ArticlePDF Available

Differential mesowear in the maxillary and mandibular cheek dentition of some ruminants (Artiodactyla)

  • Leibniz-Institut zur Analyse des Biodiversitätswandels

Abstract and Figures

The mesowear method assesses the dietary regime of herbivorous mammals based on the attrition/abrasion equilibrium by evaluating cusp shape and relief of upper second molars. The method has recently been extended to include four tooth positions, upper P4-M3, in equids. In this study we determine whether the method can be extended in ruminants by applying it to maxillary and mandibular dentitions of a browser, the giraffe (Giraffa camelopardalis) and two mixed feeders, the oribi (Ourebia ourebi) and the musk ox (Ovibos moschatus). We find that including the upper third molar in addition to the upper second molar provides consistent mesowear classifications in these species. Lower dentitions of mixed feeders score significantly differently in terms of mesowear as compared with upper dentitions. We infer that adaptive optimization in differential anisodonty is related to the composition of the diet and should be mirrored in differential mesowear signals of adjoining upper and lower molars. Our results suggest that in mixed feeders, sharpness is maximized in upper teeth, whereas in specialized feeders this is not the case.
Content may be subject to copyright.
Ann. Zool. Fennici 40: 395–410 ISSN 0003-455X
Helsinki 27 October 2003 © Finnish Zoological and Botanical Publishing Board 2003
Differential mesowear in the maxillary and mandibular
cheek dentition of some ruminants (Artiodactyla)
Tamara A. Franz-Odendaal1 & Thomas M. Kaiser2
1) Department of Zoology, University of Cape Town, Rondebosch Private Bag 7700, South Africa;
current address: Biology Department, Dalhousie University, 1355 Oxford Street, Halifax NS,
Canada B3H4JI (e-mail:
2) Zoological Institute and Museum, University of Greifswald, Johann-Sebastian-Bach Str. 11–12,
D-17489 Greifswald, Germany (e-mail:
Received 12 May 2003, revised version received 7 June 2003, accepted 7 June 2003
Franz-Odendaal, T. A. & Kaiser, T. M. 2003: Differential mesowear in the maxillary and mandibular
cheek dentition of some artiodactyls. — Ann. Zool. Fennici 40: 395–410.
The mesowear method assesses the dietary regime of herbivorous mammals based on
the attrition/abrasion equilibrium by evaluating cusp shape and relief of upper second
molars. The method has recently been extended to include four tooth positions, upper
P4–M3, in equids. In this study we determine whether the method can be extended
in ruminants by applying it to maxillary and mandibular dentitions of a browser, the
giraffe (Giraffa camelopardalis) and two mixed feeders, the oribi (Ourebia ourebi) and
the musk ox (Ovibos moschatus). We nd that including the upper third molar in addi-
tion to the upper second molar provides consistent mesowear classi cations in these
species. Lower dentitions of mixed feeders score signi cantly differently in terms of
mesowear as compared with upper dentitions. We infer that adaptive optimization in
differential anisodonty is related to the composition of the diet and should be mirrored
in differential mesowear signals of adjoining upper and lower molars. Our results sug-
gest that in mixed feeders, sharpness is maximized in upper teeth, whereas in special-
ized feeders this is not the case.
Artiodactyla and some Perissodactyla have cer-
tain biomechanical adaptations of their dentition
in common, which are closely related to their
particular herbivorous diets. In all these animals,
occlusal surfaces of opposing teeth (left and
right) are inclined toward each other in a tecti-
form fashion. This allows a one-phase upward-
inward occlusal stroke, which produces a bucco-
lingually straight occlusal surface. Our approach
here is restricted to forms with transverse chew-
ing, as the method applied here, the mesowear
method, cannot as yet be applied to animals with
chewing strokes parallel to the tooth row (such
as in elephants). A basic requirement for trans-
latory chewing is anisodonty, the differential
width of occluding teeth that allows one tooth
row to move across the other while maintain-
ing occlusal contact. That anisodonty is clearly
related to chewing dynamics is shown by the
fact that it has a strong relationship to the spac-
ing of the tooth rows or anisognathy. Anisodonty
and anisognathy are strongly correlated in forms
396 Franz-Odendaal & Kaiser ANN. ZOOL. FENNICI Vol. 40
with a two-phase occlusal morphology, whereas
in forms with a single-phase morphology, such
as hypsodont equids, the precise relationship
disappears although anisodonty itself remains
(see detailed discussion in Fortelius 1985: p. 53).
Anisodonty is a general feature of mammals
with translatory chewing and is invariably posi-
tive, upper teeth being wider than lowers.
Wear in hypsodont molars has two sources,
abrasion (tooth-to-food contact) and attrition
(tooth-to-tooth contact). While abrasion is the
result of grinding action that takes place when
food inclusions, such as phytoliths or grit that
remove softer dental tissues, contact the occlu-
sal surface under pressure, attrition is the result
of immediate tooth-to-tooth contact. Mesowear
analysis is a relatively new technique used to
assess the dietary regime of herbivorous mam-
mals based on the attrition/abrasion equilibrium
(Fortelius & Solounias 2000). The method
evaluates cusp shape and cusp relief of upper
second molars (txM2) and is able to correctly
classify species into expected feeding groups.
Mesowear analysis has been applied to fossil
and extant Equidae (Bernor et al. 1999, Kaiser et
al. 2000a, 2000b, Kaiser 2001, Kaiser & Bernor
2001a, 2001b, Kaiser 2003, Kaiser et al. 2003,
Kaiser & Fortelius 2003; T. M. Kaiser & T. A.
Franz-Odendaal unpubl. data) to fossil Cervi-
dae (Kaiser & Croitor 2002; T. M. Kaiser & R.
Croitor unpubl. data) as well as fossil giraf ds
(Franz-Odendaal 2002). Recently Kaiser and
co-authors (Kaiser & Fortelius 2003, Kaiser &
Solounias 2003) have extended the mesowear
method to include the upper tooth positions
P4–M3 for equids only.
In this study, we apply the mesowear method
to the rest of the upper dentition and to the second
molar of some ruminants in order to determine
whether the method can be extended to include
other tooth positions in these animals. We test
taxa from each of the major trophic groups
— a browser (giraffe, Giraffa camelopardalis,
Gc) and two mixed feeders (musk ox, Ovibos
moschatus, Om, and oribi, Ourebia ourebi, Oo)
— and compare differences with grazers (Equus
burchelli, Eb) for which the method has already
been extended (Kaiser & Fortelius 2003, Kaiser &
Solounias 2003). Extending the mesowear method
is important if fossil assemblages with limited
tooth specimens are to be analysed in the future.
A second focus of this paper is to assess
whether a difference in the mesowear signal is
found if upper and lower dentitions are com-
pared. Kaiser and Fortelius (2003) have observed
that after a certain amount of wear in both zebras
(Equus burchelli) and in the extinct Miocene
hipparionine horse (Hippotherium primigenium),
cusp apices of upper and lower molars show
obvious differences in their distinctness and
sharpness. These differences appear to be highly
consistent, at least in hypsodont equids. No other
ungulates however, have so far been investi-
According to Kaiser and Fortelius (2003),
gravity, which will always ensure that unloaded
food will more frequently contact the lower than
the upper teeth, is a major factor responsible for
differences observed in the mesowear signature
of upper and lower dentitions. Another factor
these authors consider is occlusal geometry.
They expect occlusal geometry to result in differ-
ent loading patterns on the cutting edges of upper
and lower teeth, for example towards versus
away from the supporting dentine (Fortelius
1985: p. 59). In this context, adaptive optimisa-
tion of enamel ultrastructure is considered a pos-
sible source of differential wear control (Rens-
berger & von Koenigswald 1980, Fortelius 1985,
von Koenigswald & Sander 1997). It seems
plausible that a general explanation of posi-
tive anisodonty (upper teeth wider than lower
teeth) could be derived from basic relationships
imposed by position and geometry. Kaiser and
Fortelius (2003) suggest that food polishing acts
on lower teeth when the teeth are not in occlu-
sion and might explain why the polarity of the
anisodonty is initially set as positive. They pro-
pose the hypothesis that an arrangement maxim-
ising sharpness in upper teeth would be superior
to one that maximises sharpness in lowers, since
lowers are more in uenced by food abrasion
between occlusal events.
In summary, this study addresses the follow-
ing questions:
Can the mesowear model be extended to
include more upper teeth in ruminants?
What is the difference in mesowear scorings
for upper and lower cheek teeth in ruminant
ANN. ZOOL. FENNICI Vol. 40 Application of mesowear to ruminants 397
• How do these relate to differential anisodonty
with regard to the composition of the diet?
Materials and methods
We investigate three assemblages of extant
ruminants — Ovibos moschatus, Ourebia ourebi
and Giraffa camelopardalis. Datasets of Equus
burchelli from Kaiser and Fortelius (2003) are
incorporated in this study for comparative pur-
poses only.
All specimens investigated are wild shot
animals. The sample of Giraffa camelopar-
dalis (giraffe) comprises 91 upper premolars and
molars and 17 lower second molars. These speci-
mens are housed at the Museum für Naturkunde,
Berlin (Germany). The sample of Ovibos mos-
chatus (musk ox) consists of 238 upper molars
and premolars and 32 lower second molars. This
population is derived from Greenland and is
curated at the Zoological Museum of the Uni-
versity of Copenhagen (Denmark). The sample
of Ourebia ourebi (oribi) is from the Museum
für Naturkunde, Berlin (Germany) and consists
of 152 upper premolars and molars and 29 lower
second molars. In total, over 500 teeth were ana-
lysed (see the appendix for accession numbers).
We compare mesowear signatures of upper
(tx) and lower (tm) second molars in these three
populations of ungulates. In addition, we test
combinations of all six upper tooth positions
(txP2–M3) for their consistency in correctly
classifying species as compared with the model
tooth position (txM2) on which the original
mesowear method of Fortelius and Solounias
(2000) was based. In this study, we follow the
methodology by Kaiser and Solounias (2003).
In order to extend the mesowear model to more
than one tooth position in ruminant populations,
we test two extended models. The four-tooth
model (txP4 + txM1 + txM2 + txM3), which is
the “extended” model proposed for hypsodont
equids (Kaiser & Solounias 2003), and a two-
tooth model (txM2 + M3), which we nd to show
high consistency in mesowear classi cations of
the three ruminant species investigated. We refer
to these two models as tx(P4–M3) and tx(M2 +
M3), respectively. As comparison data we use
the mesowear data for 28 extant species from
Fortelius and Solounias (2000) as well as the
data provided by Kaiser and Fortelius (2003) for
41 upper and 42 lower second molars of Equus
burchelli. The extant species from Fortelius and
Solounias (2000) consist of 27 “typical” species,
which they found to cluster in expected catego-
ries based on known diets, as well as one “neu-
tral” species, Ourebia ourebi.
We used Axum 6 software (licensed to TMK)
to compute Chi-square statistics and to test for
signi cance of differences observed between
individual data sets. Chi-square “corresponding
probabilities” ( p) are computed for each of the
62 combinations of single and multiple upper
tooth positions and the txM2 in order to test
whether incorporating one or more tooth posi-
tions into the mesowear model in uences the
mesowear classi cation. The resulting matrices
are sorted in descending order according to the
p value for each combination. The closer a single
model is to the txM2, the more similar is the dis-
tribution pattern of the mesowear variables. The
txM2, which is acting as the reference of these
Chi-square analyses, has the p value 1, while the
p values of all the other matrix columns (tooth
combinations) are smaller than or equal to 1.
The closer a column is to the p value of 1, the
higher is the probability that the distribution of
the mesowear variables of that particular model
of tooth combinations is not different from the
reference position, txM2.
