ArticlePDF Available

Abstract and Figures

A comparison of metrical and morphological data from more than 30 bear populations belonging to the cave bear group (U. spelaeus and its relatives U. eremus, U. ladinicus, U. ingressus as well as their predecessor U. deningeri) shows that the different species developed very different adaptations to the altitude of their habitats. Whereas there is a reduction of body size in Ursus eremus and U. ladinicus correlated with the altitude of the habitat (“alpine nanism”) no such correlation can be observed in U. ingressus. In U. ingressus there is a positive correlation of the tooth indices and the altitude of the habitat, i.e. in the higher sites the teeth are more evolved.
Content may be subject to copyright.
Die Höhle / 59. Jg. / Heft 1-4/2008 59
Morphological responses of
cave bears (Ursus spelaeus group)
to high-alpine habitats
Gernot Rabeder
University of Vienna, Institute of
Palaeontology and Naturkundliche
Station Lunz am See
gernot.rabeder@univie.ac.at
Irena Debeljak
Institute of Paleontology,
Scientific Research Centre SAZU,Novi trg 2,
SI-1000, Ljubljana, Slovenia
IrenaDe@zrc-sazu.si
Michael Hofreiter
Max Planck Institute for Evolutionary
Anthropology.
hofreite@eva.mpg.de
Gerhard Withalm
University of Vienna, Institute of Palaeonto-
logy.
gerhard.withalm@univie.ac.at
Eingelangt: 22.2.2008
Angenommen: 29.6.2008
A comparison of metrical and morphologi-
cal data from more than 30 bear populati-
ons belonging to the cave bear group (U.
spelaeus and its relatives U. eremus,U. ladi-
nicus,U. ingressus as well as their prede-
cessor U. deningeri) shows that the different
species developed very different adaptations
to the altitude of their habitats. Whereas
there is a reduction of body size in Ursus
eremus and U. ladinicus correlated with the
altitude of the habitat (“alpine nanism”) no
such correlation can be observed in U. in-
gressus. In U. ingressus there is a positive
correlation of the tooth indices and the alti-
tude of the habitat, i.e. in the higher sites
the teeth are more evolved.
Morphologische Anpassung des
Höhlenbären (Ursus spelaeus)
an hochalpine Habitate
Aus alpinen und außeralpinen Höhlen lie-
gen Höhlenbärenreste in so großen Mengen
vor, dass es möglich ist, statistische Metho-
den zu verwenden und die Bärenfaunen
untereinander zu vergleichen. Als „Höhlen-
bären“ verstehen wir, in Abwesenheit von
unterschiedlichen Trivialnamen, die Ange-
hörigen der Ursus spelaeus-Gruppe d.h. ne-
ben U. spelaeus auch die nah verwandten
Arten U. eremus,U. ladinicus,U. ingressus
sowie die mittelpleistozäne Vorläuferform
U. deningeri. Die hier untersuchten Höhlen-
bärenfaunen stammen alle aus dem Zeitbe-
reich zwischen etwa 80.000 und 25.000
Jahren vor heute. Zum Vergleich der Kör-
pergröße werden vor allem die Längen- und
Breitenwerte der Backenzähne und der Mit-
telhand- und Mittelfußknochen (Metapo-
dien) herangezogen, weil Zähne und Meta-
podien einerseits die bei weitem häufigsten
Elemente in den Fundinventaren sind und
weil andererseits die Körpergröße eng mit
der Zahn- und Metapodiengröße gekoppelt
ist. Als „morphologische Daten“ werden
ebenfalls statistisch erhobene „Indices“ ver-
standen, die uns das Evolutionsniveau einer
bestimmten Zahnkategorie angeben. Zum
Beispiel gibt der so genannte „m2-Enthy-
poconid-Index“ die durchschnittliche Anzahl
von zusätzlichen Höckern an einer be-
stimmten Stelle (= Enthypoconid) des zwei-
ten Unterkieferbackenzahnes an. Dieses
ein-, zwei oder mehrhöckerige Enthypoco-
nid wird gebildet, um dem Kaudruck des
zweiten Oberkieferbackenzahnes wirkungs-
volle Antagonisten entgegen zu setzen. Der
P4-Index besagt, wie oft zusätzliche Kaulei-
sten und Höcker entwickelt sind. Es hat sich
in den letzten 20 Jahren herausgestellt, dass
diese „Indices“ einen hohen Aussagewert
ABSTRACT ZUSAMMENFASSUNG
60 Die Höhle / 59. Jg. / Heft 1-4/2008
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
für die Evolution der Höhlenbären haben. Aus den sehr un-
terschiedlichen Indices der Backenzähne und dem geologi-
schen Alter wurde geschlossen, dass in den Alpen vor 50.000
bis 30.000 Jahren drei verschiedene Höhlenbärenarten mit
den zoologischen Namen Ursus ladinicus,U. eremus und U.
ingressus gelebt haben. Diese drei Arten zeichnen sich auch
durch unterschiedliche mitochondriale DNA Sequenzen aus,
ein Hinweis darauf, dass es sich um getrennte Populationen
handelte, zwischen denen wenig oder kein Genfluss stattge-
funden hat (Hofreiter, 2005), was für einige Fundstellen auch
nachgewiesen wurde (Hofreiter & al., 2004; 2007). Auch
wenn die genetischen Daten alleine keine Artunterscheidung
erlauben unterstützen sie doch die Trennung der verschiede-
nen Populationen in unterschiedliche Arten.
Durch den Vergleich der Daten von Höhlenbärenresten aus
über 30 verschiedenen alpinen und sechs außeralpinen Höh-
len ergibt sich, dass die einzelnen Arten sehr unterschiedliche
Anpassungen an die Höhenlage ihrer Äsungsgebiete zeigen.
Während die Faunen der Ursus eremus- und der ladinicus-Li-
nie eine höhenabhängige Reduktion ihrer Körpergröße („Ge-
birgsnanismus“) zeigen, ist eine derartige Korrelation bei der
U. ingressus-Gruppe nicht zu verzeichnen. Die fossilen Reste
der U. ladinicus-Gruppe (ladinischer Bär) und der U. eremus-
Gruppe (Rameschbär) sind durchschnittlich in den hoch gele-
genen Höhlen deutlich kleiner als in den Höhlen der tieferen
Lagen oder anders ausgedrückt: die Größenwerte sind mit der
Seehöhe der Höhleneingänge negativ korreliert. Bei der dritten
Art, dem erst um 50.000 Jahren vor heute eingewanderten
Gamssulzenbären (Ursus ingressus), ist keine Korrelation mit
der Meereshöhe zu erkennen. Die Mittelwerte aus der höchst
gelegenen Fundstelle (Potocˇka zijalka, 1650 m) sind kaum klei-
ner als aus der am tiefsten situierten Ilinka-Höhle (33 m).
Die Breite der Zahnkronen ist bei den Unterkiefermahlzähnen
mit der Seehöhe negativ korreliert, während dies bei den
Oberkiefermolaren nicht der Fall ist (U. ingressus) oder nur
angedeutet (U. ladinicus und U. eremus).
