alfano, M.J., Barron, E.J., Pollard, D., Huntley, B., Allen, J.R.M., 2003. Comparison of
climate model results with European vegetation and permafrost during oxygen
isotope stage three. Quat. Res. 59, 97e107. http://dx.doi.org/10.1016/S0033-
5894(02)00012-1.
Ambrose, S.H., 1990. Preparation and characterization of bone and tooth collagen
for isotopic analysis. J. Archaeol. Sci. 17, 431e451. http://dx.doi.org/10.1016/
0305-4403(90)90007-R.
Ambrose, S.H., DeNiro, M.J., 1986. The isotopic ecology of east African mammals.
Oecologia 69 (3), 395e406. http://dx.doi.org/10.1007/BF00377062.
Amundson, R., Austin, A.T., Schuur, E.A.G., Yoo, K., Matzek, V., Kendall, C.,
Uebersax, A., Brenner, D., Baisden, W.T., 2003. Global patterns of the isotopic
composition of soil and plant nitrogen. Glob. Biogeochem. Cycles 17 (1, 1031),
1e11. http://dx.doi.org/10.1029/2002GB001903.
Arroyo, M.K., Medina, E., Ziegler, H., 1990. Distribution and d13
C values of Portulaceae
species of the high Andes in northern Chile. Acta Bot. 103, 291e295.
Baca, M., Mackiewicz, P., Stankovic, A., Popovic, D., Stefaniak, K., Czarnogorska, K.,
Nadachowski, A., Ga˛siorowski, M., Hercman, H., Weglenski, P., 2014. Ancient
DNA and dating of cave bear remains from Niedzwiedzia cave suggest early
appearance of Ursus ingressus in Sudetes. Quat. Int. 339e340, 217e223. http://
dx.doi.org/10.1016/j.quaint.2013.08.033.
Baca, M., Stankovic, A., Stefaniak, K., Marciszak, A., Hofreiter, M., Nadachowski, A.,
We˛glenski, P., Mackiewicz, P., 2012. Genetic analysis of cave bear specimens
from Niedzwiedzia Cave, Sudetes, Poland. Palaeontol. Electron. 15 (2,21A), 1e16.
Bakwin, P.S., Tans, P.P., White, J.W.C., Andres, R.J., 1998. Determination of the isotopic
(13
C/12
C) discrimination by terrestrial biology from a global network of
observations. Glob. Biogeochem. Cycles 12, 555e562.
Baldwin, R.A., Bender, L.C., 2011. Denning chronology of black bears in eastern
Rocky mountain National Park, Colorado. West. North Am. Nat. 70 (1), 48e54.
Baryshnikov, G., 1998. Cave bears from the Palaeolithic of the greater Caucasus.
Springﬁeld. In: Saunders, J.J., Styles, B.W., Baryshnikov, G.F. (Eds.), Quaternary
Paleozoology in the Northern Hemisphere, Illinois State Museum Scientiﬁc
Papers, vol. 27, pp. 69e118.
Baryshnikov, G.F., Puzachenko, A.Yu, 2011. Craniometrical variability in the cave
bears (Carnivora, Ursidae): multivariate comparative analysis. Quat. Int. 245,
350e368. http://dx.doi.org/10.1016/j.quaint.2011.02.035.
Blant, M., Bocherens, H., Bochud, M., Braillard, L., Constandache, M., Jutzet, J.-M.,
2010. Le gisementa faune Würmienne du B€arenloch (Prealpes fribourgeoises).
Bull. la Societe Fribg. Desscience Nat. SFSN 99, 1e22.
Bocherens, H., 2000. Preservation of isotopic signals (13
C, 15
N) in Pleistocene
mammals. In: Katzenberg, M.A., Ambrose, S.H. (Eds.), Biogeochemical Approaches
to Paleodietary Analyses. Kluwer Academic/Plenum Publishers, New
York, pp. 65e88.
Bocherens, H., 2003. Isotopic biogeochemistry and the paleoecology of the
mammoth steppe fauna. In: Reumer, J.W.F., De Vos, J., Mol, D. (Eds.), Advances in
Mammoth Research (Proceedings of the Second International Mammoth Conference,
Rotterdam, May 16e20 1999), DEINSEA, 9, pp. 57e76.
Bocherens, H., 2004. Cave bear palaeoecology and stable isotopes: checking the
rules of the game. In: Philippe, M., Argant, A., Argant, J. (Eds.), Proceedings of
the 9th International Cave Bear Conference, Cahiers scientiﬁques du Centre de
Conservation et d'Etude des Collections. Museum d'Histoire naturelle de Lyon,
pp. 183e188. Hors Serie No 2.
Bocherens, H., 2015. Isotopic tracking of large carnivore palaeoecology in the
mammoth steppe. Quat. Sci. Rev. 117, 42e71. http://dx.doi.org/10.1016/
j.quascirev.2015.03.018.
Bocherens, H., Billiou, D., Patou-Mathis, M., Bonjean, D., Otte, M., Mariotti, A., 1997.
Palaeobiological implications of the isotopic signatures (13
C, 15
N) of fossil
mammal collagen in Scladina cave (Sclayn, Belgium). Quat. Res. 48, 370e380.
http://dx.doi.org/10.1006/qres.1997.1927.
Bocherens, H., Drucker, D., 2003. Trophic level isotopic enrichment of carbon and
nitrogen in bone collagen: case studies from recent and ancient terrestrial
ecosystems. Int. J. Osteoarchaeol. 13, 46e53. http://dx.doi.org/10.1002/oa.662.
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e72 69
Bocherens, H., Drucker, D.G., Billiou, D., Geneste, J.-M., van der Plicht, J., 2006. Bears
and humans in Chauvet Cave (Vallon-Pont-d'Arc, Ardeche, France): Insights
from stable isotopes and radiocarbon dating of bone collagen. J. Hum. Evol. 50,
370e376.
Bocherens, H., Drucker, D.G., Billiou, D., Moussa, I., 2005. Une nouvelle approche
pour evaluer l'etat de conservation de l'os et du collagene pour les mesures
isotopiques (datation au radiocarbone, isotopes stables du carbone et de
l'azote). L'Anthropologie 109, 557e567. http://dx.doi.org/10.1016/
j.anthro.2005.06.005.