Chi-square statistics is further used to test
for signi cance of differences observed between
upper and lower M2ʼs in the individual species
tested. The absolute frequencies of mesowear
variables (high, sharp, and round) were tested
and p values were plotted. Hierarchical clus-
ter analysis with complete linkage (furthest
neighbors) is applied following the standard
hierarchical amalgamation method of Hartigan
(1975). The algorithm of Gruvaeus and Weiner
(1972) was used to order the tree. We analyze
the three mesowear variables (% high, % sharp
and % blunt). For this analysis we use Fortelius
and Solouniasʼ (2000) original data set, the data
set of Equus burchelli from Kaiser and Fortelius
(2003) and data sets of Ovibos moschatus, Oure-
bia ourebi and Giraffa camelopardalis presented
in this study.
398 Franz-Odendaal & Kaiser ANN. ZOOL. FENNICI Vol. 40
Comparing upper and lower second
In Ovibos moschatus (Om), there are no differ-
ences in either occlusal relief scorings or in cusp
shape scorings of “blunt” (Table 1) between
upper and lower second molars. However cusp
shape scorings of “sharp” are signi cantly more
prevalent in upper M2ʼs (46%) than in lower
M2ʼs, which are dominated by round cusps
(97%) ( p = 0.029; since relief scorings are iden-
tical, this p value re ects the difference in sharp
and round scorings). In Ourebia ourebi (Oo) the
distribution pattern seen in Om for sharp and
round scorings is even more pronounced. Cusp
shape scorings of “sharp” are 58% in upper M2ʼs
versus 4% in lower M2ʼs ( p = 0.0001) while
“blunt” cusps are not present in either upper or
lower teeth. In Giraffa camelopardalis (Gc) the
distribution pattern shows an intermediate posi-
tion between the two mixed feeders described
above. Sharp cusps are more common in the
upper second molars than in the lower ones
(73% in txM2 versus 24% in tmM2, p = 0.029).
Occlusal relief is high in G. camelopardalis.
Taken together, in all three species, occlusal
relief variables are not different in upper and
lower second molars, while a signi cantly dif-
ferent mesowear pattern is observed for the cusp
shape variables “sharp” and “round”. Typically,
cusps of lower second molars tend to be less fre-
quently sharp as compared with the upper M2ʼs.
Kaiser and Fortelius (2003) observed a similar
trend in E. burchelli. It therefore appears from
these preliminary results, that all three feeding
groups (grazers, browsers and mixed feeders)
show this pattern. Corresponding probabilities
indicate that mesowear signatures in upper and
lower second molars are less likely to be differ-
ent in both the browser, G. camelopardalis ( p
= 0.029) and the grazer, Equus burchelli ( p =
0.021) and more likely to be different in the two
mixed feeders investigated (Om, p = 0.0003 and
Oo, p = 0.0001) (Fig. 1). Put another way, the
difference between uppers and lowers is more
pronounced in the two mixed feeders than in the
more specialised taxa, Giraffa and Equus.
Chi-square ranking matrices
In the sample of G. camelopardalis, the cor-
responding probability ( p) of all isolated tooth
positions and of any combination of tooth posi-
tions ranges between p = 1 for only the txM2,
the model position of Fortelius and Solounias
Table 1. Distribution of mesowear variables in upper (tx) and lower (tm) second molars of Giraffa camelopardalis
(Gc), Ourebia ourebi (Oo), Ovibos moschatus (Om) and Equus burchelli (Eb) (Equus data are from Kaiser & For-
telius 2003). n = number of specimens available, tm = mandibular tooth, tx = maxillary tooth; Mesowear variables:
l = absolute scorings low, h = absolute scorings high, s = absolute scorings sharp, r = absolute scorings round, b
= absolute scorings blunt; % l = percent low occlusal relief, % h = percent high occlusal relief, % s = percent sharp
cusps, % r = percent rounded cusps, % b = percent blunt cusps. Chi-squared statistics are shown for mesowear
variables h, s and r, comparing the upper and lower second molars.
Population n l h s r b % l % h % s % r % b
Gc(txM2) 17 0 17 8 3 0 0 100 73 27 0
Gc(tmM2) 17 0 17 4 13 0 0 100 24 76 0
Oo(txM2) 30 0 30 15 11 0 0 100 58 42 0
Oo(tmM2) 29 0 29 1 27 0 0 100 4 96 0
Om(txM2) 43 0 43 19 22 0 0 100 46 54 0
Om(tmM2) 32 0 32 1 30 0 0 100 3 97 0
Eb(txM2) 41 8 33 15 22 1 20 80 39 58 3
Eb(tmM2) 42 16 26 4 30 8 38 62 10 71 19
Gc(tx/tm) h, s, r |2 = 7.069, df = 2, p = 0.029
Oo(tx/tm) h, s, r |2 = 18.996, df = 2, p = 0.0001
Om(tx/tm) h, s, r |2 = 16.378, df = 2, p = 0.0003
Eb(tx/tm) h, s, r |2 = 7.706, df = 2, p = 0.021
ANN. ZOOL. FENNICI Vol. 40 Application of mesowear to ruminants 399
(2000), and p = 0.075 for txP2 + M1. Using all
tooth positions (txP2–M3) gives a p value of 0.6
(Fig. 2) indicating that this combination has a
low probability of giving different dietary results
for the species as compared with using txM2
alone. The txM2 + M3 model for Gc classi es
less consistently with the classical model (txM2)
than with the “extended model” (txP4–M3) for-
mulated for hypsodont equids (Kaiser & Solou-
nias 2003) (Fig. 2).
In the two mixed feeders, Ourebia ourebi and
Ovibos moschatus, the corresponding probability
( p) of all isolated tooth positions and of any
combination of tooth position ranges between
p = 1 for txM2 and p < 0.0001 for txP2 (Figs.
3 and 4). The use of all cheek tooth positions
(txP2–M3) in both species is likely to classify
the species differently to the reference position,
txM2, as p values are less than 0.05 (0.014 for
Ovibos and 0.029 for Ourebia). The “extended
model” for hypsodont equids (Kaiser & Solou-
nias 2003) using txP4–M3 is however likely to
correctly classify these species as p values are
0.327 and 0.682, respectively, for the musk ox
and oribi. In both mixed feeders the combination
of using the two most proximal tooth positions,
txM2 + M3, consistently classi es the species
closer to the reference model (using txM2 only)
than any other combination of tooth positions.
As a concluding result, we nd a larger
degree of inconsistency in the individual upper
tooth positions in the mixed feeders (Ourebi and
Ovibos) than in the browser (Giraffa). Both the
four- and two-tooth models were therefore tested
further using cluster analyses.
Cluster analysis
The cluster diagrams computed show relations of
datasets by joining them in the same clusters. The
closer the data are the smaller is the normalized
Euclidean distance (NED) at the branching point
(root-mean-squared difference). Cluster analysis
polarizes the entire set of 28 recent species and
the populations of G. camelopardalis, O. ourebi,
O. moschatus and E. burchelli into a pattern with
grazers and browsers at the extremes and with
mixed feeders in between (Figs. 5 and 6). There
are generally four main clusters, one containing
the most abrasion dominated grazers, one con-
taining the remaining grazers, one containing the
mixed feeders and one the browsers.
Comparing upper and lower second molars
(Fig. 5)
The upper tooth sample of G. camelopardalis
(Gc(txM2)) shares the same subcluster within
the browser spectrum as the comparison popu-
lation by Fortelius and Solounias (2000). The
lower tooth sample (Gc(tmM2)) however is
linked in the mixed feeder spectrum, where it
shares a subcluster of NED = 8 with the impala
(Aepyceros melampus), an abrasion dominated
mixed feeder, and the oribi (a “neutral” mixed
feeder according to Fortelius and Solounias
(2000)). The upper molars of Ourebia ourebi
(Oo(txM2)) are linked in the mixed feeder spec-
trum, but are only linked at the level of NED =
23 to the reference population of Fortelius and
Solounias (2000). This indicates a considerable
difference between the population investigated
Gc Om Oo Eb
Fig. 1. Chi-square corresponding probabilities testing
absolute mesowear scorings in upper versus lower
second molars. Mesowear variables high, sharp
and round were tested (see Table 1). Gc = Giraffa
camelopardalis, Om = Ovibos moschatus, Oo =
Ourebia ourebi, Eb = Equus burchelli.
400 Franz-Odendaal & Kaiser ANN. ZOOL. FENNICI Vol. 40
Fig. 2. Chi-square ranking matrix of the absolute frequencies of mesowear variables “low” (l), “high” (h), “sharp” (s), “round” (r) and “blunt” (b) calculated for 63 single
and multiple tooth combinations for Giraffa camelopardalis from Africa. R = ranking position; P = number of combination of single and multiple tooth positions; l, h, s, r, b,
= absolute numbers of mesowear scores; % l, % h, % s, % r, and % b = calibrated frequency (percentages) of mesowear variables; n = number of tooth specimens in a
certain model; p = Chi-square corresponding probability. The reference position for each test is the txM2 (far left column of the matrix, P = 62). The columns are sorted
by decreasing p values. Large p value (left), small p value (right). TxM2 (R = 1) always has the p value of 1, because this position is the model position of the “original”
mesowear method. hi: – (minus) = null-hypotheses of independence should be rejected at an error probability of 0.05; the distribution patterns are likely to be equal; +
(plus) null-hypotheses of independence cannot be rejected, the distribution patterns are likely to be different. Grey background: all single tooth positions txP2, txP3, txP4,
txM1, txM2 and txM3, the total of all tooth positions (txP2–M3) and the model for the “extended” mesowear method (txP4–M3) after Kaiser and Solounias (2003). The bar
charts to the bottom of each matrix plot indicate the calibrated frequencies (percentages) of mesowear variables for each column of the matrix. Gray = % h, black = % s,
% r and % b.
ANN. ZOOL. FENNICI Vol. 40 Application of mesowear to ruminants 401
Fig. 3. Chi-square ranking matrix of the absolute frequencies of mesowear variables “low” (l), “high” (h), “sharp” (s), “round” (r) and “blunt” (b) calculated for 63 single and
multiple tooth combinations for Ourebia ourebi from Africa. See Fig. 2 for details.
402 Franz-Odendaal & Kaiser ANN. ZOOL. FENNICI Vol. 40
Fig. 4. Chi-square ranking matrix of the absolute frequencies of mesowear variables “low” (l), “high” (h), “sharp” (s), “round” (r) and “blunt” (b) calculated for 63 single and
multiple tooth combinations for Ovibos moschatus from Greenland. See Fig. 2 for details.
ANN. ZOOL. FENNICI Vol. 40 Application of mesowear to ruminants 403
here and the population investigated by Fortelius
and Solounias (2000). Lower molars of O.
ourebi (Oo(tmM2)) are linked within the grazer
spectrum where they are closest to the common
waterbuck (Kobus ellipsiprymnus). Upper molars
of the Ovibos moschatus population (Om(txM2))
are linked in the mixed feeder spectrum, where
they are linked to the reference population at the
level of NED = 16. Lower molars of O. moscha-
tus (Om(tmM2)) classify together with lower
molars of O. ourebi close to the common water-
buck (Kobus ellipsiprymnus) and thus are within
the grazer spectrum. The classi cation of the E.
burchelli population has already been described
by Kaiser and Fortelius (2003) and is displayed
to show the relation to the ruminant species
investigated in this study (Fig. 5).