Bei den morphologischen Daten ist ein ganz anderer Trend
zu erkennen. Nach den verschiedenen Indices gibt es bei
U. ingressus eine starke Abhängigkeit von der Höhenlage der
Fundstellen: je höher die Höhle liegt, desto höher waren die
Evolutionsniveaus der dort hausenden Bären. Bei den beiden
anderen Arten ist diese Korrelation z.T. auch vorhanden,
z.T. aber viel schwächer. Das Evolutionsniveau einer Höhlen-
bärenfauna ist also nicht nur von der zeitlichen Stellung
sondern auch von der Seehöhe der Fundstelle abhängig.
Auch die unterschiedlichen Geschlechterverhältnisse wurden
in die Studie aufgenommen. Weder für die Mittelwerte der
Zahnmaße noch für die morphologischen Indices besteht ein
Zusammenhang mit dem Sex-Index (prozentualer Anteil an
Weibchen). Deutungsversuche für die erkennbaren Trends
werden im Kapitel „Diskussion“ gebracht, von dem sich eine
Übersetzung am Ende des Artikels findet.
PREAMBLE
For several years it was assumed that there was more
than one bear species living in the Alps in the time bet-
ween 40 and 50 ka b.p., the so called Middle-Wurmian
(Rabeder, 1995). This hypothesis has recently been
supported by DNA-analyses (Hofreiter et al., 2004; Ra-
beder et al., 2004 and Rabeder & Hofreiter, 2004), re-
sulting in the description of three new species (Rabe-
der et al., 2004). There were all in all three different spe-
cies living in the Alps: Ursus eremus,U. ladinicus and
U. ingressus, whereas the typical cave bear, U. spelaeus,
living in high-alpine habitat, is now only known from
the Préalpes in France. Moreover, this species was wi-
dely distributed outside the Alps in the mountainous
regions of Germany, France, and Spain.
Because there are many caves in the Alps which are or
were filled with bone-bearing sediments it is possible
to use statistical methods to reveal the differences bet-
ween these fossil populations (= “faunas”). The large
differences in means and distributions of the metrical
and morphological parameters cannot be explained
only by taxonomical differences, i.e. speciation. For the
parameter body size it was presumed that there is a ne-
gative correlation with the altitude of the habitat, see
Ehrenberg (1929) who introduced the term “hochalpi-
ne Kleinform”, literally translated “high-alpine small-
form”. He concluded that this kind of dwarfism is a
special adaptation to the high-alpine habitat, because
of short summers and long winters, which result in a
shortened vegetation period. A test of this hypothesis,
in combination with knowledge of the new taxonomy,
resulted only in a partial proof of Ehrenberg’s theory.
Contrary to his hypothesis, a first comparison of mor-
phodynamic indices showed a different picture, with
different trends being visible in different taxa.
In this study, we are concerned with the following que-
stions:
1. What – if any – is the correlation of the different me-
trical and morphological parameters with the taxo-
nomic attribution of the fossil bear populations on
the one hand and with the altitude of the habitat (the
cave) on the other hand?
A large number of new radiometric- and DNA ana-
lyses have shown that approximately 50 ka ago a
massive bear migrated into the Alps. This bear was
therefore named Ursus ingressus.U. ingressus mi-
grated into those regions that were formerly inhabi-
Die Höhle / 59. Jg. / Heft 1-4/2008 61
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
MATERIAL AND METHODS
This survey is based on a large amount of statistical
data obtained from fossil bear populations, which
were found in caves within and outside of the Alps.
Most of the material was excavated with novel metho-
dology. Most of these excavations were carried out un-
der the supervision of G. Rabeder (University ofVien-
na and Austrian Academy of Sciences), in cooperation
with universities, academies, museums, and collectors
from Austria and abroad, including, Croatia, France,
Italy, Slovenia, Switzerland and Ukraine. If possible
additions to the data set were made from old excava-
tions.
Only those faunas that produced enough teeth and
metapodial bones for statistical analysis were used in
this study and where the taxonomic attribution to one
of the aforementioned species is proven by means of
DNA-analyses (exceptions: Repolust Höhle, Herkova
jama and Mokriška jama) and/or trustworthy mor-
phological parameters (p4-index, P4-index, p4/4-in-
dex, m2-enthypoconid-index, M2-metaloph-index,
m3/m2-length-index, index of plumpness; see Rabe-
der (1999) and Withalm (2001).The realibility of these
indices has been tested in different ways. Thus, the
taxa belonging to diverse cave bear faunas (e.g.
Ajdovska, Divje babe, Merkensteinhöhle, Ochsenhalt-
höhle, Potocˇka zijalka, Sulzfluhhöhlen) had been re-
cognized correctly by morphodynamic indices prior
to DNA-analysis.
ted by U. eremus and probably by U. ladinicus since
80 ka b.p. (Hofreiter et al., 2004; Rabeder & Hofreiter,
2004; Rabeder et al., 2004). This interesting fact led to
the second question.
2. Are there any observable shifts in body size, plump-
ness of extremities and morphology of the teeth of
U. eremus during its replacement by U. ingressus and
are there consequently changes in the data of the lat-
ter species?
3. What are the differences of the aforementioned
three species in relation to the altitude of the habi-
tat?
4. What – if such shifts and/or differences exist – are
the probable causes of these potential adaptations
and why was U. ingressus more successful than its
predecessors were?
The question whether or not the taxa U. eremus,U. la-
dinicus and U. ingressus should be regarded as species
is not the main target of this study. However, as the ta-
xonomic distinction is of importance to the topic dis-
cussed, we would like to give a few explanations on this
subject: Ursus ingressus is defined by remarkable me-
trical, morphological, and genetic differences to U. ere-
mus. Even more significant is the fact that both species
lived side by side (sympatrically) for approximately
15 ka (from 47 ka till 31 ka b.p.) without any detectable
gene flow between the two populations. This means
that there was a behavioural mechanism to avoid in-
terbreeding between these two species of bears (Hof-
reiter et al., 2004; Rabeder et al., 2004). Consequently
this is one of the few possibilities in vertebrate palae-
ontology to apply the concept of biological species and
not only the morpho-species concept.
There is a similar situation in U. eremus and U. ladini-
cus, but the evidence is not so clear. At approximately
the same time (>50 ka b.p.) both bears lived in the “To-
tes Gebirge” (Austria). Whereas haplotypes attributa-
ble to U. ladinicus were found in Brieglersberghöhle,
those of U. eremus were found in Ramesch- and Salz-
ofenhöhle; for more details see Rabeder et al., 2005.
The problem in the case of these two forms is, that un-
til now, there is only a small number of reliable radio-
metric dates available. The reason for this is the tem-
poral limitation of the 14C-AMS-method. However, it
should be noted that occurrences of haplotypes of two
of the proposed four species in a single location have
almost never been found, and where this was the case,
it could be shown that the different species were tem-
porally separated (Hofreiter et al. 2007).This is a criti-
cal point for the analyses presented, as the four taxa
therefore represented different gene pools between
which little or most likely no gene flow took place. The-
refore, even if interbreeding had been possible, it did
not occur to an extent that would prevent different ad-
aptive responses to environmental changes, a situati-
on similar to modern brown and polar bears, which
also represent distinct gene pools, despite occasional
interbreeding between the two species.