Bocherens, H., Drucker, D.G., Bonjean, D., Bridault, A., Conard, N.J., Cupillard, C.,
Germonpre, M., H€oneisen, M., Münzel, S.C., Napierrala, H., Patou-Mathis, M.,
Stephan, E., Uerpmann, H.-P., Ziegler, R., 2011a. Isotopic evidence for dietary
ecology of cave lion (Panthera spelaea) in North-western Europe: prey choice,
competition and implications for extinction. Quat. Int. 245, 249e261. http://
dx.doi.org/10.1016/j.quaint.2011.02.023.
Bocherens, H., Drucker, D.G., Madelaine, S., 2014a. Evidence for a 15
N positive
excursion in terrestrial foodwebs at the Middle to Upper Palaeolithic transition
in south-western France: implications for early modern human palaeodiet and
palaeoenvironment. J. Hum. Evol. 69, 31e43. http://dx.doi.org/10.1016/
j.jhevol.2013.12.015.
Bocherens, H., Fizer, H.M., Mariotti, A., 1990. Mise en evidence alimentaire vegetarien
de l'ours des cavernes (Ursus spelaeus) parla biogeochemie isotopique
(13
C, 15
N) des vertebres fossiles. CR Acad Sci Paris Ser II 311, pp. 1279e1284.
Bocherens, H., Fizet, M., Mariotti, A., 1994. Diet, physiology and ecology of fossil
mammals as inferred from stable carbon and nitrogen isotope biogeochemistry:
implications for Pleistocene bears. Palaeogeogr. Palaeoclimatol. Palaeoecol. 107
(3e4), 213e225. http://dx.doi.org/10.1016/0031-0182(94)90095-7.
Bocherens, H., Fizet, M., Mariotti, A., Billiou, D., Bellon, G., Borel, J.-P., Simone, S.,
1991. Biogeochimie isotopique (13
C, 15
N, 18
O) et paleoecologie des ours
pleistocenes de la Grotte d'Aldene (Cesseras, Herault). Bull. Musee d'Anthropologie
Prehistorique Monaco 34, 29e49.
Bocherens, H., Grandal-d'Anglade, A., Hobson, K.A., 2014b. Pitfalls in comparing
modern hair and fossil bone collagen C and N isotopic data to reconstruct
ancient diets: a case study with cave bears (Ursus spelaeus). Isotopes Health
Environ. Sci. 50 (3), 291e299.
Bocherens, H., Stiller, M., Hobson, K.A., Pacher, M., Rabeder, G., Burns, J.A., Tütken, T.,
Hofreiter, M., 2011b. Niche partitioning between two sympatric genetically
distinct cave bears (Ursus spelaeus and Ursus ingressus) and brown bear (Ursus
arctos) from Austria: isotopic evidence from fossil bones. Quat. Int. 245,
238e248. http://dx.doi.org/10.1016/j.quaint.2010.12.020.
Bon, C., Berthonaud, V., Fosse, P., Gely, B., Maksud, F., Vitalis, R., Philippe, M., van der
Plicht, J., Elalouf, J.-M., 2011. Low regional diversity of late cave bears mitochondrial
DNA at the time of Chauvet Aurignacian paintings. J. Archaeol. Sci. 38,
1886e1895. http://dx.doi.org/10.1016/j.jas.2011.03.033.
Bonaﬁni, M., Pellegrini, M., Ditchﬁeld, P., Pollard, A.M., 2013. Investigation of the
‘canopy effect’ in the isotope ecology of temperate woodlands. J. Archaeol. Sci.
40, 3926e3935. http://dx.doi.org/10.1016/j.jas.2013.03.028.
Bryant, J.D., Froelich, P.N., 1995. A model of oxygen isotope fractionation in body
water of large mammals. Geochim. Cosmochim. Acta 59 (21), 4523e4537.
http://dx.doi.org/10.1016/0016-7037(95)00250-4.
Cerling, T.E., Harris, J.M., 1999. Carbon isotope fractionation between diet and
bioapatite in ungulate mammals and implications for ecological and paleoecological
studies. Oecologia 120, 347e363. http://dx.doi.org/10.1007/
s004420050868.
Chmielewski, W., 1961. Civilisation de Jerzmanowice. IHKM PAN,
WrocławeWarszawaeKrakow, pp. 1e92.
Chmielewski, W., 1975. Paleolit srodkowy i gorny. In: Chmielewski, W., Hensel, W.
(Eds.), Prahistoria Ziem Polskich II. Paleolit I Mezolit. Zakład Narodowy Imienia
Ossolinskich Wydawnictwo PAN, Wrocław, Warszawa, Krakow, Gdansk,
pp. 9e158.
Cordell, S., Goldstein, G., Meinzer, F.C., Handley, L.L., 1999. Allocation of nitrogen and
carbon in leaves of Metrosideros polymorpha regulates carboxylation capacity
and d13
C along an altitudinal gradient. Funct. Ecol. 13, 811e818.
Craine, J.M., Elmore, A.J., Wang, L., Augusto, L., Baisden, W.T., Brookshire, E.N.J.,
Cramer, M.D., Hasselquist, N.J., Hobbie, E.A., Kahmen, A., Koba, K.,
Kranabetter, J.M., Mack, M.C., Marin-Spiotta, E., Mayor, J.R., McLauchlan, K.K.,
Michelsen, A., Nardoto, G.B., Oliveira, R.S., Perakis, S.S., Peri, P.L., Quesada, C.A.,
Richter, A., Schipper, L.A., Stevenson, B.A., Turner, B.L., Viani, R.A.G., Wanek, W.,
Zeller, B., 2015. Convergence of soil nitrogen isotopes across global climate
gradients. Sci. Rep. 5, 8280. http://dx.doi.org/10.1038/srep08280.
Craine, J.M., Lee, W.G., 2003. Covariation in leaf and root traits for native and nonnative
grasses along an altitudinal gradient in New Zealand. Oecologia 134,
471e478.
Dalerum, F., Bennett, N.C., Clutton-Brock, T.H., 2007. Longitudinal differences in 15N
between mothers and offspring during and after weaning in a small cooperature
mammal, the meerkat (Suricata suricatta). Rapid Commun. Mass Spectrom.