In summary, we show that for both the “neu-
tral” (Ourebia ourebi) and “typical” (Ovibos
moschatus) mixed feeders (as reported by
0 10 20 30 40 50 60 70 80 90 100
Alces alces
Dicerorhinus sumatrensis
Diceros bicornis
Giraffa camelopardalis
Odocoileus hemionus
Odocoileus virginianus
Okapia johnstoni
Rhinoceros sondaicus
Alcelaphus buselaphus
Bison bison
Ceratotherium simum
Connochaetes taurinus
Damaliscus lunatus
Equus burchelli
Equus grevyi
Hippotragus equinus
Hippotragus niger
Kobus ellipsiprymnus
Reduncar edunca
Aepyceros melampus
Capricornis sumatraensis
Cervus canadensis
Gazella granti
Gazella thomsoni
Ovibos moschatus
Taurotragus oryx
Tragelaphus scriptus
Ourebia ourebi
mixed feeder
Fig. 5. Hierarchical cluster diagram comparing upper second molars (txM2) and lower second molars (tmM2) of
populations of Giraffa camelopardalis, Ovibos moschatus, Ourebia ourebi and Equus burchelli, based on a set of
“typical” recent species from Fortelius and Solounias (2000). Ourebia ourebi, a neutral species after Fortelius and
Solounias (2000) is included, for comparative purposes. NED = normalized Euclidean distance (root-mean-squared
difference). Oo = Ourebia ourebi, Om = Ovibos moschatus. Gc = Giraffa camelopardalis, Eb = Equus burchelli. tx
= maxillary tooth, tm = mandibular tooth. tx(P4–M3) = four tooth model comprising upper (maxillary) teeth P4, M1,
M2 and M3. This is the “extended” model by Kaiser and Fortelius (2003), tx(M2 + M3) = two-tooth model comprising
upper (maxillary) M2 and M3. Data for E. burchelli is from Kaiser and Fortelius (2003).
404 Franz-Odendaal & Kaiser ANN. ZOOL. FENNICI Vol. 40
Fortelius & Solounias 2000), the upper M2ʼs
score within the mixed feeders while the lower
M2ʼs are classi ed in the grazer range. For the
browser, G. camelopardalis, lower M2ʼs score
with the mixed feeders, while the uppers clas-
sify as browsers. The grazer population of E.
burchelli investigated by Kaiser and Fortelius
(2003) behaves like a mixed feeder based on
upper M2ʼs and like a grazer based on lowers
(for further discussion see Kaiser & Fortelius
Testing tooth models in ruminants (Fig. 6)
In the giraffe sample, the reference position
of the original mesowear method based on the
txM2 only is closely linked to the two-tooth
model (txM2 + M3). Both share a subcluster
of NED = 4 with the reference population of
Fortelius and Solounias (2000). For G. camel-
opardalis, the “extended” four-tooth model (i.e.
txP4–M3) proposed for hypsodont equids by
Kaiser and Solounias (2003) does not classify
Fig. 6. Hierarchical cluster diagrams comparing the one-tooth model using upper second molars (txM2) only, the
two-tooth model using upper second and upper third molars (txM2 + M3) and the four-tooth model (tx(P4–M3)) for
populations of Giraffa camelopardalis, Ovibos moschatus, Ourebia ourebi and Equus burchelli, based on a set of
“typical” recent species from Fortelius and Solounias (2000). See Fig. 5 for details.
0 10 20 30 40 50 60 70 80 90 100
Oo (txM2)
Alces alces
Dicerorhinus sumatrensis
Diceros bicornis
Giraffa camelopardalis
Odocoileus hemionus
Odocoileus virginianus
Okapia johnstoni
Rhinoceros sondaicus
Alcelaphus buselaphus
Bison bison
Ceratotherium simum
Connochaetes taurinus
Damaliscus lunatus
Equus burchelli
Equus grevyi
Hippotragus equinus
Hippotragus niger
Kobus ellipsiprymnus
Redunca redunca
Aepyceros melampus
Capricornis sumatraensis
Cervus canadensis
Gazella granti
Gazella thomsoni
Ovibos moschatus
Taurotragus oryx
Tragelaphus scriptus
Ourebia ourebi
mixed feeder
ANN. ZOOL. FENNICI Vol. 40 Application of mesowear to ruminants 405
the population consistently. Using this model Gc
is classi ed in the mixed feeder spectrum close
to the bushbuck (Tragelaphus scriptus) and the
eland (Taurotragus oryx).
The sample of O. ourebi second molars
(txM2) is linked with both the two-tooth model
(txM2 + M3) and the four-tooth model proposed
by Kaiser and Solounias (2003) at the level of
NED = 11. This cluster is shared with extant
mixed feeders like the bushbuck (Tragelaphus
scriptus), eland (Taurotragus oryx) and the
wapiti (Cervus canadensis). In O. moschatus,
the second mixed feeder in this study, all three
models tested are linked at the level of NED =
13. The original one-tooth model (txM2), how-
ever, is more closely linked to the two-tooth
model (txM2 + M3) at a distance of NED = 11
than to the four-tooth model (txP4–M3). While
the two-tooth model classi es the Ovibos sample
consistently with the one tooth model (txM2) and
together with other mixed feeders (i.e. Trage-
laphus scriptus and Cervus canadensis), the
four-tooth model is linked to a different group
of mixed feeders, comprising the two gazelles
Gazelle granti and Gazelle thomsoni.
Extending the number of model tooth positions
available for mesowear evaluation in ruminants
would signi cantly improve the mesowear
method especially if fossil assemblages are to
be investigated in the future. Most fossil assem-
blages contain low numbers of tooth specimens
and restricting the mesowear method to just
one tooth position considerably limits sample
sizes. Our preliminary observations indicate
that the “extended” four-tooth model suggested
by Kaiser and Solounias (2003) for hypsodont
equids (tx(P4–M3)) can not be applied to the
ruminant species under study, because it does
not result in a consistent classi cation of the
species as compared with the one-tooth model
of the original mesowear method by Fortelius
and Solounias (2000). We observed that in the
ruminants we investigated, the differences in
mesowear signature between the individual
tooth positions is larger than in E. burchelli and
the Miocene hipparionine Hippotherium primi-
genium (Kaiser & Solounias 2003) presumably
as a result of a much more pronounced het-
erodonty in the ruminants. These observations,
supported by chi-square analysis and cluster
diagrams, indicate that a two-tooth model con-
sisting of the upper M2 and M3 is preferred as
a preliminary “extended” model for ruminants.
This two-tooth model was found to consistently
match the dietary classi cation of the one-tooth
(tx(M2)) model. This extended model, incorpo-
rating only one additional tooth position (i.e. the
upper M3), is likely to take the heterodonty of
ruminants into account and should therefore be
considered in the future application of the mes-
owear method to ruminant artiodactyls.
In comparing mesowear signatures in upper
and lower second molars we make two important
observations. Firstly, cusps of lower teeth tend
to be less frequently sharp than those of upper
molars, and secondly, mesowear signatures for
upper and lower molars are more similar in
the specialized feeders (the browsing giraffe
and the grazing zebra) as compared with those
in the mixed feeders (the oribi and musk ox).
According to Kaiser and Fortelius (2003), grav-
ity causes more pronounced additional wear in
the lower teeth of highly hyspodont dentitions
such as those of Equus burchelli. This gravita-
tional force will always ensure that unloaded
food will much more frequently contact the lower
than the upper teeth. The comparatively higher
degree of similarity between upper and lower
mesowear signatures in Equus may therefore be
the result of compensative evolutionary effects
that reduce differential wear in order not to allow
the adjoining tooth antagonists to diverge. We
therefore expect adaptive optimization of enamel
ultrastructure (Rensberger & von Koenigswald
1980, von Koenigswald & Sander 1997) to be
particularly pronounced in Equus as compared
with some ruminant species. This hypothesis
should be further tested and is considered a possi-
ble source of differential wear control in grazers.
We nd a similar effect in the giraffe, the
browser in this comparison. Since both the
giraffe and the zebra are specialized feeders,
one would expect a similar functional adapta-
tion forcing upper and lower dentitions to remain
406 Franz-Odendaal & Kaiser ANN. ZOOL. FENNICI Vol. 40
similar in terms of mesowear signatures. In the
giraffe however, the degree of anisodonty is
much less pronounced than in the zebra. The less
abrasive food eaten by the giraffe presumably
causes far less additional unloaded abrasion to
lower cheek teeth than is noted in the zebra. This
could explain why the giraffe shows the small-
est degree of differential mesowear, a hypoth-
esis that is supported by the comparatively large
p value (Table 1 and Fig. 1).
The mixed feeders, on the other hand, are
required to cope with a wide variety of food
items, including abrasive grass, grit loaded
matter and tough foliage. We would therefore
expect their lower dentitions to be subjected to
a considerable amount of additional unloaded
abrasion, which would favor differential wear
in upper and lower cheek teeth. It appears as if
lower molars “loose” their sharpness, while the
uppers are optimized in terms of sharp cutting
edges. This optimization is possible because of
the lack of gravitational forces (and additional
loading) acting on upper teeth. The degree of
functional anisodonty is therefore likely to be
most pronounced in the mixed feeders. This
interpretation is further supported by a study
by T. M. Kaiser (unpubl.), which indicates that
enamel ridge alignments in ruminant cheek teeth
are highly correlated with diet. T. M. Kaiser
(unpubl.) nds that compared with browsers
and grazers the mixed feeding ruminants have a
particularly high proportion of cutting blades ori-
ented at 75°–85° towards chewing direction. This
observation is interpreted as re ecting an adapta-
tion to cope with the extreme heterogeneity of
plant materials eaten. Depending on body size,
grazers and browsers however are found more
frequently to have cutting blades oriented paral-
lel or in low angle alignments to chewing direc-
tion. In terms of keeping upper teeth sharp, this
is further evidence that maintaining sharp cutting
blades in upper molars is the most critical adap-
tation of mixed-feeders, which lack many of the
speci c adaptations grazers have, but at the same
time have to cope with grass as an extremely
abrasive and ber rich component in their diet.
In summary, although the signi cance of the
results presented here with regards to the mes-
owear method are clear, understanding the proc-
esses behind differential wear control in upper
and lower teeth is far more complex. We have
attempted to address these based on our pre-
liminary data. Our results seem to suggest that in
mixed feeders, sharpness is maximized in upper
teeth, whereas in specialized feeders this is not
the case to the same extent. Kaiser and Fortelius
(2003) observe a similar pattern in mixed feed-
ing horses. No doubt further investigations will
shed light on these hypotheses and provide a
better understanding of the evolutionary proc-
esses involved in differential wear control.
We thank the Deutsche Akademische Austauschdienst
(DAAD) for a scholarship awarded to TFO. The Zoological
Museum, University of Copenhagen (Denmark) is acknowl-
edged for providing access to the collection of Ovibos
moschatus, in particular we wish to thank Ellen Schulz
(Greifswald University, Germany) for moulding the Ovibos
specimens at the Zoological Museum, Copenhagen and Prof.
Kim Aaris-Sørensen for the support he gave to ES during
her stay in Copenhagen. The Copenhagen Biodiversity
Centre (COBICE) is gratefully acknowledged for providing
a grant from the European Commissionʼs “Improving Human
Potential” programme, which made it possible for ES to
mould the Ovibos specimens at the Zoological Museum of
the University of Copenhagen. We further thank the Museum
für Naturkunde Berlin for providing access to the giraffe
and oribi specimens. Mrs Karin Meyer and Mrs Heidrun
Dähn (Greifswald University, Germany) are acknowledged
for making casts of the fossil teeth and for maintaining and
curating our growing mesowear database.
Bernor, R. L., Kaiser, T. M., Kordos, L. & Scott, R. S. 1999:
Stratigraphic Context, Systematic Position and Paleoecol-
ogy of Hippotherium sumegense Kretzoi, 1984 from MN
10 (Late Vallesian of the Pannonian Basin). — Mitteilung
der Bayerischen Staatssammlung für Paläontologie und
historische Geologie, München 39: 1–35.
Fortelius, M. 1985: Ungulate check teeth: developmental,
functional, and evolutionary interrelations. — Acta
Zoologica Fennica 180: 1–76.
Fortelius, M. & Solounias, N. 2000: Functional character-
ization of ungulate molars using the Abrasion-Attrition
wear gradient: A new method for reconstructing paleodi-
ets. — Am. Mus. Novit. 3301: 1–36.
Franz-Odendaal, T. A. 2002: Analysis of dental pathologies
in the Pliocene herbivores of Langebaanweg and their
palaeoenvironmental implications. — Ph.D. thesis,
University of Cape Town, South Africa.