Finally, it should be noted that it is very likely that the-
re are older synonyms for U. ingressus, which migrated
from East towards West as, for instance, Ursus spela-
eus (odessanus)” Nordmann (1858). “Odessanus” is not
a valid taxonomic name because its author, Alexander
Nordmann used this term only for a geographic des-
cription of this bear and did not erect a distinct spe-
cies or subspecies. In fact he explicitely wrote: “2. Der
Odessaer fossile Bär ist identisch mit dem Höhlenbä-
ren des übrigen Europas.” (The fossil bear from Odes-
sa is identical to the cave bear in other parts of
Europe.).
62 Die Höhle / 59. Jg. / Heft 1-4/2008
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
Fig.1: Examples of tooth evolution: one primitive and one advanced form of P4 sup. (upper row) and of M2 sup. (lower row)
Beispiele der Zahnevolution: Eine urtümliche und eine hoch entwickelte Form des P4 sup. (obere Reihe) und des M2 sup.
(untere Reihe).
Fig. 2: Metacarpale 5 dex, anterior
view from Ursus deningeri (1)
and Ursus eremus (2);
length = measured length;
deb = distal epiphyseal breath.
Messstrecken am Metacarpale 5 dex
(in Vorderansicht) von Ursus
deningeri (1) und Ursus eremus (2);
length = Länge, deb = distale
Breite an den Epiphysen.
Die Höhle / 59. Jg. / Heft 1-4/2008 63
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
Table 1: List of cave bear sites included in this study.
Name of Site Abbreviation Country Altitude Species Literature Chronology
Alpine No.
Divje babe Db SLO 450 ingressus (1, 14, 15) Middle to Late Wurmian
Brettstein-Bärenhöhle BS A, 1625/33 1700 ladinicus + eremus (2) Middle Wurmian
Brieglersberghöhle BB A 1625/24 1960 ladinicus (2, 3) Middle Wurmian
Ander dles Conturines Cu I 2800 ladinicus (4, 5) Middle Wurmian
Gamssulzenhöhle GS A, 1637/3 1300 ingressus (2, 5, 6) Middle to Late Wurmian
Hartelsgrabenhöhle HG A, 1714/1 1230 ingressus (2, 7) Middle Wurmian
Herdengelhöhle 200-330 HD 4-6 A, 1823/4 878 ingressus (2, 5) Middle to Late Wurmian
Herdengelhöhle 330-360 HD 3 A, 1823/4 878 eremus (2, 5) Middle Wurmian
Herkova jama Hj SLO 520 deningeri (18) Middle Pleistocene
Kugelstein-Tropfsteinhöhle Kst 2 A, 2784/3 482 ingressus (2) Late Wurmian
Lieglloch LL A, 1622/1 1290 ingressus (2, 5) Middle to Late Wurmian
Merkensteinhöhle Mst A, 1911/32 441 ingressus (2) Middle Wurmian
Grotte Merveilleuse Mv F 1100 ladinicus new Middle Wurmian
Mixnitz, Drachenhöhle MixJ A, 2839/1 949 ingressus (2, 8) Middle to Late Wurmian
Mokriška jama Mj SLO 1500 ingressus (17) Middle Wurmian
Nixloch NL A, 1665/1 770 ingressus (2, 5) Late Wurmian
Ochsenhalthöhle OH A, 1624/40 1660 eremus new Middle Wurmian
Potocˇka zijalka PZ SLO 1650 ingressus (9) Middle to Late Wurmian
Grotte Préletang PLE F 1225 ladinicus new Middle Wurmian
Ramesch 1, 0-50 RK 1 A, 1636/8 1960 eremus (2, 5) Middle Wurmian
Ramesch 2, 50-100 RK 2 A, 1636/8 1960 eremus (2, 5) Middle Wurmian
Ramesch 3, 100-200 RK 3 A, 1636/8 1960 eremus (2, 5) Middle Wurmian
Ramesch 4, 200-250 RK 4 A, 1636/8 1960 eremus (2, 5) Middle Wurmian
Repolusthöhle Rep A, 2837/1 525 deningeri (2, 5) Middle Pleistocene
Salzofenhöhle SO A, 1624/31 2005 eremus (2, 5) Middle Wurmian
Schreiberwandhöhle Sr A, 1543/27 2250 eremus (2) Middle Wurmian
Schwabenreithhöhle SW A, 1823/32 959 eremus (2, 5) Early Wurmian
Sulzfluhhöhlen SF CH 2300 ladinicus (2, 10) Middle Wurmian
Windener Bärenhöhle Wi A, 2911/1 190 ingressus (2) Late Wurmian
Extra-alpine
Ajdovska jama Aj SLO 244 ladinicus new Early Wurmian?
Ilinka near Odessa Ilinka UA 33 ingressus new Middle Wurmian
Krizˇna jama Kj SLO 675 ingressus (11) Middle Wurmian
Loutraki Aridea Cave LAC GR 540 ingressus (12) Middle Wurmian
Grotta di Pocala Po I 139 eremus (13) Early Wurmian?
Vindija G Vi G HR 275 ingressus (14) Late Wurmian
Relevant references: (1) Debeljak, 2002a; (2) Döppes & Rabeder, 1997; (3) Rabeder et al., 2005; (4) Rabeder, 1991; (5) Rabeder,
1999; (6) Rabeder, 1995; (7) Rabeder, 2005; (8) Abel & Kyrle, 1931; (9) Pacher et al., 2004; (10) Rabeder, 2004; (11) Pohar et al.,
2002; (12) Tsoukala & Rabeder, 2006; (13) Calligaris et al., 2006; (14) Wild et al., 2001; (15) Debeljak, 2002b; (16) Turk, 2007;
(17) Rakovec, 1976; (18) Pohar, 1981.
RESULTS
General Remarks: In the following, the results will be
presented as diagrams. All data points are significant
means in statistical sense. These data points represent
mean lengths and widths of molars and metapodial
bones as well as morphodynamic indices; see Rabeder
(1999) and Withalm (2001). For better comparability
these parameters are standardized against the means
of U. ingressus from Gamssulzenhöhle (Austria), see
64 Die Höhle / 59. Jg. / Heft 1-4/2008
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
Rabeder (1999). The used altitudes in the diagrams al-
ways represent the altitude of the main entrance of the
cave in meters above sea level.The applied regression
lines show partly low R2-values and should be regar-
ded as trend lines only. For almost all caves, DNA data
exist with regard to the question of taxa identification.
Consequently, exclusion of the few sites for which no
DNA data are available (partially due to insufficient
preservation) does not change the overall picture or
the conclusions.
Length of molars
There is a negative correlation of the molar length with
the altitude of the cave entrance in Ursus eremus and
U. ladinicus, see fig. 3 and 4. With respect to this para-
meter, the term“hochalpine Kleinform” sensu Ehren-
berg (1929) is true only for these two species. There is
no correlation between molar length and altitudes of
the entrance of the caves at all in Ursus ingressus, see
also fig. 3 and 4.
Width of molars
There is also a negative correlation between the rela-
tive width of molars and the altitude of the site. The
correlation of the aforementioned character is only
weak for the upper molars with two exceptions: the
bears from Conturines- and Schreiberwand cave. See
fig. 5 for the example of the upper M1. In general, the
correlation between the width and the altitude is much
stronger with the lower molars but is dependent on the
species: the strongest correlation is visible in U. ladi-
nicus; in contrast, the weakest correlation can be ob-
served in U. ingressus, see fig. 6.