21, 1889e1892. http://dx.doi.org/10.1002/rcm.3032.
De Smet, S., Balcaen, A., Claeys, E., Boeckx, P., Van Cleemput, O., 2004. Stable carbon
isotope analysis of different tissues of beef animals in relation to their diet.
Rapid Commun. Mass Spectrom. 18, 1227e1232. http://dx.doi.org/10.1002/
rcm.1471.
DeNiro, M.J., 1985. Postmortem preservation and alteration of in vivo bone collagen
isotope ratios in relation to palaeodietary reconstruction. Nature 317, 806e809.
http://dx.doi.org/10.1038/317806a0.
DeNiro, M.J., Epstein, S., 1978. Inﬂuence of diet on the distribution of carbon
isotopes in animals. Geochim. Cosmochim. Acta 42, 495e506. http://dx.doi.org/
10.1016/0016-7037(78)90199-0.
DeNiro, M.J., Epstein, S., 1981. Inﬂuence of diet on the distribution of nitrogen
isotopes in animals. Geochim. Cosmochim. Acta 45, 341e351. http://dx.doi.org/
10.1016/0016-7037(81)90244-1.
Diedrich, C.J., 2009. Upper Pleistocene Panthera leo spelaea (Goldfuss, 1810) remains
from the Bilstein Caves (Sauerland Karst) and contribution to the steppe lion
taphonomy, palaeobiology and sexual dimorphism. Ann. Paleontol. 95, 117e138.
http://dx.doi.org/10.1016/j.annpal.2009.07.001.
Diefendorf, A.F., Mueller, K.E., Wing, S.L., Koch, P.L., Freeman, K.H., 2010. Global
patterns in leaf 13
C discrimination and implications for studies of past and
future climate. PNAS 107, 5738e5743.
D€oppes, D., Rabeder, G. (Eds.), 1997. Plioz€ane und pleistoz€ane Faunen €Osterreichs.
Ein Katalog der wichtigsten Fossilfundstellen und ihrer Faunen. Mitt. Komm.
Quart€arforsch. €Osterr. Akad. Wiss. 10: 1e411, Wien. (130).
Drucker, D.G., Bridault, A., Cupillard, C., Hujic, A., Bocherens, H., 2011. Evolution of
habitat and environment of red deer (Cervus elaphus) during the Late-Glacial
and Early Holocene in eastern France (French Jura and the western Alps) using
multi-isotope analysis (d13
C, d15
N, d18
O, d34
S) of archaeological remains.
Quat. Int. 245, 268e278. http://dx.doi.org/10.1016/j.quaint.2011.07.019.
Drucker, D.G., Bridault, A., Hobson, K.A., Szuma, E., Bocherens, H., 2008. Can carbon-
13 abundances in large herbivores track canopy effect in temperate and boreal
ecosystems? Evidence from modern and ancient ungulates. Palaeogeogr.
Palaeoclimatol. Palaeoecol. 266, 69e82. http://dx.doi.org/10.1016/
j.palaeo.2008.03.020.
Du, B., Kang, H., Zhu, Y., Zhou, X., Yin, S., Burgess, P.J., Liu, C., 2015. Variation of
oriental oak (Quercus variabilis) leaf d13
C across temperate and subtropical
China: spatial patterns and sensitivity to precipitation. Forests 6, 2296e2306.
http://dx.doi.org/10.3390/f6072296.
Ehrenberg, K., 1927. Die Variabilit€at der Molaren des H€ohlenb€aren. Palaeontol. Z. 9,
240e250. http://dx.doi.org/10.1007/BF03041893.
Fejfar, O., 1953. Zprava o výskumu Medvedí jeskyne ve Slovenskem raji (Report on
the research of the Medvedia jaskyna in the Slovenský raj Mts). Krasy Slov. 30,
1e3.
Fernandez-Mosquera, D., Vila-Taboada, M., Grandal-d’Anglade, A., 2001. Stable
isotopes data (d13
C, d15
N) from the cave bear (Ursus spelaeus): a new approach
to its palaeoenvironment and dormancy. Proc. R. Soc. B Biol. Sci. 268 (1472),
1159e1164. http://dx.doi.org/10.1098/rspb.2001.1639.
Figueirido, B., Palmqvist, P., Perez-Claros, J.A., 2009. Ecomorphological correlates of
craniodental variation in bears and paleobiological implications for extinct
taxa: an approach based on geometric morphometrics. J. Zoology 277, 70e80.
http://dx.doi.org/10.1111/j.1469-7998.2008.00511.x.
Fizet, M., Mariotti, M., Bocherens, H., Lange-Badre, B., Vandermeersch, B., Borel, J.P.,
Bellon, G., 1995. Effect of diet, physiology and climate on carbon and nitrogen
stable isotopes of collagen in a Late Pleistocene anthropic paleoecosystem
(France, Charente, Marillac). J. Archaeol. Sci. 22, 67e79. http://dx.doi.org/
10.1016/S0305-4403(95)80163-4.
Fletcher, W.J., Sanchez Go~ni, M.F., Allen, J.R.M., Cheddadi, R., CombourieuNebout,
N., Huntley, B., Lawson, I., Londeix, L., Magri, D., Margari, V., Müller, U.C.,
Naughton, F., Novenko, E., Roucoux, K., Tzedakis, P.C., 2010. Millennial-scale
variability during the last glacial in vegetation records from Europe. Quat. Sci.
Rev. 29, 2839e2864. http://dx.doi.org/10.1016/j.quascirev.2009.11.015.
Fogel, M.L., Tuross, N., Owsley, D.W., 1989. Nitrogen isotope tracers of human
lactation in modern and archaeological populations. Annual report of the director,
geophysical laboratory. Carnegie Inst. Wash. 89, 111e117.
Fricke, H., 2007. Stable isotope geochemistry of bonebed fossils: reconstructing
paleoenvironments, paleoecology, and paleobiology. In: Rogers, R.R.,
Eberth, D.A., Fiorillo, A.R. (Eds.), Bonebeds. Genesis, Analysis, and Paleobiological
Signiﬁcance. The University of Chicago Press, Chicago and London,
pp. 437e490.