Gruvaeus, G. & Weiner, H. 1972: Two additions to hierarchi-
cal cluster analysis. — The British Journal of Mathemat-
ANN. ZOOL. FENNICI Vol. 40 Application of mesowear to ruminants 407
ical and Statistical Psychology 25: 200–206.
Hartigan, J. A. 1975: Clustering algorithms. — John Wiley
& Sons, Inc., New York.
Kaiser, T. M. 2001: Ecological signi cance of wear gradients
in the enamel ridge morphology of the cheek dentition of
the Miocene horse Cormohipparion occidentale (Equi-
dae, Perissodactyla). — Mamm. Biol. 66: 18.
Kaiser, T. M. 2003: The dietary regimes of two contempora-
neous populations of Hippotherium primigenium (Peris-
sodactyla, Equidae) from the Vallesian (upper Miocene)
of Southern Germany. — Palaeogeogr. Palaeocol. [In
Kaiser, T. M. & Bernor, R. L. 2001a: Dietary adaptations
in the Neogene genus Hippotherium (Perissodactyla,
Equidae) from the late Miocene of Central Europe. — J.
Morphol. 248: 247.
Kaiser, T. M. & Bernor, R. L. 2001b: Extending the mesowear
method to small assemblages of fossil herbivore mam-
mals. 3,5 Milliarden Jahre Biodiversität. — Terra Nostra,
Schriften der Alfred Wegener Stiftung 01: 181.
Kaiser, T. & Croitor, R. 2002: Functional morphology and
diet preferences of fossil deer and paleolandscape recon-
struction on early Pleistocene of Ceyssaguet. — Verhan-
dlungen der Gesellschaft für Ökologie: 465.
Kaiser, T. M. & Fortelius, M. 2003: Differential mesowear in
occluding upper and lower molars — opening mesowear
analysis for lower molars and premolars in hypsodont
equids. — J. Morphol. [In press].
Kaiser, T. M. & Solounias, N. 2003: Extending the tooth
mesowear method to extinct and extant equids. — Geo-
diversitas 25: 321–345.
Kaiser, T. M., Bernor, R., Fortelius, M. & Scott, R. 2000a:
Ecological diversity in the Neogene genus Hippotherium
(Perissodactyla, Equidae) from the late Miocene of Cen-
tral Europe. — J. Vertebr. Palaeontol. 20: 51A.
Kaiser, T. M., Bernor, R. L., Franzen, J. L., Scott, R. &
Solounias, N. 2003: New Interpretations of the System-
atics and Palaeoecology of the Dorn-Dürkheim 1 Hippa-
rions (Late Miocene, Turolian Age [MN11]), Rheinhes-
sen, Germany. — Senckenbergiana lethaea. [In press].
Kaiser, T. M., Solounias, N., Fortelius, M., Bernor, R. L. &
Schrenk, F. 2000b: Tooth mesowear analysis on Hip-
potherium primigenium from the Vallesian Dinotherien-
sande (Germany) — A blind test study. — Carolinea
— Beiträge zur naturkundlichen Forschung in
Südwestdeutschland 58: 103–114.
Rensberger, J. M. & von Koenigswald, W. 1980: Functional
and phylogenetic interpretation of enamel microstructure
in rhinoceroses. — Paleobiol. 6: 477–495.
von Koenigswald, W. & Sander, P. M. 1997: Schmelzmuster
differentiation in leading and trailing edges, a spe-
ci c biomechanical adaptation in rodents. — In: von
Koenigswald, W. & Sander, M. (eds.), Tooth enamel
microstructure: 259–266. A. A. Balkema, Rotterdam.
Dental specimens included in this investigation. ZMB = Museum für Naturkunde Berlin (Germany),
ZMUK = Zoological Museum, University of Copenhagen (Denmark), tx = maxillary tooth, tm =
mandibular tooth, s = left, d = right.
Ourebia ourebi: ZMB-17109 txP2 s, ZMB-20942 txP2 d, ZMB-36759 txP2 s, ZMB-36760 txP2 d,
ZMB-40299 txP2 s, ZMB-40301 txP2 d, ZMB-41815 txP2 s, ZMB-4758 txP2 d, ZMB-56878 txP2
d, ZMB-56882 txP2 d, ZMB-56893 txP2 s, ZMB-56894 txP2 d, ZMB-56924 txP2 s, ZMB-56927
txP2 d, ZMB-56931 txP2 s, ZMB-56933 txP2 s, ZMB-56936 txP2 s, ZMB-56943 txP2 d, ZMB-
57319 txP2 s, ZMB-57727 txP2 d, ZMB-17109 txP3 s, ZMB-20942 txP3 d, ZMB-36759 txP3 s,
ZMB-36760 txP3 d, ZMB-40299 txP3 s, ZMB-40301 txP3 d, ZMB-41815 txP3 s, ZMB-4758 txP3
d, ZMB-56878 txP3 d, ZMB-56882 txP3 d, ZMB-56893 txP3 s, ZMB-56894 txP3 d, ZMB-56924
txP3 s, ZMB-56927 txP3 d, ZMB-56931 txP3 s, ZMB-56933 txP3 s, ZMB-56936 txP3 s, ZMB-
56938 txP3 s, ZMB-56943 txP3 d, ZMB-56946 txP3 s, ZMB-57319 txP3 s, ZMB-57727 txP3 d,
ZMB-17109 txP4 s, ZMB-20942 txP4 d, ZMB-36759 txP4 s, ZMB-36760 txP4 d, ZMB-40299 txP4
s, ZMB-40301 txP4 d, ZMB-41815 txP4 s, ZMB-4758 txP4 d, ZMB-56878 txP4 d, ZMB-56882 txP4
d, ZMB-56885 txP4 d, ZMB-56893 txP4 s, ZMB-56894 txP4 d, ZMB-56924 txP4 s, ZMB-56927
txP4 d, ZMB-56931 txP4 s, ZMB-56933 txP4 s, ZMB-56936 txP4 s, ZMB-56938 txP4 s, ZMB-
56943 txP4 d, ZMB-56946 txP4 s, ZMB-57319 txP4 s, ZMB-57727 txP4 d, ZMB-17109 txM1 s,
ZMB-20942 txM1 d, ZMB-36759 txM1 s, ZMB-36760 txM1 d, ZMB-40299 txM1 s, ZMB-40301
txM1 d, ZMB-41815 txM1 s, ZMB-4758 txM1 d, ZMB-56875 txM1 s, ZMB-56878 txM1 d, ZMB-
56882 txM1 d, ZMB-56885 txM1 d, ZMB-56893 txM1 s, ZMB-56894 txM1 d, ZMB-56897 txM1
d, ZMB-56927 txM1 d, ZMB-56928 txM1 s, ZMB-56929 txM1 d, ZMB-56930 txM1 s, ZMB-56931
txM1 s, ZMB-56933 txM1 s, ZMB-56934 txM1 s, ZMB-56936 txM1 s, ZMB-56938 txM1 s, ZMB-
408 Franz-Odendaal & Kaiser ANN. ZOOL. FENNICI Vol. 40
56943 txM1 d, ZMB-56946 txM1 s, ZMB-57127 txM1 s, ZMB-57319 txM1 s, ZMB-57727 txM1 d,
ZMB-17109 txM2 s, ZMB-20942 txM2 d, ZMB-36759 txM2 s, ZMB-36760 txM2 d, ZMB-40299
txM2 s, ZMB-40301 txM2 d, ZMB-41815 txM2 s, ZMB-4758 txM2 d, ZMB-56875 txM2 s, ZMB-
56878 txM2 d, ZMB-56882 txM2 d, ZMB-56885 txM2 d, ZMB-56893 txM2 s, ZMB-56894 txM2
d, ZMB-56897 txM2 d, ZMB-56924 txM2 s, ZMB-56927 txM2 d, ZMB-56928 txM2 s, ZMB-56929
txM2 d, ZMB-56930 txM2 s, ZMB-56931 txM2 s, ZMB-56933 txM2 s, ZMB-56934 txM2 s, ZMB-
56936 txM2 s, ZMB-56938 txM2 s, ZMB-56943 txM2 d, ZMB-56946 txM2 s, ZMB-57127 txM2 s,
ZMB-57319 txM2 s, ZMB-57727 txM2 d, ZMB-17109 txM3 s, ZMB-20942 txM3 d, ZMB-36759
txM3 s, ZMB-36760 txM3 d, ZMB-40299 txM3 s, ZMB-40301 txM3 d, ZMB-41815 txM3 s, ZMB-
4758 txM3 d, ZMB-56875 txM3 s, ZMB-56878 txM3 d, ZMB-56882 txM3 d, ZMB-56885 txM3 d,
ZMB-56893 txM3 s, ZMB-56894 txM3 d, ZMB-56924 txM3 s, ZMB-56927 txM3 d, ZMB-56928
txM3 s, ZMB-56929 txM3 d, ZMB-56930 txM3 s, ZMB-56931 txM3 s, ZMB-56933 txM3 s, ZMB-
56936 txM3 s, ZMB-56938 txM3 s, ZMB-56943 txM3 d, ZMB-56946 txM3 s, ZMB-57127 txM3 s,
ZMB-57319 txM3 s, ZMB-57727 txM3 d, ZMB-17109 tmM2 s, ZMB-20942 tmM2 d, ZMB-36759
tmM2 s, ZMB-36760 tmM2 d, ZMB-40299 tmM2 s, ZMB-40301 tmM2 d, ZMB-41815 tmM2 s,
ZMB-4758 tmM2 d, ZMB-56875 tmM2 s, ZMB-56878 tmM2 d, ZMB-56882 tmM2 d, ZMB-56885
tmM2 d, ZMB-56893 tmM2 s, ZMB-56893 tmM2 d, ZMB-56894 tmM2 d, ZMB-56924 tmM2 s,
ZMB-56927 tmM2 d, ZMB-56928 tmM2 s, ZMB-56929 tmM2 d, ZMB-56930 tmM2 s, ZMB-56931
tmM2 s, ZMB-56933 tmM2 s, ZMB-56936 tmM2 s, ZMB-56938 tmM2 s, ZMB-56943 tmM2 d,
ZMB-56946 tmM2 s, ZMB-57127 tmM2 s, ZMB-57319 tmM2 s, ZMB-57727 tmM2 d.