Greatest length of metapodial bones
The greatest length of metapodial bones displays a
negative correlation with the altitude of the cave
entrance, and unlike to the situation of the molars,
all three cave-bear lineages, including U. ingressus,
follow this trend. This is shown in fig. 7, which is a
scatter-plot of means of greatest length against altitu-
de of site.
Plumpness of metapodial bones
There is a weak correlation between the index of
plumpness (ip = (distal epiphyseal width / length) x
100) of the fifth metacarpal bone and the altitude of
the site in U. ingressus and an even weaker correlation
in U. eremus and U. ladinicus, see fig. 8.
Morphodynamic indices
All morphodynamic indices of teeth show a very diffe-
rent behaviour than length and width in showing a
Fig. 3: Scatter-plot showing the
correlation of the mean length
of the upper M2 and the altitude
of the cave entrances.
Die Abhängigkeit der Zahnlänge
(Mittelwerte) des Oberkiefer-M2
von der Höhenlage der Höhlen-
eingänge.
Die Höhle / 59. Jg. / Heft 1-4/2008 65
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
Fig. 4: Scatter-plot showing the
correlation of the mean length of
the lower m2 and the altitude of
the cave entrances. The values are
standardized against the mean
tooth length of U. ingressus from
Gamssulzen cave (Lower Austria).
Die Abhängigkeit der Zahnlänge
des Unterkiefer-m2 von der
Höhenlage der Höhleneingänge.
Als Standard dienen die
Mittelwerte von U. ingressus aus
der Gamssulzenhöhle.
Fig. 5: Scatter-plot showing the
correlation of ratio of width and
length of the upper M1 and the
altitude of the cave entrances.
Das Breiten-Längen-Verhältnis des
ersten Oberkiefermolaren und die
Höhenlage der Höhleneingänge.
positive correlation with the altitude of the site. Howe-
ver, there are differences in the strength of correlation:
p4- and P4-index
There is a slow increase and only a weak correlation
between these indices and the altitude of the site in U.
eremus and U. ladinicus, but a strong one in U. ingres-
sus. The morphology of the fourth premolars has sin-
ce a long time been known to reflect the evolution of
cave bears.The emergence of additional cusps and ed-
ges is basically correlated with the chronology of the
fossil bearing strata, a fact that has been shown in the
evolution from the Middle Pleistocene U. deningeri to
the highly evolved U. ingressus of the Late Pleistocene.
66 Die Höhle / 59. Jg. / Heft 1-4/2008
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
Fig. 6: Scatter-plot showing that
the relative width of molars
depends only on the altitude
of the cave entrance and not
on the specific attribution to
an evolutionary line.
Die relative Breite der Unterkiefer-
molaren ist abhängig von der
Höhenlage der Höhlen und nicht
von der Zugehörigkeit zu einer
bestimmten Evolutionslinie.
Fig. 7: Scatter-plot showing the
correlation of greatest length of
the 5th metacarpal bone and the
altitude of the cave entrances.
Der Gesamtlänge des 5. Meta-
carpale und die Höhenlage der
Höhlenfundstellen.
However, what is different is the speed of the evolution
in the different cave bear lineages. Moreover, in addi-
tion to the temporal correlation, the morphology of the
fourth premolar also depends on the altitude of the
site; see fig. 9 and 10.
m2-enthypoconid-index
The strongest correlation is observable in U. ingressus,
with a slightly weaker correlation in U. eremus. There is
a strange correlation in U. ladinicus: while there is only
avery weak correlation between this index and the al-
titude of the site the absolute value exceeds those of all
the other bears at different altitudes.The values of en-
thypoconid-index are astonishingly high in U. ladini-
cus and are the best possibility to morphologically dif-
ferentiate between this species and U. eremus. Until
now a conclusive interpretation of this phenomenon
is missing, see fig. 11.
M2-metaloph-index
All the species of bears show a positive correlation of
this index and the altitude of the site, see fig. 12.
Die Höhle / 59. Jg. / Heft 1-4/2008 67
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
Distribution of sexes versus altitude of site
The hypothesis that the different means of metric pa-
rameters of sites in the alpine region and in the low-
lands can be explained by unequal distribution of sexes
was published by Spahni (1954). This was disproved by
Rabeder (2001) using material from high-alpine sites
(Conturines cave and Ramesch Knochenhöhle) and by
means of analyses of fossil DNA; see Hofreiter et al.
(2004) and Rabeder et al. (2004). Seventeen caves have
enough canines to obtain realistic and thus correct re-
sults. When reading the literature one can find two dif-
ferent terms for the distribution of sexes: sex ratio and
sex index.
Fig. 8: Scatter plot showing the
correlation of plumpness index
of 5th metacarpal bone and the
altitude of the cave entrances.
Der Plumpheitsindex des
5. Metacarpale und die
Hoehenlage der
Hoehlenfundstellen.
Fauna Canines n Sex- Sex- Altitude m1-length Species
Female Male ratio index in m a.s.l. standard.
Ajdovska 12 12 24 1.00 50.00 244 95.46 eremus
Brieglersberg 22 7 29 3.14 75.86 1960 96.09 ladinicus
Conturines 33 33 66 1.00 50.00 2800 93.02 ladinicus
Divje babe 322 288 610 1.12 52.79 450 100.56 ingressus
Gamssulzen 61 22 83 2.77 73.49 130 100.00 ingressus
Herdengel 24 30 54 0.80 44.44 780 97.63 eremus
Herkova jama 22 16 38 1.38 57.89 511 88.03 deningeri
Križna jama 11 34 45 0.32 24.44 670 100.82 ingressus
Loutraki 181 63 244 3.44 74.18 540 99.44 ingressus
Mixnitz ~3000 ~9000 ~12000 0.33 25.00 959 103.34 ingressus
Mokriška jama 273 477 750 0.57 36.40 1500 100.36 ingressus
Ochsenhalt 46 41 87 1.12 52.87 1650 96.32 eremus
Pocala 326 410 736 0.80 44.29 139 102.71 eremus
Potocˇka zijalka 16 47 63 0.34 25.40 1650 103.07 ingressus
Ramesch 83 71 154 1.17 53.90 1960 93.05 eremus
Repolust 23 29 52 0.79 44.23 525 88.45 deningeri
Schreiberwand 15 16 31 0.94 48.39 2250 92.86 eremus
Schwabenreith 51 28 79 1.82 64.56 960 96.00 eremus
Σwithout Mixnitz 1450 1593 3043 0.91 47.65 ——
Table 2: Distribution of sexes, size of cave bears and altitude of the sites in use.
68 Die Höhle / 59. Jg. / Heft 1-4/2008
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
Sex ratio means the quotient of female and male
canines, i.e. sex ratio = f/m where f is the number
of female and m the number of male canines.
The sex ratio follows a hyperbolic function where
rsex = f/m. A more suitable parameter is the sex index
which is calculated as isex = f/(f+m) x 100 and re-
presents the relative abundance of females in percent.
Table 2 gives the usable sites along with the number of
canines and the means of the length of the lower first
molar.