Frischauf, C., Argant, A., Kavcik, N., Marin, J., Pacher, M., Philippe, M., Rabeder, G.,
Refﬁenna, B., Withalm, G., 2010. In: The cave bears of the Massif du Vercors
(France). 16th International Cave Bear and Lion Symposium Aze (Sa^one-et
eLoire) September 22nde26th, Aze, pp. 76e77.
García García, N., Feranec, R.S., Arsuaga, J.L., Bermúdez de Castro, J.M., Carbonell, E.,
2009. Isotopic analysis of the ecology of herbivores and carnivores from the
Middle Pleistocene deposits of the Sierra De Atapuerca, northern Spain.
J. Archaeol. Sci. 36, 1142e1151. http://dx.doi.org/10.1016/j.jas.2008.12.018.
Garten Jr., C.T., 1993. Variation in foliar 15
N abundance and the availability of soil N
on the Walker branch watershed. Ecology 74, 2098e2113.
Garten Jr., C.T., Migroet, H., 1994. Relationship between soil nitrogen dynamics and
natural 15
N abundance in plant foliage from Great Smoky Mountains National
Park. Can. J. For. Res. 24, 1636e1645.
Grandal d’Anglade, A., Fernandez Mosquera, D., 2008. Hibernation can also cause
high d15
N values in cave bears: a response to Richards et al. Proc. Natl. Acad. Sci.
U. S. A. 105 (11), E15.
Hammer, O., 1999e2014. PAST PAleontological STatistics Version 3.04, Reference
Manual. University of Oslo, p. 224.
Hammer, O., Harper, D.A.T., Ryan, P.D., 2001. PAST: paleontological statistics software
for education and data analysis. Plaeontologia Electron. 4 (1), 9.
Heaton, T.H.E., Vogel, J.C., Chevallerie, G.V.l., Collett, G., 1986. Climatic inﬂuence on
the isotopic composition of bone nitrogen. Nature 322, 822e824. http://
dx.doi.org/10.1038/322822a0.
Helmens, K.F., 2014. The last interglacial-glacial cycle (MIS 5-2) re-examined based
on long proxy records from central and northern Europe. Quat. Sci. Rev. 86,
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e7270
115e143. http://dx.doi.org/10.1016/j.quascirev.2013.12.012.
Hilderbrand, G.V., Farley, S.D., Robbins, C.T., Hanley, T.A., Titus, K., Servheen, C., 1996.
Use of stable isotopes to determine diets of living and extinct bears. Can. J.
Zoology 74, 2080e2088. http://dx.doi.org/10.1139/z96-236.
Hindelang, M., Peterson, R.O., Maclean, A.L., 2002. Compensatory bone remodelling
in moose: a study of age, sex, and cross-sectional cortical bone dimensions in
moose at Isle Royale National Park. Int. J. Osteoarchaeol. 12, 343e348.
Hobbie, E.A., Macko, S.A., Shugart, H.H., 1998. Patterns of N dynamics and N isotopes
during primary succession in Glacier Bay, Alaska. Chem. Geol. 152, 3e11. http://
dx.doi.org/10.1016/S0009-2541(98)00092-8.
Hobson, K.A., McLellan, B.N., Woods, J.G., 2000. Using stable carbon (d13
C) and nitrogen
(d15
N) isotopes to infer trophic relationships among black and grizzly
bears in the upper Columbia river basin, British Columbia. Can. J. Zoology 78,
1332e1339. http://dx.doi.org/10.1139/z00-069.
Hofreiter, M., Capelli, Ch, Krings, M., Waits, L., Conard, N., Münzel, S., Rabeder, G.,
Nagel, D., Paunovic, M., Jambresic, G., Meyer, S., Weiss, G., P€a€abo, S., 2002.
Ancient DNA analyses reveal high mitochondrial DNA sequences diversity and
parallel morphological evolution of Late Pleistocene cave bears. Mol. Biol. Evol.
19 (8), 1244e1250.
Hofreiter, M., Münzel, S., Conard, N., Pollack, J., Slatkin, M., Weiss, G., P€a€abo, S., 2007.
Sudden replacement of cave bear mitrochondrial DNA in the late Pleistocene.
Curr. Biol. 17 (4), R1eR3.
Hofreiter, M., Rabeder, G., Jaenicke-Despres, V., Withalm, G., Nagel, D., Paunovic, M.,
Jambresic, G., P€a€abo, S., 2004a. Evidence for reproductive isolation between
cave bear populations. Curr. Biol. 14 (1), 40e43.
Hofreiter, M., Serre, D., Rohland, N., Rabeder, G., Nagel, D., Conard, N., Münzel, S.,
P€a€abo, S., 2004b. Lack of phylogeography in European mammals before the last
glaciation. PNAS 101 (35), 12963e12968.
Horacek, M., Frischauf, C., Pacher, M., Rabeder, G., 2012. Stable isotopic analyses of
cave bear bones from the Conturines cave (2800 m, south Tyrol, Italy). Braunschweiger
Naturkundl. Schriften 11 (1), 47e52.
Houmark-Nielsen, M., 2009. MIS 3 Interstadial Climate and Rapid Ice Advances in
the Southwestern Baltic. Svensk K€arnbr€anslehantering AB. Swedish Nuclear
Fuel and Waste Management Co, Stockholm, pp. 1e8.
Huja, S.S., Beck, F.M., 2007. Bone remodeling in maxilla, mandible, and femur of
young dogs. Anat. Rec. 291, 1e5.
Hultine, K.R., Marshall, J.D., 2000. Altitude trends in conifer leaf morphology and
stable carbon isotope composition. Oecologia 123, 32e40.
Huntley, B., Alfano, M.J., Allen, J.R.M., Pollard, D., Tzedakis, P.C., de Beaulieu, J.-L.,
Grüger, E., Watts, B., 2003. European vegetation during marine oxygen isotope
stage-3. Quat. Res. 59, 195e212. http://dx.doi.org/10.1016/S0033-5894(02)
00016-9.
Janacik, P., Schmidt, Z., 1965. Medvedia jaskyna v Slovenskej hornatine (Slovenský
raj) (the Medvedia jaskyna in the Slovenska hornatina Mts. (the Slovenský raj
Mts.)). Slov. Kras. 5, 10e36.