Ovibos moschatus: ZMUK-R-1725 txP2 d, ZMUK-R-1726 txP2 s, ZMUK-R-1732 txP2 d, ZMUK-
R-1734 txP2 d, ZMUK-R-1736 txP2 s, ZMUK-R-1737 txP2 s, ZMUK-R-2412 txP2 s, ZMUK-R-
2414 txP2 s, ZMUK-R-2416 txP2 s, ZMUK-R-2419 txP2 s, ZMUK-R-2423 txP2 d, ZMUK-R-2430
txP2 s, ZMUK-R-2431 txP2 s, ZMUK-R-3012 txP2 s, ZMUK-R-3015 txP2 d, ZMUK-R-3471 txP2
s, ZMUK-R-3472 txP2 s, ZMUK-R-3472 txP2 d, ZMUK-R-3479 txP2 s, ZMUK-R-3607 txP2
d, ZMUK-R-3608 txP2 d, ZMUK-R-3610 txP2 d, ZMUK-R-3614 txP2 s, ZMUK-R-3615 txP2
s, ZMUK-R-3617 txP2 d, ZMUK-R-3619 txP2 d, ZMUK-R-3707 txP2 s, ZMUK-R-3708 txP2
s, ZMUK-R-3762 txP2 s, ZMUK-R-3769 txP2 s, ZMUK-R-3772 txP2 s, ZMUK-R-3777 txP2 s,
ZMUK-R-3778 txP2 s, ZMUK-R-779 txP2 d, ZMUK-R-997 txP2 s, ZMUK-R-998 txP2 d, ZMUK-
R-1725 txP3 d, ZMUK-R-1726 txP3 s, ZMUK-R-1727 txP3 s, ZMUK-R-1727 txP3 s, ZMUK-R-
1732 txP3 d, ZMUK-R-1734 txP3 d, ZMUK-R-1736 txP3 s, ZMUK-R-1737 txP3 s, ZMUK-R-2412
txP3 s, ZMUK-R-2414 txP3 s, ZMUK-R-2416 txP3 s, ZMUK-R-2419 txP3 s, ZMUK-R-2423 txP3
d, ZMUK-R-2430 txP3 s, ZMUK-R-2431 txP3 s, ZMUK-R-3012 txP3 s, ZMUK-R-3015 txP3
d, ZMUK-R-3471 txP3 s, ZMUK-R-3472 txP3 s, ZMUK-R-3472 txP3 d, ZMUK-R-3479 txP3
s, ZMUK-R-3607 txP3 d, ZMUK-R-3608 txP3 d, ZMUK-R-3610 txP3 d, ZMUK-R-3614 txP3
s, ZMUK-R-3615 txP3 s, ZMUK-R-3617 txP3 d, ZMUK-R-3619 txP3 d, ZMUK-R-3707 txP3
s, ZMUK-R-3708 txP3 s, ZMUK-R-3762 txP3 s, ZMUK-R-3769 txP3 s, ZMUK-R-3772 txP3 s,
ZMUK-R-3777 txP3 s, ZMUK-R-3778 txP3 s, ZMUK-R-779 txP3 d, ZMUK-R-997 txP3 s, ZMUK-
R-998 txP3 d, ZMUK-R-1725 txP4 d, ZMUK-R-1726 txP4 s, ZMUK-R-1727 txP4 s, ZMUK-R-
1727 txP4 s, ZMUK-R-1732 txP4 d, ZMUK-R-1734 txP4 d, ZMUK-R-1736 txP4 s, ZMUK-R-1737
txP4 s, ZMUK-R-2412 txP4 s, ZMUK-R-2414 txP4 s, ZMUK-R-2416 txP4 s, ZMUK-R-2419 txP4
s, ZMUK-R-2423 txP4 d, ZMUK-R-2430 txP4 s, ZMUK-R-2431 txP4 s, ZMUK-R-3012 txP4 s,
ZMUK-R-3015 txP4 d, ZMUK-R-3471 txP4 s, ZMUK-R-3472 txP4 s, ZMUK-R-3472 txP4 d,
ZMUK-R-3479 txP4 s, ZMUK-R-3607 txP4 d, ZMUK-R-3608 txP4 d, ZMUK-R-3610 txP4 d,
ZMUK-R-3614 txP4 s, ZMUK-R-3615 txP4 s, ZMUK-R-3617 txP4 d, ZMUK-R-3619 txP4 d,
ZMUK-R-3707 txP4 s, ZMUK-R-3708 txP4 s, ZMUK-R-3762 txP4 s, ZMUK-R-3769 txP4 s,
ZMUK-R-3772 txP4 s, ZMUK-R-3777 txP4 s, ZMUK-R-3778 txP4 s, ZMUK-R-777 txP4 d, ZMUK-
R-779 txP4 d, ZMUK-R-997 txP4 s, ZMUK-R-998 txP4 d, ZMUK-R-1725 txM1 d, ZMUK-R-1726
txM1 s, ZMUK-R-1727 txM1 s, ZMUK-R-1727 txM1 s, ZMUK-R-1732 txM1 d, ZMUK-R-1734
ANN. ZOOL. FENNICI Vol. 40 Application of mesowear to ruminants 409
txM1 d, ZMUK-R-1736 txM1 s, ZMUK-R-1737 txM1 s, ZMUK-R-2412 txM1 s, ZMUK-R-2414
txM1 s, ZMUK-R-2416 txM1 s, ZMUK-R-2419 txM1 s, ZMUK-R-2423 txM1 d, ZMUK-R-2430
txM1 s, ZMUK-R-2431 txM1 s, ZMUK-R-3012 txM1 s, ZMUK-R-3015 txM1 d, ZMUK-R-3471
txM1 s, ZMUK-R-3472 txM1 s, ZMUK-R-3472 txM1 d, ZMUK-R-3479 txM1 s, ZMUK-R-3607
txM1 d, ZMUK-R-3608 txM1 d, ZMUK-R-3610 txM1 d, ZMUK-R-3614 txM1 s, ZMUK-R-3615
txM1 s, ZMUK-R-3617 txM1 d, ZMUK-R-3619 txM1 d, ZMUK-R-3707 txM1 s, ZMUK-R-3708
txM1 s, ZMUK-R-3762 txM1 s, ZMUK-R-3769 txM1 s, ZMUK-R-3772 txM1 s, ZMUK-R-3777
txM1 s, ZMUK-R-3778 txM1 s, ZMUK-R-777 txM1 d, ZMUK-R-779 txM1 d, ZMUK-R-997
txM1 s, ZMUK-R-998 txM1 d, ZMUK-R-1725 txM2 d, ZMUK-R-1726 txM2 s, ZMUK-R-1727
txM2 s, ZMUK-R-1727 txM2 s, ZMUK-R-1732 txM2 d, ZMUK-R-1734 txM2 d, ZMUK-R-1736
txM2 s, ZMUK-R-1737 txM2 s, ZMUK-R-2412 txM2 s, ZMUK-R-2414 txM2 s, ZMUK-R-2416
txM2 s, ZMUK-R-2419 txM2 s, ZMUK-R-2423 txM2 d, ZMUK-R-2430 txM2 s, ZMUK-R-2431
txM2 s, ZMUK-R-3012 txM2 s, ZMUK-R-3015 txM2 d, ZMUK-R-3471 txM2 s, ZMUK-R-3472
txM2 s, ZMUK-R-3472 txM2 d, ZMUK-R-3476 txM2 d, ZMUK-R-3479 txM2 s, ZMUK-R-3607
txM2 d, ZMUK-R-3608 txM2 d, ZMUK-R-3610 txM2 d, ZMUK-R-3614 txM2 s, ZMUK-R-3615
txM2 s, ZMUK-R-3617 txM2 d, ZMUK-R-3619 txM2 d, ZMUK-R-3707 txM2 s, ZMUK-R-3708
txM2 s, ZMUK-R-3762 txM2 s, ZMUK-R-3767 txM2 d, ZMUK-R-3769 txM2 s, ZMUK-R-3772
txM2 s, ZMUK-R-3777 txM2 s, ZMUK-R-3778 txM2 s, ZMUK-R-777 txM2 d, ZMUK-R-779
txM2 d, ZMUK-R-992 txM2 d, ZMUK-R-996 txM2 s, ZMUK-R-997 txM2 s, ZMUK-R-998 txM2
d, ZMUK-R-1725 txM3 d, ZMUK-R-1726 txM3 s, ZMUK-R-1727 txM3 s, ZMUK-R-1727 txM3
s, ZMUK-R-1732 txM3 d, ZMUK-R-1734 txM3 d, ZMUK-R-1736 txM3 s, ZMUK-R-1737 txM3
s, ZMUK-R-2412 txM3 s, ZMUK-R-2414 txM3 s, ZMUK-R-2416 txM3 s, ZMUK-R-2419 txM3
s, ZMUK-R-2423 txM3 d, ZMUK-R-2430 txM3 s, ZMUK-R-2431 txM3 s, ZMUK-R-3012 txM3
s, ZMUK-R-3015 txM3 d, ZMUK-R-3471 txM3 s, ZMUK-R-3472 txM3 s, ZMUK-R-3472 txM3
d, ZMUK-R-3476 txM3 d, ZMUK-R-3479 txM3 s, ZMUK-R-3607 txM3 d, ZMUK-R-3608 txM3
d, ZMUK-R-3610 txM3 d, ZMUK-R-3614 txM3 s, ZMUK-R-3615 txM3 s, ZMUK-R-3617 txM3
d, ZMUK-R-3619 txM3 d, ZMUK-R-3707 txM3 s, ZMUK-R-3708 txM3 s, ZMUK-R-3762 txM3
s, ZMUK-R-3767 txM3 d, ZMUK-R-3769 txM3 s, ZMUK-R-3772 txM3 s, ZMUK-R-3777 txM3
s, ZMUK-R-3778 txM3 s, ZMUK-R-777 txM3 d, ZMUK-R-779 txM3 d, ZMUK-R-992 txM3 d,
ZMUK-R-996 txM3 s, ZMUK-R-997 txM3 s, ZMUK-R-998 txM3 d, ZMUK-R-1725 tmM2 s,
ZMUK-R-1726 tmM2 s, ZMUK-R-1727 tmM2 s, ZMUK-R-1732 tmM2 s, ZMUK-R-1734 tmM2
d, ZMUK-R-1736 tmM2 s, ZMUK-R-2412 tmM2 s, ZMUK-R-2414 tmM2 s, ZMUK-R-2416 tmM2
s, ZMUK-R-2419 tmM2 s, ZMUK-R-2423 tmM2 s, ZMUK-R-2423 tmM2 d, ZMUK-R-2430 tmM2
s, ZMUK-R-2431 tmM2 s, ZMUK-R-3012 tmM2 s, ZMUK-R-3471 tmM2 s, ZMUK-R-3471 tmM2
d, ZMUK-R-3472 tmM2 s, ZMUK-R-3479 tmM2 s, ZMUK-R-3707 tmM2 s, ZMUK-R-3708 tmM2
s, ZMUK-R-3762 tmM2 s, ZMUK-R-3769 tmM2 s, ZMUK-R-3772 tmM2 s, ZMUK-R-3777 tmM2
s, ZMUK-R-3778 tmM2 s, ZMUK-R-775 tmM2 s, ZMUK-R-777 tmM2 d, ZMUK-R-779 tmM2 d,
ZMUK-R-993 tmM2 s, ZMUK-R-997 tmM2 s, ZMUK-R-998 tmM2 d.
Giraffa camelopardalis: ZMB-15552 txP2 s, ZMB-17391 txP2 d, ZMB-20997 txP2 s, ZMB-31781
txP2 d, ZMB-32236 txP2 d, ZMB-32372 txP2 s, ZMB-32373 txP2 d, ZMB-42103 txP2 d, ZMB-
48440 txP2 d, ZMB-84923 txP2 s, ZMB-84954 txP2 d, ZMB-84955 txP2 s, ZMB-84962 txP2 s,
ZMB-84967 txP2 d, ZMB-15552 txP3 s, ZMB-17391 txP3 d, ZMB-20997 txP3 s, ZMB-31781 txP3
d, ZMB-32236 txP3 d, ZMB-32372 txP3 s, ZMB-32373 txP3 d, ZMB-42103 txP3 d, ZMB-48440
txP3 d, ZMB-84923 txP3 s, ZMB-84954 txP3 d, ZMB-84955 txP3 s, ZMB-84962 txP3 s, ZMB-
84967 txP3 d, ZMB-15552 txP4 s, ZMB-17391 txP4 d, ZMB-20997 txP4 s, ZMB-31781 txP4 d,
ZMB-32236 txP4 d, ZMB-32372 txP4 s, ZMB-32373 txP4 d, ZMB-42103 txP4 d, ZMB-48440 txP4
d, ZMB-84923 txP4 s, ZMB-84954 txP4 d, ZMB-84955 txP4 s, ZMB-84962 txP4 s, ZMB-84967
txP4 d, ZMB-15552 txM1 s, ZMB-17391 txM1 d, ZMB-20997 txM1 s, ZMB-31393 txM1 d, ZMB-
31781 txM1 d, ZMB-32236 txM1 d, ZMB-32372 txM1 s, ZMB-32373 txM1 d, ZMB-42103 txM1 d,
410 Franz-Odendaal & Kaiser ANN. ZOOL. FENNICI Vol. 40
ZMB-48440 txM1 d, ZMB-84923 txM1 s, ZMB-84942 txM1 d, ZMB-84954 txM1 d, ZMB-84955
txM1 s, ZMB-84961 txM1 d, ZMB-84962 txM1 s, ZMB-84967 txM1 d, ZMB-15552 txM2 s, ZMB-
17391 txM2 d, ZMB-20997 txM2 s, ZMB-31393 txM2 d, ZMB-31781 txM2 d, ZMB-32236 txM2 d,
ZMB-32372 txM2 s, ZMB-32373 txM2 d, ZMB-42103 txM2 d, ZMB-48440 txM2 d, ZMB-84923
txM2 s, ZMB-84942 txM2 d, ZMB-84954 txM2 d, ZMB-84955 txM2 s, ZMB-84961 txM2 d, ZMB-
84962 txM2 s, ZMB-84967 txM2 d, ZMB-15552 txM3 s, ZMB-17391 txM3 d, ZMB-20997 txM3 s,
ZMB-31781 txM3 d, ZMB-32236 txM3 d, ZMB-32372 txM3 s, ZMB-32373 txM3 d, ZMB-42103
txM3 d, ZMB-48440 txM3 d, ZMB-84923 txM3 s, ZMB-84942 txM3 d, ZMB-84954 txM3 d, ZMB-
84955 txM3 s, ZMB-84962 txM3 s, ZMB-84967 txM3 d, ZMB-15552 tmM2 s, ZMB-17391 tmM2 d,
ZMB-20997 tmM2 s, ZMB-31393 tmM2 d, ZMB-31781 tmM2 d, ZMB-32236 tmM2 d, ZMB-32372
tmM2 s, ZMB-32373 tmM2 d, ZMB-42103 tmM2 d, ZMB-48440 tmM2 d, ZMB-84923 tmM2 s.