Fig. 13 is a scatter-plot of sex index against altitude of
the site and fig. 14 is a scatter-plot of sex index against
m1-length as an indicator of body size, including the
specific attribution. In both diagrams, it is obvious that
there is no or only a weak correlation at all. Fig. 14
shows only a weak correlation between the length of the
m1 and the sex index for U.ingressus, which means that
the slightly higher values from Mixnitz, Križna jama and
Potocˇka zijalka can be explained by a domination of
males.Within U. eremus the correlation is even weaker.
There are alpine faunas with a female predomination,
Brieglersberg cave for instance, and such with a male
predomination, as it is the case in Potocˇka zijalka and
such caves with an approximately equal distribution of
Fig. 9, 10: Scatter-plot showing the
correlation of the morphologic
indices of the 4th premolars and
the altitude of the cave entrances.
Die morphologischen Indices der
4. Prämolaren im Vergleich zur
Höhenlage der Höhleneingänge.
Die Höhle / 59. Jg. / Heft 1-4/2008 69
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
the sexes. Examples of the latter are Conturines cave,
Ramesch Knochenhöhle and Ochsenhalt cave.
Fig. 14 deals with the correlation between the sex in-
dex and the mean body size. When the faunas with U.
ingressus are taken into account one would tend to
confirm a correlation between sex index and mean
body size: the faunas from Potocˇka zijalka and Križna
jama are dominated by males and show slightly higher
means than their female dominated counterparts –
Gamssulzen cave and Loutraki Aridea cave. The U. la-
dinicus-faunas on the other hand show that the influ-
ence of the altitude exceeds that of the sex index. It is
obvious that the female-dominated fauna from Brieg-
lersberg cave displays higher means of m1-length than
that of the Conturines cave. At this point it should be
mentioned that the ratio of male and female bears
could be affected by a lack of data. For more informa-
tion on this topic see Rabeder (2001).
However, it is remarkable that unbalanced sex ratios
were found only in U. ingressus-faunas whereas most
of the faunas of the two other two cave-bear lineages
show roughly equal numbers of sexes.
Fig. 11: Scatter-plot showing the
correlation of the enthypoconid-
index of the lower m2 and the
altitude of the cave entrances.
Die Abhängigkeit des m2-Enthypo-
conid-Index von der Höhenlage
der Bärenhöhlen.
Fig. 12: Scatter-plot showing the
correlation of the metaloph-index
of the upper M2 and the altitude
of the cave entrances.
Die Abhängigkeit des M2-Meta-
loph-Index von der Höhenlage
der Bärenhöhlen.
70 Die Höhle / 59. Jg. / Heft 1-4/2008
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of Cave Bears to High-alpine Habitats
The reduction of body size in high-alpine cave bears
(alpine nanism) is commonly seen as an adaptation
to the less favourable climate, i.e. short summers
and long winters. It is easier for smaller animals to
nd enough food and thus it is easier for them to fulfil
the needs of metabolism during winter. For two linea-
ges of cave bears, U. eremus and U. ladinicus, it is
possible to follow the hypothesis of Ehrenberg (1929)
because there are no contradictory data. Ursus ingres-
sus on the other hand, found another way of adapting
to the (high-) alpine conditions: an improved per-
formance of mastication is visible in all morpho-
dynamic indices and is obviously aimed at an in-
creased daily food intake during the vegetation period
and thus at a sufficient fat storage before autumn.
The variability of morphologic parameters is very high
in U. ingressus, especially those of the indices of
the fourth premolars, which is much higher than in the
other taxa. A possible explanation for this pheno-
menon can be found in the fast adaptation of
U. ingressus-group to high-alpine habitats. This stra-
tegy was obviously successful: U. ingressus survived
DISCUSSION
Fig. 13: The sex ratio of cave
bears in relation to the altitude
of the cave entrances.
Das Geschlechterverhältnis der
Höhlenbären im Bezug zur
Höhenlage der Fundstellen.
Fig. 14: The sex ratio of cave
bears in relation to the length
of the lower m1.
Geschlechterverhältnis der
Höhlenbären im Bezug zur
m1-Länge.
Die Höhle / 59. Jg. / Heft 1-4/2008 71
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
Abel, O. & Kyrle, G., (1931): Die Drachenhöhle bei Mixnitz. –
Speläolog. Monogr., 7-9, Wien.
Calligaris, R., Rabeder, G, & Salcher,T. (2006): Neue
paläontologische Grabungen in der Grotta Pocala bei
Triest. – [in:] Ambros, D., Gropp, C., Hilpert, B. & Kaulich,
B. (eds.) Neue Forschungen zum Höhlenbären in
Europa. – Naturforsch. Ges. Nürnberg, Abh., 45/2005:49–
56, Nürnberg.
Debeljak, I. (2002a): Mortality Dynamics and Paleoecology
of Cave Bear from Divje babe I Site. – Ph. D. Thesis,
Faculty of Natural Sciences and Engineering, University
of Ljubljana.
Debeljak, I. (2002b): Fossil population structure of the cave
bear from Divje babe I site, Slovenia: Preliminary results.
– Abh. zur Karst- und Höhlenkunde (München), 34: 41–
48.
Döppes, D. & Rabeder, G. (eds.) (1997): Pliozäne und
pleistozäne Faunen Österreichs. Ein Katalog der
wichtigsten Fossilfundstellen und ihrer Faunen. – Mitt.
Komm. Quartärforsch. Österr. Akad. Wiss., Wien, 10: 267.
Ehrenberg, K. (1929): Die Ergebnisse der Ausgrabungen in
der Schreiberwandhöhle am Dachstein. – Paläont. Z., 11
(3):261–268.
Hofreiter, M. (2005): Evolutionsgeschichte alpiner
Höhlenbären aus molekulargenetischer Sicht. – Mitt.
Komm. Quartärforsch. Österr. Akad.Wiss., 14: 67-72.
Hofreiter, M., Rabeder, G., Jaenicke, V., Withalm, G., Nagel,
D., Paunovic, M., Jambr si , G. & Pääbo, S. (2004):
Evidence for reproductive isolation between cave bear
populations. – Current Biology, 14 (1): 40–43.
Hofreiter, M., Münzel, S., Conard, N., Pollack, J.,Weiss, G. &
Pääbo, S. (2007): Sudden replacement of cave bear
mitochondrial DNA in the Late Pleistocene. – Current
Biology, 17 (4).
Nordmann, A. von, (1858): Palaeontologie Südrusslands.
Helsingfors.
Pacher, M., Pohar,V. & Rabeder, G. (eds.) (2004): Potocˇka
zijalka. Palaeontological and archeological results of the
campaigns 1997-2000. – Mitt. Komm. Quartärforsch.
Österr. Akad. Wiss., 13: 1–245.
Pohar,V. (1981): La faune pléistocene de la cavité de jama
pod Herkovimi pecmi. – Geologija 24(2): 241–284.
Pohar,V., Rabeder, G. Kralj, P. & Misic, M. (2002): Cave
sediments and fossil mammal remains in Križna jama,
Southern Slovenia. – [in:] Rosendahl,W. Morgan, M. &
Lopez Correa – Abh. Karst- Höhlenkde, München, 34:
49–51.
Rabeder, G. (1991): Die Höhlenbären von Conturines.
Entdeckung und Erforschung einer Dolomiten-Höhle in
2800 m Höhe. – (Athesia-Verl.), Bozen.
Rabeder G., (ed.) (1995): Die Gamssulzenhöhle im Toten
Gebirge. – Mitt. Komm. Quartärforsch. Österr. Akad.