Jenkins, S.G., Partridge, S.T., Stephenson, T.R., Farley, S.D., Robbins, C.T., 2001. Nitrogen
and carbon isotope fractionation between mothers, neonates, and
nursing offspring. Oecologia 129, 336e341. http://dx.doi.org/10.1007/
s004420100755.
Kjellstr€om, E., Brandefelt, J., N€aslund, J.-O., Smith, B., Strandberg, G., Voelker, A.H.L.,
Wohlfarth, B., 2010. Simulated climate conditions in Europe during the marine
isotope stage 3 stadial. Boreas 39, 436e456. http://dx.doi.org/10.1111/j.1502-
3885.2010.00143.x.
Knapp, M., Rohland, N., Weinstock, J., Baryshnikov, G., Sher, A., Nagel, D.,
Rabeder, G., Pinhasi, R., Schmidt, H.A., Hofreiter, M., 2009. First DNA sequences
from Asian cave bear fossils reveal deep divergences and complex phylogeographic
patterns. Mol. Ecol. 18 (6), 1225e1238. http://dx.doi.org/10.1111/j.1365-
294X.2009.04088.x.
Koch, P.L., 2007. Isotopic study of the biology of modern and fossil vertebrates. In:
Michener, R., Lajtha, K. (Eds.), Stable Isotopes in Ecology and Environmental
Science. Blackwell Publishing, pp. 99e154. http://dx.doi.org/10.1002/
9780470691854.ch5.
Kogami, H., Hanba, Y.T., Kibe, T., Terashima, I., Masuzawa, T., 2001. CO2 transfer
conductance, leaf structure and carbon isotope composition of Polygonum
cuspidatum leaves from low and high altitudes. Plant Cell Environ. 24, 529e538.
Kohn, M.J., 2010. Carbon isotope compositions of terrestrial C3 plants as indicators
of (paleo)ecology and (paleo)climate. PNAS 107, 19691e19695.
Komar, M., Łanczont, M., Madeyska, T., 2009. Spatial vegetation patterns based on
palynological records in the loess area between the Dnieper and Odra rivers
during the last interglacialeglacial cycle. Quat. Int. 198, 152e172. http://
dx.doi.org/10.1016/j.quaint.2008.04.008.
K€orner, C., Farquhar, G.D., Roksandic, Z., 1988. A global survey of carbon isotope
discrimination in plants from high altitude. Oecologia 74, 623e632.
K€orner, C., Newmayer, M., Palaez Menendez-Reidl, S., Smeets-Scheel, A., 1989.
Functional morphology of mountain plants. Flora 182, 353e383.
K€orner, Ch, Farquhar, G.D., Wong, S.C., 1991. Carbon isotope discrimination by plants
follows latitudinal and altitudinal trends. Oecologia 88, 30e40. http://
dx.doi.org/10.1007/BF00328400.
Kowalski, K., 1961. Pleistocene rodents from Nietoperzowa cave in Poland. Folia
Quat. 5, 1e19.
Krajcarz, M.T., Krajcarz, M., 2014. The 200,000 year long record of stable isotopes
(d18
O, d13
C) of cave bear (Ursus spelaeus) teeth from Bisnik Cave, Poland. Quat.
Int. 339e340, 119e130. http://dx.doi.org/10.1016/j.quaint.2013.07.022.
Krajcarz, M.T., Madeyska, T., 2010. Application of the weathering parameters of
bones to stratigraphical interpretation of the sediments from two caves
(Deszczowa cave and Nietoperzowa cave, Krakow-Cze˛stochowa Upland,
Poland). Stud. Quat. 27, 43e54.
Kurten, B., 1976. The Cave Bears Story: Life and Death of a Vanished Animal.
Columbia University Press, New York, N.Y, pp. 11e63.
Lee-Thorp, J.A., Sponheimer, M., 2003. Three case studies used to reassess the
reliability of fossil bone and enamel isotope signals for paleodietary studies.
J. Anthropol. Archaeol. 22 (3), 208e216. http://dx.doi.org/10.1016/
j.cbpa.2007.11.012.
Liden, K., Angerbjӧrn, A., 1999. Dietary change and stable isotopes: a model of
growth and dormancy in cave bears. Proc. R. Soc. B Biol. Sci. 266, 1779e1783.
http://dx.doi.org/10.1098/rspb.1999.0846.
Lindner, L. (Ed.), 1992. Czwartorze˛d: Osady, Metody Badan, Stratygraﬁa. Wydawnictwo
PAE, Warszawa, pp. 1e683.
Loreille, O., Orlando, L., Patou-Mathis, M., Philippe, M., Taberlet, P., H€anni, C., 2001.
Ancient DNA analysis reveals divergence of the cave bear, Ursus spelaeus, and
brown bear, Ursus arctos, lineages. Curr. Biol. 11, 200e203.
Madeyska, T., 1981. Srodowisko człowieka w srodkowym i gornym paleolicie na
ziemiach polskich w swietle badan geologicznych. Stud. Geol. Pol. 69, 1e125.
Madeyska, T., 1982. The stratigraphy of Palaeolithic sites of the Cracow Upland.
Acta Geol. Pol. 32, 227e242.
Madeyska-Niklewska, T., 1969. Gornoplejstocenskie osady jaskin Wy_zyny Krakowskiej.
Acta Geol. Pol. 19 (2), 341e390.
Manchi, S., Swenson, J.E., 2005. Denning behavior of Scandinavian brown bears
Ursus arctos. Wildl. Biol. 11, 123e132.
M€annel, T.T., Auerswald, K., Schnyder, H., 2007. Altitudinal gradients of grassland
carbon and nitrogen isotope composition are recorded in the hair of grazers.
Glob. Ecol. Biogeogr. 16, 583e592. http://dx.doi.org/10.1111/j.1466-
8238.2007.00322.x.
Mariotti, A., Pierre, D., Vedy, J.C., Burckert, S., Guillemot, J., 1980. The abundance of
natural nitrogen 15 in the organic matter of soils along an altitudinal gradient
(Chablais, Haute- Savoie, France). Catena 7, 293e300. http://dx.doi.org/10.1016/
0341-8162(80)90014-4.