... Not all time bins or taxa yielded large sample sizes, however, with the most notable being E. occidentalis in pits 91 and 13, and C. hesternus in all pits (low sample size is defined as n < 10, following Fortelius and Solounias, 2000; Table 1). Recent mesowear studies have attempted to resolve issues of low sample sizes by including adult lower cheek teeth, with mixed success (Franz-Odendaal and Kaiser, 2003;Rivals et al., 2009;Fraser et al., 2014;Dumouchel and Bobe, 2019). Adult lower cheek teeth tend to have blunter cusps than adult upper cheek teeth in the same specimens, leading to an interpretation of increased grazing from lower cheek teeth compared to upper cheek teeth (Franz-Odendaal and Kaiser, 2003). ...
... Recent mesowear studies have attempted to resolve issues of low sample sizes by including adult lower cheek teeth, with mixed success (Franz-Odendaal and Kaiser, 2003;Rivals et al., 2009;Fraser et al., 2014;Dumouchel and Bobe, 2019). Adult lower cheek teeth tend to have blunter cusps than adult upper cheek teeth in the same specimens, leading to an interpretation of increased grazing from lower cheek teeth compared to upper cheek teeth (Franz-Odendaal and Kaiser, 2003). However, mesowear for adult upper and adult lower cheek teeth may give slightly different signals for the same individual (Franz-Odendaal and Kaiser, 2003;Kaiser and Fortelius, 2003). ...
... Adult lower cheek teeth tend to have blunter cusps than adult upper cheek teeth in the same specimens, leading to an interpretation of increased grazing from lower cheek teeth compared to upper cheek teeth (Franz-Odendaal and Kaiser, 2003). However, mesowear for adult upper and adult lower cheek teeth may give slightly different signals for the same individual (Franz-Odendaal and Kaiser, 2003;Kaiser and Fortelius, 2003). At RLB, low sample sizes were in large part due to a high percentage of juvenile specimens lacking worn adult teeth (Friscia et al., 2008;Jefferson and Goldin, 1989). ...
The Rancho La Brea locality is world famous for asphaltic deposits that trapped and preserved late Pleistocene megafauna over the last 50,000 years. This wealth of paleontological data allows for detailed investigation into paleoecological changes through the last glacial maximum into the Holocene. Here, we used dental mesowear analyses to infer dietary behavior in Bison antiquus, Equus occidentalis, and Camelops hesternus from five deposits (“pits”) spanning the latest Pleistocene: Pits 77, 91, 13, 3, and 61/67. Mesowear was compared among pits for each taxon and discriminant function and posterior probability analyses were conducted using a modern dataset to predict dietary categories at Rancho La Brea. Published mesowear scores from late Pleistocene Bison, Equus and Camelops from other localities were included in the discriminant function and posterior probability analyses to assess dietary variability among regions. Mesowear for each taxon did not differ among pits. Posterior probabilities and discriminant function analyses recovered E. occidentalis as a strict grazer with B. antiquus and C. hesternus recovered as mixed feeders. The stability of mesowear scores through the latest Pleistocene suggests average diets of these herbivores did not significantly change at Rancho La Brea. This is in contrast to documented changes in climate and flora proxies of southern California. However, it is unclear whether these proxies are representative of climate and floral changes at Rancho La Brea. Mesowear scores from late Pleistocene populations of Equus, Bison, and Camelops indicate little variability in diet in Equus, modest variability in Bison, and high variability in Camelops. These analyses suggest large ungulates may have been more opportunistic in their feeding strategies and highlights the need for using multiple proxies to clarify dietary behavior of herbivores
... ; EM: enhanced mesowear method established by . Table 1), expanding it to more teeth (Franz-Odendaal and Kaiser, 2003;Kaiser and Fortelius, 2003), and adapting the method to specific taxa (Fraser and Theodor, 2010;Purnell and Jones, 2012;Taylor et al., 2013;Butler et al., 2014;Saarinen et al., 2015;Saarinen and Karme, 2017). Some, deeming OR a redundant measure, simplified mesowear by only using categories of CS (Mihlbachler and Solounias, 2006;Widga, 2006), while others simplified the technique by combining OR and CS into a single score (Rivals and Semprebon, 2006;Croft and Weinstein, 2008;) -these simplified versions of the original mesowear technique were deemed "mesowear II" by . ...
... In the original mesowear scoring system, only the upper second molar was used, and CS was scored as sharp, round or blunt and OR as high or low Kaiser, 2000). Since then, the system was expanded to include upper and lower molars (Franz-Odendaal and Kaiser, 2003;Kaiser and Fortelius, 2003;Kaiser and Solounias, 2003), as well as a higher number of differentiated scoring states (e.g. . However, a more simplified version of the scoring system using a set of gauges has also been introduced . ...
... Originally developed by , the mesowear method consisted of two tooth characters: the cusp shape and the cusp relief of ungulate upper molars, scored for one apex per individual. This was further expanded to lower molars for equids (Kaiser & Fortelius 2003) and artiodactyls (Franz-Odendaal & Kaiser 2003). The original technique and its expanded version were termed "mesowear I" by . ...
Full-text available
The effects created by diet on an animal's teeth can be very informative to palaeontologists and neontologists alike. Herbivores have been the main focus of tooth wear research, as the abrasive components of their diet can create changes on their teeth at macroscopic or microscopic scales, which can in turn be used to reconstruct palaeodiets and palaeoenvironments of extinct species. Grazing herbivores, for example, sustain high amounts of tooth abrasion caused by grit, or phytoliths internal to plants; while browsers have a more attrition-dominated tooth signature caused by tooth on tooth contact when chewing a softer diet. The main proxies used to infer diet from dental wear are microwear and mesowear. However, a full understanding of the timespan, resolution and correlation of these proxies can only be achieved by controlled feeding experiments. This body or work focuses on two long-term feeding experiments conducted on goats and sheep. By adding various sizes and concentrations of abrasives to experimental diets, we were able to test the effect of each diet with several dental wear proxies, allowing us to observe the development of wear over time by means of CT scans taken over the course of the experiments. Through this body of work, we determined mesowear to be much more of a general lifetime signal than previously thought, excluding the use of this type of proxy for seasonal dietary reconstruction, at least in small ruminants. Furthermore, comparing microscopic wear to the same dataset leads us to believe that these two proxies do not measure the same process, but rather different processes on different scales, the link between which remains to be explored. Following ingested abrasives through the gastrointestinal system on CT scans allowed for the description of the ruminant washing mechanism, used to mitigate tooth wear by washing off external abrasives in the rumen and providing a less abrasive food bolus for regurgitation. Finally, we described the phenomena of root growth compensating for crown wear, through volumetric absolute wear measurements. Systems such as the washing mechanism or high-crowned teeth in herbivores emphasize tooth abrasion as one of the main pressures for herbivore evolution and diversification.
... Although dental mesowear was originally applied to upper second molars , subsequent work has shown that it can also be applied to the rest of the upper (Franz-Odendaal & Kaiser, 2003;Kaiser & Solounias, 2003) and lower molars (Blondel et al., 2010). In archaeological work, specific distinction between sheep and goats is much more difficult on the upper molars than on the lower ones (e.g. ...
... The number of samples was based on the recommendations of Fortelius and Solounias (2000) indicating a minimum of 10 individuals per group. The analysis of the dental mesowear score consists into a macroscopic and qualitative analysis of dental cusps located on the lingual side of lower molars Franz-Odendaal & Kaiser, 2003;Kaiser & Solounias, 2003). Cusps are classified into seven categories, where '0' is characterised by high relief and pointed cusps, corresponding to the least abrasive wear, and '6' is characterised by low relief and flattened (blunt) cusps, corresponding to the most abrasive wear (Mihlbachler et al., 2011;Rivals et al., 2007) (Fig. 2A). ...
Full-text available
Dental mesowear is a widely used tool in archaeology and palaeontology for the reconstruction of the overall diet of mammals. This method is based on the characterisation of the height and relief of dental cusps, as they vary according to diet. The use of this method on domestic ungulates presents limitations because (1) currently, very few reference frameworks are available, and (2) the qualitative categorisation of cusps limits our observations and interpretations. In this work, we introduce the analysis of quantitative dental mesowear, based on the measurement of the angle and height of cusps from photographs. The use of this method on 26 present-day domestic sheep with two different feeding strategies showed very significant differences: the group with a dominant rangeland diet has high cusps with acute angles, and the group with a dominant grassland diet has lower cusps with obtuse angles. This method has been compared with the traditional technique of mesowear score, which consists in the qualitative categorisation of cusps into seven categories. However, the analysis showed that mesowear score is more prone to observer error than quantitative mesowear. In addition, quantitative mesowear better reflects the variability in dietary behaviours of extant sheep since observations are not limited into seven categories. To test quantitative mesowear in archaeological samples, we analysed caprine mesowear at two Iron Age sites (sixth-fifth centuries BC) from the northeast of the Iberian Peninsula: Empúries and Ullastret. The results show that caprids at the Ullastret site have a slightly more abrasive diet than at Empúries, which is consistent with data obtained from dental microwear in previous work. Overall, we show the potential of this technique, which can be combined with traditional techniques, to distinguish past caprine feeding strategies.
... While grazers typically show abrasion-dominated tooth wear, resulting in low, blunt cusps on the molar crowns, browsers cover the other end of the herbivorous dietary spectrum with attrition-dominated tooth wear, showing higher, sharper molar cusps. Though successfully applied in many instances to observational data of extant as well as extinct species (Franz-Odendaal & Kaiser 2003, Clauss et al. 2007, Rivals et al. 2007, Schulz et al. 2007, Croft & Weinstein 2008, mesowear has rarely been applied to animals fed experimental diets in controlled conditions (Solounias et al. 2014, Kropacheva et al. 2017, Ackermans et al. 2018, so understanding the extent of the dietary signal assessed by mesowear scoring is therefore still limited. ...
... Originally developed by Fortelius and Solounias (2000), the mesowear method consisted of two tooth characters: the cusp shape and the cusp relief of ungulate upper molars, scored for one apex per individual. This was further expanded to lower molars for equids (Kaiser & Fortelius 2003) and artiodactyls (Franz-Odendaal & Kaiser 2003). The original technique and its expanded version were termed "mesowear I" by Solounias et al. (2014). ...