Wiss., 9: 1–133.
Rabeder, G. (1999) Die Evolution des Höhlenbärengebisses.
– Mitt. Quartärkomm. Österr. Akad. Wiss., 11: 1–102.
Rabeder, G. (2001) Geschlechtsdimorphismus und
Körpergröße bei hochalpinen Höhlenbärenfaunen. –
Beitr. Paläont., 26: 117–132.
Rabeder, G. (2004) Die Höhlenbären der Sulzfluh-Höhlen. –
Vorarlberger Naturschau (Dornbirn), 15: 103-114.
Rabeder, G., (2005): Neue paläontologische Daten von der
Bärenhöhle im Hartelsgraben (1714/1), Gesäuseberge,
Steiermark. – Die Höhle, 56 (1): 44–46.
Rabeder, G. & Hofreiter, M. (2004): Der neue Stammbaum
der Höhlenbären. – Die Höhle 55, (1-4): 58–77.
Rabeder, G., Hofreiter, M. Nagel, D. &Withalm G. (2004):
New Taxa of Alpine Cave Bears (Ursidae, Carnivora). –
Cahiers scientif. / Dép. Rhône - Mus. Lyon, Hors série,
n° 2 (2004): 49–67.
Rabeder, G., Hofreiter, M. &Wild, E. (2005): Die Bären
der Brieglersberghöhle (1625/24). Die Höhle 56, (1):
36-43.
We are deeply indebted to Prof. David K. Ferguson for
the correction of the manuscript. We are also grateful
for the reviews of our colleagues Alain Argant (Greno-
ble), Gennady Baryshnikov (St. Petersburg) and Vida
Pohar (Ljubljana). Moreover, we would like to thank
Bernard Rafienna (St. Pierre de Chartreuse, France) for
the possibility to study the fossil bears from the caves
of the Massif Vercors.
This study is part of the project F.A.C.E. (Fossil Animals
of Caves in Europe) Austria of the Commission of Qua-
ternary Research, Austrian Academy of Sciences and
was partly financed by the Department of Cultural
Affairs of the Government of Lower Austria (Project
LNOe0038).
the Plenigacial in the alpine regions and only became
extinct approximately 15 ka bp. In contrast, U. eremus
became extinct approximately 30 ka bp and U. ladini-
cus even earlier.
In U. ingressus there is also a high degree of variability
in the metric parameters, but they are not correlated
with the altitude of the habitat. The factors that influ-
ence this development are to date unknown. It could
be possible that there are differences in the diet of the
cave bear lineages which could be unveiled by means
of stable isotope analysis (13C, 15N, 18O). Actually, the
information on stable isotopes is available for only two
alpine cave bear sites. Future projects will be targeted
on this subject.
ACKNOWLEDGEMENTS
REFERENCES
DISKUSSION
Rakovec, I. (1976):The cave bear from the Mokrica cave in
the Savinja Alps (Slovenia,Yugoslavia).) – Razpr. IV. razr.
SAZU 10:123–203.
Spahni, J.-C. (1954) : Les gisements à ursus spelaeus de
l’Autriche et leurs problèmes. – Bull. Soc. Préhist. Franc.
(Le Mans), LI, 7:346–367.
Tsoukala, E. & Rabeder, G. (2006): Cave bears and Late
Pleistocene associated faunal remains from Loutraki
Aridea (Pella, Macedonia, Greece). – [in:] Ambros, D.,
Gropp, C., Hilpert, B. & Kaulich, B., (eds.). Neue
Forschungen zum Höhlenbären in Europa. – Natur-
forsch. Ges. Nürnberg, Abh., 45/2005: 225–236.
Turk, I. (ed.) (2007): Divje babe I – Upper Pleistocene
Palaeolithic site in Slovenia; Part I: Geology and
Palaeontology, ZRC SAZU (opera instituti archaeologici
Sloveniae), Ljubljana.
Wild, E., Paunovic, M., Rabeder, G., Steffan I. & Steier, P.
(2001): Age determination of fossil bones from the
Vindija Neandertal site in Croatia. – Radiocarbon, 43, 2B:
1021–1028.
Withalm, G. (2001): Die Evolution der Metapodien in der
Höhlenbären-Gruppe (Ursidae, Mammalia). – Beitr.
Paläont., 26: 126–249.
Die Verkleinerung der Körperdimensionen bei hochal-
pinen Bären („Gebirgsnanismus“) wird allgemein als
Anpassung an das ungünstigere Klima (kurze Sommer,
lange Winter) gedeutet. Kleinere Tiere können ihren
Bedarf an Nahrung und damit an Energie leichter dek-
ken als große. Für die beiden Linien von Ursus eremus
und von Ursus ladinicus kann diese Hypothese beibe-
halten werden, weil keine widersprüchlichen Daten
vorliegen. Die Bären der Ursus ingressus-Gruppe hin-
gegen haben einen anderen Weg der Gebirgsanpas-
sung beschritten. Eine Verbesserung der Kauleistung
ist von allen morphodynamischen Indices abzulesen
und zielt darauf ab, die Menge der pro Vegetationstag
aufgenommenen Nahrung zu steigern und damit auch
eine größere Fettspeicherung bis zum Herbst zu ge-
währleisten. Bei U. ingressus variieren die morphologi-
schen Werte, besonders die Indices der Prämolaren,
wesentlich stärker als bei den anderen Arten. Mögli-
cherweise ist dies auf die höhere Evolutionsgeschwin-
digkeit zurück zu führen, mit denen sich die Bären der
U. ingressus-Gruppe an das Gebirgsleben angepasst
haben. Offensichtlich war diese Strategie erfolgreich:
U. ingressus überlebte in den Alpen das Hochglazial
und starb erst vor etwa 15.000 Jahren BP aus, während
U. eremus schon vor ca. 30.000 Jahren verschwunden
ist, U. ladinicus wahrscheinlich schon früher.
Auch bei U. ingressus streuen die metrischen Daten
beträchtlich, sie sind aber nicht höhenabhängig,
sondern werden offensichtlich von Faktoren gesteuert,
die noch unbekannt sind. Es wäre möglich, dass es
auch Unterschiede in der Ernährung gibt, die man
eventuell durch die Analyse der stabilen Isotopen (13C,
15N, 18O) herausfinden könnte. Derzeit sind nur die
Isotopenverhältnisse von zwei alpinen Höhlen be-
kannt. Geplante Projekte sollen dieser Fragestellung
nachgehen.
Rabeder, Debeljak, Hofreiter, Withalm / Morphological responses of cave bears to high-alpine habitats
72 Die Höhle / 59. Jg. / Heft 1-4/2008
... Also according to this index, the fauna of Medvedia jaskyňa Cave is within the height-dependent distribution of the values of alpine cave bear faunas (Fig. 4). The development of a metaloph at the M2 in the faunas of Ursus ingressus is positively correlated with the height of the cave entrances RABEDER et al., 2008;LAUGHLAN, 2012). ...
... 4). Vývoj metalofu na M2 vo faunách s Ursus ingressus pozitívne koreluje s nadmorskou výškou vchodov do jaskýň RABEDER et al., 2008;LAUGHLAN, 2012). ...