Meinzer, F.C., Rundel, P.W., Goldstein, G., Shariﬁ, M.R., 1992. Carbon isotope
composition in relation to leaf gas exchange and environmental conditions in
Hawaiian Metrosideros polymorpha populations. Oecologia 91, 305e311.
Miller, A.E., Bowman, W.D., 2002. Variation in nitrogen-15 natural abundance and
nitrogen uptake traits among co-occurring alpine species: do species partition
by nitrogen form? Oecologia 130, 609e616. http://dx.doi.org/10.1007/s00442-
001-0838-8.
Miller, J.B., Tans, P.P., White, J.W.C., Conway, T.J., Vaughn, B.W., 2003. The atmospheric
signal of terrestrial isotopic discrimination and its implication for partitioning
of carbon ﬂuxes. Tellus Ser. B 55, 197e206.
Minagawa, M., Wada, E., 1984. Stepwise enrichment of 15
N along food chains:
further evidence and the relation between d15
N and animal age. Geochim.
Cosmochim. Acta 48 (5), 1135e1140. http://dx.doi.org/10.1016/0016-7037(84)
90204-7.
Mojski, J.E., 1985. Geologia Polski, I Stratygraﬁa, 3b Kenozoik, Czwartorze˛d.
Wydawnictwo Geologiczne, Warszawa, pp. 1e396.
Morecroft, M.D., Woodward, F.I., 1990. Experimental investigations on the environmental
determination of d13
C at different altitudes. J. Exp. Bot. 41,
1303e1308.
Münzel, S.C., Conard, N.J., 2004. Cave Bear hunting in the Hohle Fels, a cave site in
the Ach Valley, swabian jura. Rev. Paleobiol. 23 (2), 877e885.
Münzel, S.C., Hofreiter, M., Stiller, M., Mittnik, A., Conard, N.J., Bocherens, H., 2011.
Pleistocene bears in the Swabian Jura (Germany): genetic replacement,
ecological displacement, extinctions and survival. Quat. Int. 245, 225e237.
http://dx.doi.org/10.1016/j.quaint.2011.03.060.
Münzel, S.C., Rivals, F., Pacher, M., D€oppes, D., Rabeder, G., Conard, N.J.,
Bocherens, H., 2014. Behavioural ecology of late Pleistocene bears (Ursus spelaeus,
Ursus ingressus): insight from stable isotopes (C, N, O) and tooth microwear.
Quat. Int. 339e340, 148e163. http://dx.doi.org/10.1016/
j.quaint.2013.10.020.
Nelson, D.E., Angerbj€orn, A., Liden, K., Turk, I., 1998. Stable isotopes and the
metabolism of the European cave bear. Oecologia 116, 177e181. http://
dx.doi.org/10.1007/s004420050577.
Nelson, D.E., Ku, T.-L., 1997. Radiocarbon dating of bone and charcoal from Divje
Babe I Cave. In: Turk, I. (Ed.), Mousterian Bone Flute and Other Finds from Divje
Babe I Cave Site in Slovenia. Znanstvenoraziskovalni Centre Sazu, Ljubljana,
pp. 51e65.
Nelson, R.A., Jones, J.D., Wahner, H.W., McGill, D.V.B., Code, C.F., 1975. Nitrogen
metabolism in bears: urea metabolism in summer starvation and in winter
sleep and role of urinary bladder in water and nitrogen conservation. Mayo Clin.
Proc. 50, 141e146.
Orlando, L., Bonjean, D., Bocherens, H., Thenot, A., Argant, A., Otte, M., H€anni, C.,
2002. Ancient DNA and the population genetics of cave bears (Ursus spelaeus)
through space and time. Mol. Biol. Evol. 19, 1920e1933.
Pacher, M., 2003. Upper Pleistocene cave assemblages at Alpien sites in Austria and
adjacent regions. Preistoria Alp. 39, 115e127.
Pacher, M., Bocherens, H., D€oppes, D., Frischauf, C., Rabeder, G., 2012. First results of
stable isotopes from Drachenloch and Wildenmannlisloch, Swiss Alps. In:
D€oppes, D., Joger, U., Rosendahl, W. (Eds.), Proceedings of the 17th International
Cave Bear Symposium, 2011 September 15e18 ; Herzberg. Braunschweiger
Naturkundliche Schriften, vol. 11, pp. 101e110.
Pacher, M., Stuart, A.J., 2009. Extinction chronology and palaeocology of the cave
bear Ursus spelaeus. Boreas 38 (2), 189e206. http://dx.doi.org/10.1111/
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e72 71
bor.2009.38.issue-2/issuetoc.
Page, H.M., 1995. Variation in the natural abundance of 15
N in the halophyte, Salicornia
virginica, associated with groundwater subsidies of nitrogen in a
southern California salt-marsh. Oecologia 104, 181e188. http://dx.doi.org/
10.1007/BF00328583.
Pardo, L.H., Nadelhoffer, K.J., 2010. Using nitrogen isotope ratios to assess terrestrial
ecosystems at regional and global scales. In: West, J.B., Bowen, G.J., Dawson, T.E.,
Tu, K.P. (Eds.), Isoscapes: Understanding Movement, Pattern, and Process on
Earth Through Isotope Mapping. Springer ScienceþBusiness Media B.V.,
Springer, Dordrecht Heidelberg London New York, pp. 221e249. http://
dx.doi.org/10.1007/978-90-481-3354-3.
Passey, B.H., Robinson, T.F., Ayliffe, L.K., Cerling, T.E., Sponheimer, M., Dearing, M.D.,
Roeder, B.L., Ehleringer, J.R., 2005. Carbon isotope fractionation between diet,
breath CO2, and bioapatite in different mammals. J. Archaeol. Sci. 32,
1459e1470. http://dx.doi.org/10.1016/j.jas.2005.03.015.
Peigne, S., Goillot, C., Germonpre, M., Blondel, C., Bignon, O., Merceron, G., 2009.
Predormancy omnivory in European cave bears evidenced by a dental microwear
analysis of Ursus spelaeus from Goyet, Belgium. PNAS 106, 15390e15393,
10.1073pnas.0907373106.