Conference Paper
Musk oxen are large tundra‐dwelling bovids that, when in rut, can strike heads at speeds around 30 mph. The wide base of their large horns armor their skull against such impacts, although it is unknown if they sustain brain trauma. There is no literature describing concussion, or other adverse repercussions of severe cranial contact in musk oxen. Our research explores the brain and skull anatomy of combative bovids, specifically focusing on whether these animals sustain brain injury after head‐butting, and if not, what features most likely provide protection. Achieving a better understanding of how these extreme animals may avoid brain injury will provide insight to develop strategies for the reduction of traumatic brain injury in humans. An archived brain of an adult male musk ox was used in this preliminary study. It had been collected after humane euthanasia and preserved in formalin following a goring injury in the wild where it had been observed clashing heads with another musk ox bull in the moments before its death. The brain was MR scanned and examined histologically for evidence of traumatic brain injury. The right hemisphere was fixed in paraformaldehyde (4%, buffered) and prepared using a variety of exploratory histological protocols, to investigate abnormalities in neurons, microglia, astrocytes, and blood vessels. Specifically, Tau protein, a biomarker found in the cerebrospinal fluid and in neurodegenerative lesions, was used to detect any brain trauma‐related cellular consequence of chronic or acute head clashing. Preliminary histological results indicate no apparent abnormalities. A high‐resolution 7‐T MRI scan additionally revealed no abnormal neuropathological changes. This brain being a single sample does not allow us to make sweeping assumptions. However, these preliminary investigations lead us to believe that musk oxen do not appear to suffer from chronic or acute brain trauma after head clashing. A more in‐depth exploration with a larger sample size is proposed to better understand the combination of anatomical structures contributing to brain protection in these animals.
... We employed two different tooth wear proxies (mesowear and microwear) to reconstruct the probable diet of the hipparion(s) from Cioburciu. Mesowear analysis has been used extensively to interpret paleodiet (e.g., Kaiser et al. 2000;Kaiser & Fortelius 2003;Kaiser & Solounias 2003;Franz-Odendaal & Kaiser 2003;Semprebon et al. 2004a;Mihlbachler & Solounias 2006;Rivals & Semprebon 2006;Semprebon & Rivals 2007;Mihlbachler et al. 2011, Ackermans et al. 2020. Two variables are assessed using the mesowear technique: molar cusp shape and occlusal relief (the relative difference in height between cusp apices and intercusp valleys) in lateral (buccal) view. ...
Full-text available
The Cioburciu hipparions, Republic of Moldova, are included in a Turolian assemblage, approximately dated between 9 and 7 million years. We assess herein their taxonomic position, systematics, biogeography and paleodietary habits. We have undertaken standard equid measurements as well as accessing the Vera Eisenmann website for measurements and images and analysed craniodental and postcranial elements. This assemblage has been determined to be of a medium-sized hipparion with an elongated muzzle, well developed preorbital fossa that is dorsoventrally extensive and placed close to the orbit, lacking a caninus fossa and having a prominent and deep buccinator fossa. As such, this assemblage is referable to Cremohipparion moldavicum Gromova, 1952 common to the Western Ukraine, Balkans, Romania, Republic of Georgia, Turkey and Iran. We have employed a combination of gross cheek tooth wear morphology utilizing the mesowear method and a microscopic analysis of occlusal enamel scars utilizing the light microscope microwear technique. These complementary paleodietary methods indicate that these hipparions engaged in a mixed feeding dietary behavior and that the Cioburciu sample of C. moldavicum likely alternated its diet between browsing and grazing seasonally and/or regionally. A hierarchical cluster analysis based on average scratch and pit numbers positions this taxon among extant mixed feeding ungulates. Large pitting and gouging assessed through the microwear technique indicates occasional consumption of relatively coarser foods than typical mixed feeders or grazers or grit-laden food just prior to death while mesowear indicates that this was not a lifetime habit.
... Vegetation varies with soil structure, available water and photoperiodicity (Happolds and Happolds, 1987;Calandra et al, 2016). Dental-based method as suggested by Fortelius and Solounias (2000), Kaiser (2002) and Franz-Odendaal and Kaiser (2003) allows for dietary reconstructions through the use of mesowear method on dental structures alone and has been employed in comparison-inference based differences in food availability and habitat structure (Schulz et al, 2007). This study captions age class identity diet type by mesowear equilibrium and age segregations in relation to fauna quality (ecologic zones). ...
Full-text available
Tooth-wear signatures obtained from maxillary carnassial fourth premolar teeth of raccoons in three ecologic regions in Nigeria testified to segregations in diet of the species with more abrasive diet in specimens from coastal south-western areas compared to more vegetal diet content of those from middle belt and northern areas. Endoloph assessments showed sexually dimorphic mesowear signals between and within locations suggestive that males are more exposed to dental wears compared to females; Male and female specimens from rainforest zone had 40.2% and 34.2% respectively, Sudan Savanna zone had 46.8% and 40.6% for females and males while 67.6% and 44.3% for Sahel zone specimens in similar order. We investigated dietary resource use for sustained survivability within limits of interspecific spatial overlaps using seasonal rainfall indices between two years. There was 86% per high dental occlusal surface relief in the specimens from the savannas while 32% per low relief was observed in South-Western badgers teeth samples. This study observed a change in habitat use as a predisposing factor to sub-regional dental wear differences among age groups as well as sexes of species from three geographic climatic areas. The richness of the eco-habitat/life expectancy found in the rain forest can be ascribed to diet availability which is reduced in the savanna areas. The study suggests minimal change in habitat use as a predisposing factor in sub-regional species dental relief differences observed among age groups and sexes of the species from three geographic climatic areas and also represents quality of the eco-habitats.
... Relatively small sample series (less than 10 specimens) give reliable and significant results (Fortelius and Solounias, 2000). Although in most cases, only M2 is used in mesowear analysis, here all three upper molars were scored for mesowear on the buccal side of the tooth to increase the sample size (Franz-Odendaal and Kaiser, 2003; see also Van Asperen and Kahlke, 2015). The occlusal relief was scored as either high or low. ...
Full-text available
The wooly rhinoceros (Coelodonta antiquitatis) and forest rhinoceros (Stephanorhinus kirchbergensis), were prominent representatives of the Middle and Late Pleistocene glacial and interglacial faunas of Eurasia. Their diet has traditionally been inferred on functional morphology of the dentition and skull. In rare cases, food remains are preserved in the fossas of the teeth or as gut content. New approaches to infer diet include the study of isotopes and mesowear. Here we apply all four methods to infer the diet of these emblematic rhinoceros' species and compare the food actually taken with the food available, as indicated by independent botanical data from the localities where the rhinoceros' fossils were found: Gorzów Wielkopolski (Eemian) and Starunia (Middle Vistulian) as well as analysis of literature data. We also made inferences on the season of death of these individuals. Our results indicate that the woolly rhino in both Europe and Asia (Siberia) was mainly a grazer, although at different times of the year and depending on the region its diet was also supplemented by leaves of shrubs and trees. According to the results of isotope studies, there were important individual variations. The data show a clear seasonal variation in the isotope composition of this rhino's diet. In contrast, Stephanorhinus kirchbergensis was a browser, though its diet included low-growing vegetation. Its habitat consisted of various types of forests, from riparian to deciduous and mixed forests, and open areas. The diet of this species consisted of selected items of vegetation, also including plants growing near both flowing and standing waters. The food remains from the fossae of the teeth indicated flexible browsing, confirming the previous interpretations based on functional morphology and stable isotopes. Long-term data from mesowear and microwear across a wider range of S. kirchbergensis fossils indicate a more mixed diet with a browsing component. The different diets of both of rhinoceros reflect not only the different habitats, but also climate changes that occurred during the Late Pleistocene.
The dietary guild structure of ungulate communities is a useful paleoecological tool for understanding the context of hominin paleobiology and evolution. Ungulates are well represented in the fossil record, and their dietary preferences reflect those of major habitat types. However, paleoecology relies on modern ecological patterns as analogs for recreating ecologies of the past. It has previously been suggested that for much of the Pliocene, no such modern analogs exist for the herbivore communities associated with hominins in eastern Africa. This study aims to determine whether the ungulate community associated with A. afarensis at the Pliocene site of Laetoli, Tanzania, shares similarities with extant communities or whether it lacks a modern analog. Our multiproxy approach using mesowear, hypsodonty, and stable carbon isotopes of tooth enamel to infer the diets of ungulates in the Upper Laetolil Beds shows that this community is dominated by browsers and mixed feeders and has a very low prevalence of grazers and frugivores. This dietary guild composition distinguishes the Upper Laetolil Beds from modern African communities and suggests either that the Upper Laetolil Beds had a unique vegetation structure which was able to support a higher diversity of browsing ungulates than that exists in African ecosystems today or that it retained an ungulate community that was resilient to environmental change. The Upper Laetolil Beds ungulate community is also unique relative to other mid-Pliocene communities in eastern Africa, some of which are similar to extant communities, while others, such as Laetoli, lack modern counterparts. This suggests that A. afarensis was a eurytopic species that inhabited a variety of ecosystems, including those with and without modern analogs. The co-occurrence of both analog and nonanalog communities in the Pliocene suggests that the transformation toward ungulate communities of modern aspect occurred asynchronously in eastern Africa.
The ecological preferences of ruminant artiodactyls are commonly used to reconstruct the paleoenvironment of Neogene fossil localities throughout Africa. However, comparatively little research has focused on the ecology of ruminant artiodactyls from the Miocene of Africa. Here, we contribute new molar mesowear and hypsodonty data for the ruminant artiodactyls from the early and middle Miocene of Kenya and Uganda. Macroscopic dental characteristics of 608 tragulids, stem pecorans, giraffoids, and bovids dated to between 20 and 13.7 Ma were analyzed. Our hypsodonty results reveal that, whereas tragulids remain brachydont throughout the early and middle Miocene, pecoran ruminants experience an increase in hypsodonty due to the appearance of high-crowned bovids and climacoceratids that migrate into eastern Africa in the middle Miocene. Results from dental mesowear analysis suggest that all tragulids and pecorans were likely browsers, with only one taxon showing mesowear values that overlap with both browsers and mixed feeders in both the upper and lower molars (Canthumeryx sirtensis). None of the taxa analyzed had mesowear scores indicative of a grazing diet. Surprisingly, middle Miocene bovids and climacoceratids, despite possessing gross tooth morphologies adapted to abrasive diets, were largely utilizing a browsing diet. Although the early and middle Miocene habitats of eastern Africa were likely very heterogenous, none of the ruminant artiodactyls present in these habitats is interpreted as having incorporated grasses into their diet in significant quantities.
Full-text available
Published mesowear data was reviewed from the year 2000 to November 2019 (211 publications, 707 species, 1,396 data points). Mesowear is a widely applied tooth wear technique that can be used to infer a herbivore's diet by scoring the height and sharpness of molar tooth cusps with the naked eye. Established as a fast and efficient tool for paleodiet reconstruction, the technique has seen multiple adaptations, simplifications, and extensions since its establishment, which have become complex to follow. The present study reviews all successive changes and adaptations to the mesowear technique in detail, providing a template for the application of each technique to the research question at hand. In addition, the array of species to which mesowear has been applied, along with the equivalent recorded diets have been compiled here in a large dataset. This review provides an insight into the metrics related to mesowear publication since its establishment. The large dataset overviews whether the species to which the various techniques of mesowear are applied are extant or extinct, their phylogenetic classification, their assigned diets and diet stability between studies, as a resource for future research on the topic. Subjects Evolutionary Studies, Paleontology
Full-text available
Addresses problems of coronal morphogenesis, amelogenesis, food comminution and digestion, amastication, tooth eruption and wear, in order to identify functional intrrelations and developmental constraints in the evolution of cheek tooth morphology. Many aptive features probably or certainly did not arise for their current functions, but are one-time constraints which have become incorporated into functional systems (exaptations rather than adaptations).-from Author
Full-text available
Worn enamel surfaces of the cheek teeth in fossil and recent rhinoceroses are characterized by fine, parallel ridges aligned perpendicular to the enamel-dentin interface. We show that these ridges result from an unusual enamel ultrastructure in which a primitively horizontal layering of the prisms has become vertical. The new structure apparently appeared between early and middle Eocene, at the time when the superfamilies of perissodactyls were rapidly diverging. Similar modifications of the enamel structure occurred in certain parts of the cheek teeth in tapiroids, chalicotherioids and brontotherioids, but hardly at all in the equids. The modified enamel structure, where it occurs in groups other than rhinocerotoids, is associated with lophs but not cusps. Experimental evidence shows that the modified enamel is more resistant to wear than the unmodified enamel. The consistent association with thin lophs rather than cusps suggests that the modified enamel evolved to prolong the life of the lophs, where occlusal pressures are highest and attrition greatest. The dominance of modified enamel in rhinocerotoids correlates with the higher degree of compression of the cusps and extreme lophodonty in this group. The absence of modified enamel structure in the equids, even in the ectoloph, correlates with the lesser importance of the ectoloph in equids relative to brontotherioids, chalicotherioids and rhinocerotoids.