Chapter
Full-text available
The main subject of a recent study of cave bears from Medvedia jaskyňa Cave in the Slovenský raj Mts. is the dentition that was found in large numbers in the cave sediments, including hundreds of deciduous and permanent teeth. During the field campaign in 2007 – 2009, many isolated permanent teeth have been excavated together with these situated still in jaws, or in their fragments . This large number allow representative statements about the evolutionary level of the bears and their sex-ratios.
... The mean values of the standardized length and width values are 7-9 % below the standards of Ursus ingressus from the Gamssulzen cave (Rabeder, 1999) and correspond to the type of a high alpine cave bear, which was also called »high alpine small form«, sensu Ehrenberg. There is a positive correlation between the dimensions of the cave bear teeth and the height of the sites for Ursus ingressus, and a negative correlation for the subspecies of Ursus spelaeus (Rabeder et al., 2008). In comparison with other alpine and extra-alpine cave bear faunas, the Potentialschacht (PS) lies in the common cluster of Ursus s. eremus and U. s. ladinicus. ...
Article
Full-text available
Abstract: On the Hochschwab karst massif in the east of the Northern Calcareous Alps (NCA), the remains of four cave bear cubs were found in the subvertical cave Potentialschacht that opens at 2,070 m a.s.l. There is severe pathology on two individually related mandibular fragments that indicate that the cubs could have fallen into the cave through narrow gaps. There are no remains of adult bears, however. According to the morphology of the teeth, DNA analysis and 14C data, these juvenile bones and teeth belong to Ursus spelaeus eremus, which lived here at least from 46,000 to 39,000 years before present. This proves that all large plateaus of the NCA were inhabited by cave bears of Ursus spelaeus group during the so-called »Cave Bear Era« of the middle Late Pleistocene. It will be discussed, which palaeoclimatological conclusions can be drawn from it.
... Zur besseren Vergleichbarkeit der einzelnen morphodynamischen Indices werden die Werte standardisiert, wobei man als Standards die Werte der Gamssulzenfauna (Typusfauna von U. ingressus) heranzieht (RabeDeR1999). Die Höhenlage der Fossilfundstellen ist ein wichtiger Faktor der Evolutionsforschung alpiner Höhlenbären, weil sich gezeigt hat, dass nicht nur die Dimension sondern auch die morphologischen Indices z. T. sehr eng mit der Höhenlage korreliert sind (RabeDeR et al. 2008). In Diagrammen werden die Gesamtwerte der einzelnen Höhlenbärenfaunen verglichen. ...
Article
Full-text available
Zusammenfassung Die taxonomische Zuordnung des Höhlenbärenmaterials aus der Schusterlucke ist wegen seines relativ geringen Umfanges nicht eindeutig. Die taxonomische Stellung wird an Hand von drei Diagrammen diskutiert. Die Zuordnung zu Ursus ingressus erscheint wahrscheinlicher als zu Ursus spelaeus ladinicus. Abstract The cave bears from Schusterlucke in the Krems valley (Lower Austria) discussion of taxonomic assignment The numerical extent of the cave bear material from Schusterlucke is too small for a clear determination. The taxonomic position is checked by means of three diagrams. An assignment to Ursus ingressus seems more likely than to Ursus spelaeus ladinicus.
... It can be argued that the similar high variable intraspecific genetic structure of the brown bear U. arctos (Hirata et al., 2013) warrants caution when identifying putative species status for distinct cave bear haplogroups . This considerable and evident morphological variation between bear populations and the controversial nature of taxonomic affinities in cave bear groups is most clearly illustrated by the diminutive 'Alpine' Spelaean form Rabeder et al., 2008;Grandal d'Anglade, 1993;Rossi & Santi, 2001;Carlis et al., 2005;Withalm, 2005;Baryshnikov & Puzachenko, 2011;Seetah et al., 2012). A north-south geographic separation was noted following morphometric analysis of cheek teeth from Belgium, Poland and Russia (northern assemblages) and Spain, France and Ukraine (southern sites) (Baryshnikov et al., 2003). ...
Thesis
An important debate amongst cave bear researchers is the diet and ecoethology of these animals. A great controversy surrounds the phylogenetic resolution of the cave bear lineage. Previous investigations have focussed on teeth, skull and mandibular bones. This study focuses on the morphology of metapodia as an alternative, and for the first time 3D digital methods have been applied on these elements. This study involved 3D scanning of metacarpals and metatarsals belonging to Ursus spelaeus, Ursus arctos and U. deningeri, followed by landmarks assignment onto the surface of virtual bones...
... The metaloph index of M2 sup describes the characteristics of the connection between metaconus and the lingual cusp of the protoconus respectively hypoconus (Rabeder 1999). This index is in all taxa directly correlated with the altitude of the cave bear site (Rabeder et al. 2008). The data point of the Torrener Bärenhöhle lies far outside the 'ingressus cluster' (figure 9; Suppl. ...
Article
The Torrener Bärenhöhle (Cave of Torren) is an alpine bear cave in the limestone Alps near Salzburg (Austria). The entrance of this cave is located in the riverbed of a periodically flowing stream that floods the caverns during snowmelt or after heavy rainfall. Due to these flooding events, the fossiliferous layers were repeatedly destroyed and the fossil remains – mainly cave bear bones – were distributed over the cave’s entrance hall. The fossil collections have taken place since 1924. During the last decades, numerous bones scattered over the cave floor have been collected again and again. In this study, we conduct metric and morphological analyses of this fossil material as well as DNA analyses in an attempt to clarify the taxonomic position of the cave bear remains. The chronological status of the bear remains has not yet been clarified, because the few samples that have been analysed so far were beyond the range of 14C dating, indicating they are older than 49,000 cal yr BP. The following taxa are represented in small numbers: Brown bear (Ursus arctos L.), wolf (Canis lupus L.), cave lion (Panthera spelaeus Goldfuss, 1810), European bison (Bison bonasus L.) and beaver (Castor fiber L.).
... The foraging strategy of an animal is largely determined by body mass via energetic needs 72 . Cave bear populations were likely adapted to their local habitats, with different altitudes being associated with different features, such as body size, tooth morphology, and tooth wear patterns (e.g., ref. 21,33,73 ). Although the tooth microwear and the isotopic composition of bone collagen reflect the diet of the analyzed individuals at different pre-mortem time periods (e.g., several weeks vs several years or more before the death 74 ), the observed variation in the bulk collagen δ 15 N values for cave bears in the Romanian region possibly reflected a difference in foraging strategy on plants associated with different body sizes 24 . ...
Article
Full-text available
An amendment to this paper has been published and can be accessed via a link at the top of the paper.
... Numerous recent studies of ancient mtDNA show significant genetic diversity in the cave bear group (Valdiosera et al. 2006;Rabeder et al. 2008Rabeder et al. , 2010Knapp et al. 2009;Dabney et al. 2013;Stiller et al. 2014;Knapp 2018;Gretzinger et al. 2019). There are several recognized evolutionary lineages or separate species: Ursus spelaeus Rosenm€ uller, 1794 (Late Pleistocene, western Europe and west Siberia), including subspecies U. s. eremus Rabeder, Hofreiter, Nagel et Withalm, 2004 (the Alps and Altai Mountains) and U. s. ladinicus Rabeder et al., 2004 (highlands in Alps), U. deningeri von Reichenau, 1904 (Middle Pleistocene, Europe and central Asia), including subspecies U. d. batyrovi Baryshnikov 2007 (Kirghizia), U. kanivetz Vereshchagin, 1973(= U. ingressus Rabeder et al., 2004 (Late Pleistocene, central and eastern Europe and the Urals), and U. kudarensis Baryshnikov, 1985 including subspecies U. k. praekudarensis (Baryshnikov 1998) (Middle and Late Pleistocene, Caucasus and, probably, eastern Siberia). ...