Perez-Rama, M., Fernandez-Mosquera, D., Grandal-d’Anglade, A., 2011. Recognizing
growth patterns and maternal strategies in extinct species using stable isotopes:
the case of the cave bear Ursus spelaeus Rosenmüller. Quat. Int. 245,
302e306. http://dx.doi.org/10.1016/j.quaint.2010.09.009.
Popovic, D., Baca, M., Stefaniak, K., Lipecki, G., Nadachowski, A., Marciszak, A.,
Sabol, M., Ridush, B., Roblíckova, M., Kana, V., Obada, T., Stankovic, A.,
Mackiewicz, P., 2015. Genetics studies of cave bear (Ursus ingressus) population
from central to eastern Europe. In: 21st International Cave Bear Symposium.
Universiteit Leiden, Leiden, p. 28.
Quiles, J., Petrea, C., Moldovan, O., Zilh~ao, J., Rodrigo, R., Rougier, H., Constantin, S.,
Milota, S¸ ., Gherase, M., Sarcina, L., Trinkaus, E., 2006. Cave bears (Ursus spelaeus)
from the Pes¸ tera cu Oase (Banat, Romania): paleobiology and taphonomy.
Comptes Rendus Palevol 5, 927e934. http://dx.doi.org/10.1016/
j.crpv.2006.09.005.
Rabeder, G., Debeljak, I., Hofreiter, M., Withalm, G., 2008. Morphological responses
of cave bears (Ursus spelaeus group) to high-alpine habitats. Die H€ohle 59 (1e4),
59e72.
Rabeder, G., Nagel, D., Pacher, M., 2000. Der H€ohlenb€ar. Jan Thorbecke Verlag,
Stuttgart, p. 152. Thorbecke Species 4.
Richards, M.P., Pacher, M., Stiller, M., Quiles, J., Hofreiter, M., Constantin, S., Zilhao, J.,
Trinkaus, E., 2008. Isotopic evidence for omnivory among European cave bears:
late Pleistocene Ursus spelaeus from the Pes¸ tera cu Oase, Romania. PNAS 105,
600e604. http://dx.doi.org/10.1073/pnas.0711063105.
Riga, A., Van Praag, H.J., Brigode, N., 1971. Rapport isotopic naturel de l'azote dans
quelques sols forestiers et agricoles de Belgique soumis a divers traitements
culturaux. Geoderma 6, 213e222.
Robu, M., Fortin, J.K., Richards, M.P., Schwartz, C.C., Wynn, J.G., Robbins, C.T.,
Trinkaus, E., 2013. Isotopic evidence for dietary ﬂexibility among European late
Pleistocene cave bears (Ursus spelaeus). Can. J. Zoology 91, 227e234. http://
dx.doi.org/10.1139/cjz-2012-0222.
Rodiere, E., Bocherens, H., Angibault, J.-M., Mariotti, A., 1996. Particularites isotopiques
chez le chevreuil (Capreolus capreolus L.): Implications pour les reconstitutions
paleoenvironnementales. Comptes Rendus l'Academie Sci. Paris
323, 179e185.
Roth, J.D., Hobson, K.A., 2000. Stable carbon and nitrogen isotopic fractionation
between diet and tissue of captive red fox: implications for dietary reconstruction.
Can. J. Zoology 78, 848e852. http://dx.doi.org/10.1139/z00-008.
Ro_zycki, S.Z., 1967. Plejstocen Polski Srodkowej. PWN, Warszawa, p. 256.
Sabol, M., D€oppes, D., Pacher, M., Rabeder, G., Withalm, G., 2008. Cave Bears from
the Medvedia jaskyna in the Slovensky raj mountains (Slovakia): preliminary
results. Stalactite 58 (2), 74e77.
Sah, S.P., Brumme, R., 2003. Altitudinal gradients of natural abundance of stable
isotopes of nitrogen and carbon in the needles and soil of a pine forest in Nepal.
J. For. Sci. 49, 19e26.
Schoeninger, M.J., 1985. Trophic level effects on 15
N/14
N and 13
C/12
C ratios in bone
collagen and strontium levels in bone mineral. J. Hum. Evol. 14, 515e525. http://
dx.doi.org/10.1016/S0047-2484(85)80030-0.
Schulze, E.-D., Williams, R.J., Farquhar, G.D., Schulze, W., Landgridge, J., Miller, J.M.,
Walker, B.H., 1998. Carbon and nitrogen isotope discrimination and nitrogen
nutrition of trees along a rainfall gradient in northern Australia. Aust. J. Plant
Physiol. 25, 413e425. http://dx.doi.org/10.1071/PP97113.
Schuur, E.A.G., Matson, P.A., 2001. Net primary productivity and nutrient cycling
across a mesic to wet precipitation gradient in Hawaiian montane forest.
Oecologia 128, 431e442.
Seminoff, J.A., Jones, T.T., Eguchi, T., Hastings, M., Jones, D.R., 2009. Stable carbon
and nitrogen isotope discrimination in soft tissues of the leatherback turtle
(Dermochelys coriacea): insights for trophic studies of marine turtles. J. Exp.
Mar. Biol. Ecol. 381, 33e41. http://dx.doi.org/10.1016/j.jembe.2009.08.018.
Seryodkin, I.V., Kostyria, A.V., Goodrich, J.M., Miquelle, D.G., Smirnov, E.N.,
Kerley, L.L., Quigley, H.B., Hornocker, M.G., 2003. Denning ecology of brown
bears and asiatic black bears in the Russian Far East. Ursus 14 (2), 153e161.
Sigman, D.M., Boyle, E.A., 2000. Glacial/interglacial variations in atmospheric carbon
dioxide. Nature 407, 859e869.
Sp€otl, C., Reimer, P.J., Rabeder, G., Scholz, D., 2014. Presence of cave bears in western
Austria before the onset of the last Glacial maximum: new radiocarbon dates
and palaeoclimatic considerations. J. Quat. Sci. 29 (8), 760e766. http://
dx.doi.org/10.1002/jqs.2747.