Article
Full-text available
The variability of lower 6776 cheek teeth of spelaeoid (Ursus spelaeus, U. kanivetz, U. deningeri, U. kudarensis, U. savini and U. rossicus) and arctoid lineages (extant and extinct U. arctos) recovered from 176 palaeontological localities was analysed by methods of univariate and multivariate statistics. For comparison, teeth of U. minimus and U. etruscus species, which are ancestral taxa for cave bears and brown bear, were studied. The performed analyses indicate great variability of tooth crown in lower cheek teeth; no distinct general trend was identified in the changes, except for an increase in p4 roundness and in m3 size from U. minimus to big cave bears. Centroids of U. minimus and U. etruscus are at a great distance from those of cave bears; U. arctos occupies an intermediate position between the two groups. Molar characteristics in general are close in U. savini and U. rossicus. The bear species are differentiated in the morphospaces by morphometric variations in the lower cheek teeth but less markedly than in the morphospaces of the upper cheek teeth. A test of the inhibitory developmental cascade model showed that a linear relationship between lower molar dimensions became clear only after premolar p4 was included in the model. Evolutionary trends were found in the tooth changes with time (differently expressed in individual species and in different teeth) depending on environments, in particular on the elevation of the locality above the sea level. The allometric pattern differentiation does not always coincide with phylogenetic relationships between taxa, being most often revealed at the subspecies level. There is a dominant covariance between the upper and lower rows of teeth in accordance with the occlusion scheme. It may vary in manifestation depending on the particular species.
... The foraging strategy of an animal is largely determined by body mass via energetic needs 72 . Cave bear populations were likely adapted to their local habitats, with different altitudes being associated with different features, such as body size, tooth morphology, and tooth wear patterns (e.g., ref. 21,33,73 ). Although the tooth microwear and the isotopic composition of bone collagen reflect the diet of the analyzed individuals at different pre-mortem time periods (e.g., several weeks vs several years or more before the death 74 ), the observed variation in the bulk collagen δ 15 N values for cave bears in the Romanian region possibly reflected a difference in foraging strategy on plants associated with different body sizes 24 . ...
Article
Full-text available
Heavy reliance on plants is rare in Carnivora and mostly limited to relatively small species in subtropical settings. The feeding behaviors of extinct cave bears living during Pleistocene cold periods at middle latitudes have been intensely studied using various approaches including isotopic analyses of fossil collagen. In contrast to cave bears from all other regions in Europe, some individuals from Romania show exceptionally high δ¹⁵N values that might be indicative of meat consumption. Herbivory on plants with high δ¹⁵N values cannot be ruled out based on this method, however. Here we apply an approach using the δ¹⁵N values of individual amino acids from collagen that offsets the baseline δ¹⁵N variation among environments. The analysis yielded strong signals of reliance on plants for Romanian cave bears based on the δ¹⁵N values of glutamate and phenylalanine. These results could suggest that the high variability in bulk collagen δ¹⁵N values observed among cave bears in Romania reflects niche partitioning but in a general trophic context of herbivory.
Article
Out of all extinct megafaunal mammals of the Quaternary, the cave bear Ursus spelaeus is one of the best represented in the fossil record. This species has been found to exhibit skeletal morphological adaptations when exploiting varied environmental niches, be that spatially or temporally. Here, we employ geometric morphometrics and phenotypic trajectory analysis to explore temporal morphological changes across the entire lower molar tooth row from the infilling of Scladina Cave, Belgium. We show that molar tooth size increases from Marine Isotope Stage (MIS) 5 – MIS 3, with cusp position varying temporally in relation to a larger talonid grinding platform in later time periods. Phenotypic trajectory analyses further show similar evolutionary shape trajectories in the first and second molars, but not in the third molar. Morphological changes related to a larger grinding platform are found in the second and third molars, with the divergent morphological change of M3 suggesting that this tooth is less constrained and appears much more responsive to environmental changes. The need to cope with harder fibrous plant material present in the palaeoenvironment potentially constrained morphological evolution of the cave bear until its disappearance throughout Europe.
Article
Full-text available
This article gives a perspective of several years of morphological and morphometrical studies of the paleontological collection of Ursus spelaeus’s remains founds in the Pocala cave, kept in the Civic Museum of Natural History of Trieste. The aim of this research is to develop a new metrical approach for the study of the skull and dentition of Ursus spelaeus, based on previous morphological and morphometrical studies. Goal of the present paper is an easy sexing of the remains and giving a hint of the homogeneity of the population researched.
Thesis
Full-text available
For this investigation an amount of 4459 metapodial bones of Ursus deningeri and Ursus spelaeus was treated. Up to eight measurements were taken and four indices were calculated to follow the evolution of metacarpus and metatarsus and the development of the metapodials in particular. The material was taken from eight different sites in Austria and Southern Tyrol (Italy), differing in age and altitude. These sites are: Hundsheimer Spalte (Lower M–Pleistocene, n < 30), Repolust cave (Upper M–Pleistocene), Schwabenreith cave (Early Wurmian), Herdengel cave (Early – Middle Wurmian), Ramesch–Knochenhöhle (Middle Wurmian, two separated stratigraphic levels, high alpine), Conturines cave (Early – Middle Wurmian, Southern Tyrol, high alpine), Windener Bärenhöhle (Middle Wurmian) and Gamssulzen cave (Late Wurmian, the standard for cave bear faunas, all other faunas will be related to the population of this cave). The following points were considered: 1. Is there an evolution of metapodial bones leading from Ursus deningeri to Ursus spelaeus and if so, is this evolution related to the evolution of the P4/4–Index ? 2. Are there any peculiarities in the material from the high alpine cave bear faunas ? Which position do these bears obtain within the cave bear group ? 3. Is it possible to determine the sex–ratio by means of metapodial bones ? 4. As it became obvious that there is one more species of bear in the Windener Bärenhöhle these findings are described as well.
Article
Full-text available
Vindija cave in Croatia is famous for the Neanderthal bones found in layer G of its sediment profile. Radiocar- bon dating has been performed mainly on this layer due to the great interest in its fossils. In addition to Neanderthal remains , the sediment in layer G contains bones from the cave bear. Cave bear bones are found also in other layers of the sediment pro- file and offer the possibility of studying the bears' evolutionary mode. Therefore, we tried to determine the time span covered by the entire profile. The U/Th age determination method was applied to cave bear bones from different layers of the profile. For the younger part of the profile, the U/Th ages were compared with the results of the 14C and the amino-acid racemization method. The agreement of the different methods indicates that closed-system behavior can be assumed for the fossil bones from Vindija cave. From this finding it may be deduced that bones from the lower sediment layers are also closed systems and that the U/Th ages of these layers are reliable. This conclusion is corroborated by the stratigraphy of the cave profile.