Stiller, M., Molak, M., Prost, S., Rohland, N., Pacher, M., Rabeder, G., Baryshnikov, G.,
Rosendahl, W., Münzel, S., Bocherens, H., Grandal-d'Anglade, A., Hilpert, B.,
Germonpre, M., Stasyk, O., Pinhasi, R., Tintori, A., Ho, S.Y.W., Hofreiter, M.,
Knapp, M., 2014. Mitochondrial DNA diversity and evolution of the Pleistocene
cave bear complex. Quat. Int. 339e340, 224e231. http://dx.doi.org/10.1016/
j.quaint.2013.09.023.
Stiller, M., Baryshnikov, G., Bocherens, H., Grandal d’Anglade, A., Hilpert, B.,
Münzel, S.C., Pinhasi, R., Rabeder, G., Rosendahl, W., Trinkaus, E., Hofreiter, M.,
Knapp, M., 2010. Withering away e 25,000 years of genetic decline preceded
cave bear extinction. Mol. Biol. Evol. 27, 975e978.
Stiner, M.C., Achyuthan, H., Arsebük, G., Howell, F.C., Josephson, S., Juell, K., Pigati, J.,
Quade, J., 1998. Reconstructing cave bear paleoecology from skeletons: a cross
disciplinary study of middle Pleistocene bears from Yarimburgaz cave, Turkey.
Paleobiology 24, 74e98.
Suits, N.S., Denning, A.S., Berry, J.A., Still, C.J., Kaduk, J., Miller, J.B., Baker, I.T., 2005.
Simulation of carbon isotope discrimination of the terrestrial Biosphere. Glob.
Biogeochem. Cycles 19, GB1017. http://dx.doi.org/10.1029/2003GB002141.
Tieszen, L.L., Boutton, T.W., Tesdahl, K.G., Slade, N.A., 1983. Fractionation and
turnover of stable carbon isotopes in animal tissues: implications for d13
C
analysis of diet. Oecologia 57, 32e37.
Trinkaus, E., Richards, M.P., 2013. Stable isotopes and dietary patterns of the faunal
species from the Pes¸ tera cu Oase. In: Trinkaus, E., Constantin, S., Zilh~ao, J. (Eds.),
Life and Death at the Pes¸ tera Cu Oase. A Setting for Modern Human Emergence
in Europe. Oxford University Press, New York, pp. 211e226.
Van de Water, P.K., Leavitt, S.W., Betancourt, J.L., 2002. Leaf d13
C variability with
elevation, slope aspect, and precipitation in the southwest United States.
Oecologia 132, 332e343.
Van der Merwe, N.J., Medina, E., 1991. The canopy effect, carbon isotope ratios and
foodwebs in Amazonia. J. Archaeol. Sci. 18, 249e259.
Van Heteren, A., Maclarnon, A., Rae, T.C., Soligo, C., 2009. Cave bears and their
closest living relatives: a 3D geometric morphometrical approach to the functional
morphology of the cave bear Ursus spelaeus. Slov. Kras. Acta Carsologica
Slovaca 47 (1), 33e46.
Van Meerbeeck, C.J., Renssen, H., Roche, D.M., Wohlfarth, B., Bohncke, S.J.P.,
Bos, J.A.A., Engels, S., Helmens, K.F., Sanchez-Go~ni, M.F., Svensson, A.,
Vandenberghe, J., 2011. The nature of MIS 3 stadialeinterstadial transitions in
Europe: new insights from modeledata comparisons. Quat. Sci. Rev. 30
(25e26), 3618e3637. http://dx.doi.org/10.1016/j.quascirev.2011.08.002.
Vila Taboada, M., 1998. Nota sobre el estudio de ADN antiguo en restos oseos de
macromamíferos cuaternarios de Galicia. Cuad. do Lab. Xeoloxico Laxe Coru~na
23, 263e270.
Vila Taboada, M., Fernandez Mosquera, D., Grandal d’Anglade, A., 2001. Cave bear's
diet: a new hypothesis based on stable isotopes. Cuad. do Lab. Xeoloxico Laxe
Coru~na 26, 431e439.
Vila Taboada, M., Fernandez-Mosquera, A., Lopez Gonzalez, F., Grandal
d'Anglade, A., Vidal Romaní, J.R., 1999. Paleoecological implications inferred
from stable isotopic signatures (d13
C, d15
N) in bone collagen of Ursus spelaeus
ROS.-HEIN. Cuad. do Lab. Xeoloxico Laxe Coru~na 24, 73e87.
Vitousek, P.M., Field, C.B., Matson, P.A., 1990. Variation in foliar d13
C in Hawaiian
Metrosideros polymorpha d a case of internal resistance. Oecologia 84,
362e370.
Warren, C.R., McGrath, J.F., Adams, M.A., 2001. Water availability and carbon isotope
discrimination in conifers. Oecologia 127, 476e486.
Willis, K.J., van Andel, T.H., 2004. Trees or no trees? the environments of central and
eastern Europe during the last glaciation. Quat. Sci. Rev. 23, 2369e2387. http://
dx.doi.org/10.1016/j.quascirev.2004.06.002.
Wojtal, P., 2007. Zooarchaeological Studies of the Late Pleistocene Sites in Poland.
Institute of Systematics and Evolution of Animals, Polish Academy of Science,
Krakow, p. 189.
Wojtal, P., Wilczynski, J., Nadachowski, A., Münzel, S.C., 2015. Gravettian hunting
and exploitation of bears in central Europe. Quat. Int. 359e360, 58e71. http://
dx.doi.org/10.1016/j.quaint.2014.10.017.
Wysoczanski-Minkowicz, T., 1969. An attempt at relative age determination of fossil
bones by ﬂuorine-chlorine-apatite method. Stud. Geol. Pol. 28, 1e76.
Zhou, Y.C., Fan, J.W., Zhong, H.P., Zhang, W.Y., 2013. Relationships between altitudinal
gradient and plant carbon isotope composition of grassland communities
on the Qinghai-Tibet plateau, China. Sci. China Earth Sci. 56 (2), 311e320. http://
dx.doi.org/10.1007/s11430-012-4498-9.
Zhu, Y., Siegwolf, R.T.W., Durka, W., K€orner, C., 2010. Phylogenetically balanced
evidence for structural and carbon isotope responses in plants along elevational
gradients. Oecologia 162, 853e863. http://dx.doi.org/10.1007/s00442-009-
1515-6.
M. Krajcarz et al. / Quaternary Science Reviews 131 (2016) 51e7272