Full-text available
The tooth morphology in herbivorous mammals with a one-phased power stroke shows frequently a functional symmetry between upper and lower dentitions. In large mammals the schmelzmuster normally does not differentiate between leading and trailing edges. In contrast such differentiations were found in some rodents with evergrowing molars. The highest degree of schmelzmuster functional symmetry occurs in voles (Arvicolinae). The biomechanic value of radial enamel is demonstrated by its regular occurrence in the push side of each cutting edge. ZUSSAMMENFASSUNG: Die Zahnmorphologie vieler Pflanzenfresser mit einem einphasigen Kraftschluß der Kiefer zeigen eine funktionale Symmetrie zwischen den Bezahnungen des Ober-und des Unterkiefers. Bei den Großsäugern lassen sich Luv-und Leeseite nicht im Schmelzmuster unterscheiden. Dagegen wurden solche Unter-schiede in den immerwachsenden Molaren einiger Nagetier gefunden. Die Wühl-mäuse (Arvicolinae) zeigen dabei den höchsten Grad von funktionaler Symmetrie im Schmelzmuster. Der biomechanische Wert des radialen Schmelzes wird durch dessen regelmäßigen Einbau auf der Druck-Seite des Schmelzbandes hervorgehoben.
Full-text available
A new approach of reconstructing ungulate diet, the mesowear method was recently introduced by FORTELlUS & SOLOUNIAS (in press). Expressions of tooth wear were found to have strong diagnostic capabilities for ungulate diets. The present study is the first test of the mesowear method in two ways: (1) to reconstruct the dietary regime of Hippotherium primigenium, an equid from the Vallesian Dinotheriensande (Germany) applying the mesowear method; (2) to test the robustness of the mesowear method by applying a blind test approach where several researchers scored the same sample of teeth independently of each other. As a consensus dietary diagnosis for Hippotherium primigenium, a mixed diet with grassy components similar to the diet of the impala (Aepyceros melampus) is suggested. We find the mesowear method to be efficient and robust. Kurzfassung Palaodiat-Analyse an Hippotherium primigenium aus den vallesischen Dinotheriensanden (Rheinhessen) mit der Mesowearmethode-eine Blindteststudie Ein neuer Ansatz zur Rekonstruktion der Palaodiat von Huftie-ren, die Mesowearmethode, wurde kurzlich von FORTELlUS & SOLOUNIAS (im Druck) beschrieben. Ein groBes diagnosti-sches Potential fur die Ernahrungsweise von Huftieren wurde in Merkmalen der Zahnabnutzung auf der Okklusalflache er-kannt. Die vorliegende Untersuchung ist in zweifacher Hin-sicht der erste Test der Mesowearmethode. (1) Es wird die Diat des hipparionten Equiden Hippotherium primigenium aus den vallesischen Dinotheriensanden (Rheinhessen, Deutsch-land) unter Anwendung der Mesowearmethode rekonstruiert. (2) Um die Robustheit der Methode zu uberprufen, wird eine Blindteststudie durchgefuhrt, in der die 5 Autoren dieselbe Sammlung oberer zweiter Molaren unabhangig voneinander untersuchen. Als Konsensusdiat fUr Hippotherium primigenium, wird eine gemischte Nahrungszusammensetzung mit Gras-anteil, ahnlich der des Impala (Aepyceros melampus) vor-geschlagen. Die Mesowearmethode hat sich als effektiv und robust erwiesen.
Full-text available
Sümeg is a late Vallesian (MN10) karst-fissure locality situated about 60 kilometers north of the western end of Lake Balaton. We update the biochronologic ranking of critical late Miocene (MN9-MN12) Hungarian localities below based on the stage-of-evolution of murid, cricetid and anomalomyid lineages in order to securely place Sümeg's chronologic position. This diverse vertebrate fauna includes two species of hipparionine horses that we refer here to Hippotherium sumegense and "Hipparion" sp. small. Hippotherium sumegense has short, wide and shallow metapodials and is believed to be a late derived form of the Central European Hippotherium s.s. lineage. This species is believed to be the same as the one that appears in the Vallesian Austrian locality of Götzendorf. "Hipparion" sp. small is represented by very little material and as such has an indefinite phylogenetic position, but is plausibly related to the small radicle of the Cremohipparion lineage, and as such may represent an immigrant from the eastern Mediterranean. Our various analyses suggest that the larger species Hippotherium sumegense was a non-cursorial forest denizen with a significant browse component in its diet while "Hipparion" sp. small was likely a cursorial form that had a mixed graze-browse diet.
Full-text available
The Turolian vertebrate locality of Dorn-Dürkheim is situated near the city of Mainz, SW-Germany. The mammalian fauna is significantly younger than most other late Miocene Central European faunas that have yielded an equally rich assemblage of hipparion remains. The Dorn-Dürkheim hipparion fauna consists of isolated teeth and postcranial skeletal elements. We use the Vallesian (MN9) samples ofHippotherium primigenium from Eppelsheim and Höwenegg (Germany) as standards for comparison. Based on cheek tooth occlusal dimensions and continuous variables of the astragali, we identify two clusters of dental and postcranial specimens. We argue that these two clusters belong to two populations of hipparions that differ from one another principally in their body size. We further investigate the populations from Eppelsheim (EPhP), the total of all Dorn-Dürkheim specimens (DDpPall) and the two sub-populations from Dorn-Dürkheim (DDhPri and DDhPsm) with respect to their dietary preferences. For this we use the microwear and mesowear methods. The paleodietary signals for each sub-population are found to be quite different. The larger sized population of Dorn-Dürkheim (DDhPri) was a mixed feeder, while the small sized population (DDhPsm) is interpreted to have been a dedicated browser. In addition, we redefine the calculation of indices of hypsodonty so that they are more appropriate to equine horses. In comparing the hypsodonty indices of the hipparion populations from Eppelsheim and the total of the Dorn-Dürkheim specimens we find no differences in hypsodonty. In our comparison of Dorn-Dürkheim metacarpal III’s (MC III’s) with MC III’ from other pertinent European localities, we find two MC III’s from Dorn-Dürkheim to be similar to the Höwenegg population ofH. primigenium. However, eight Dorn-Dürkheim specimens differ from the Höwenegg population in having a relatively expanded crista sagittalis and reduced lateral and medial condyles. The only complete Dorn-Dürkheim MC III is relatively longer than the those from the Höwenegg population. This leads us to conclude that the MC III morphology of the smaller sized sub-population from Dorn-Dürkheim (DDhPsm) exhibits an adaptation for more cursorial locomotion than the Höwenegg hipparions, while the larger sized sub-population ventured into less forested habitats and was less cursorial. Based on the peculiarities of metapodial build and of cheek tooth dimensions, we recognize the population DDhPsm from Dorn-Dürkheim as belonging to a new species of hipparionine horse,Hippotherium kammerschmitti. New species:Cormohipparion n. sp.,Hippotherium kammerschmitti n. sp. Die turolische Wirbeltierlokalität Dorn-Dürkheim liegt nahe von Mainz in Südwestdeutschland. Die Säugetierfauna ist deutlich jünger als die meisten übrigen obermiozänen mitteleuropäischen Faunen und hat ein reiches, überwiegend aus isolierten Zähnen und postcranialen Skelettelementen bestehendes Hipparionenmaterial geliefert. Als Referenz für Vergleiche wird das vallesische Fundgut vonHippotherium primigenium von den Lokalitäten Eppelsheim und Höwenegg herangezogen. Basierend auf den occlusalen Dimensionen der Backenzähne und den Abmessungen von Astragali und Metapodien werden zwei Gruppen dentaler und postcranialer Individuen identifiziert. Diese zwei Gruppen werden zu zwei verschiedenen Unterpopulationen hipparioner Pferde gestellt, die sich im wesentlichen durch ihre Körpergröße unterscheiden. In Bezug auf ihre Nahrungspräferenzen werden ferner die Populationen von Eppelsheim (EPhP), die Gesamtheit aller Dorn-Dürkheim-Individuen (DDhPall) sowie die zwei Unterpopulationen von Dorn-Dürkheim (DDhPri and DDhPsm) untersucht. Hierfür kommen die Mikrowear- und die Mesowearmethode zum Einsatz. Die großwüchsige Population von Dorn-Dürkheim (DDhPri) wird als Mischkostfresser (mixed feeder) erkannt, während die kleinwüchsige Population (DDhPsm) als ausgesprochener Konzentratselektierer (browser) interpretiert wird. In der vorliegenden Untersuchung wird ferner die Berechnung des Hypsodontie-Index neu definiert, um den besonderen Gegebenheiten der Equiden besser gerecht zu werden. Zwischen den Hipparion-Populationen von Eppelsheim und der Gesamtheit der Dorn-Dürkheim-Individuen bestehen keine Unterschiede im Hypsodontie-Index. Im Vergleich der Metacarpalia III (MC III) von Dorn-Dürkheim mit Hipparionen verschiedener europäischer Lokalitäten sind zwei MC III’s von Dorn-Dürkheim der Höwenegg-Population vonH. primigenium im Bau sehr ähnlich. Acht MC III’s von Dorn-Dürkheim zeigen jedoch eine verhältnismäßig ausgedehnte Crista sagittalis und reduzierte laterale und mediale Condylen. Das einzige vollständig erhaltene Metacarpale III von Dorn-Dürkheim ist im Verhältnis zur Höwenegg Population relativ lang gestreckt. Hieraus wird geschlossen, dass die eine Subpopulation von Dorn-Dürkheim (DDhPsm) eine Adaptation zeigt, die gegenüber den Höwenegg-Hipparionen auf verstärkte kursoriale Lokomotion verweist. Dem gegenüber war die großwüchsigere Subpopulation weniger cursorial adaptiert. Basierend auf den Besonderheiten im Bau der Metacarpalia und in den Abmessungen der Backenzähne wird die Population DDhPsm als eine neue Art hipparioner Pferde erkannt,Hippotherium kammerschmitti.
A new approach of reconstructing ungulate diets, the mesowear method, was introduced by Fortelius, Solounias (2000). Mesowear is based on facet development on the occlusal surfaces of the teeth. Restricting mesowear investigation on the M2 as has previously been suggested would limit application of the mesowear methodology to large ungulate assemblages. Most of the fossil, subfossil and recent ungulate assemblages that have been assigned to a single taxon have a smaller number of individuals. This results in the demand to extend the mesowear method to further tooth positions in order to obtain stable dietary classifications of fossil taxa. The focus of this paper is to test, if a consistent mesowear classification is obtainable for the remaining positions of the upper cheek tooth dentition (P2, P3, P4, M1 and M3) and for combinations of these tooth positions. For statistical testing, large assemblages of isolated cheek teeth of the Vallesian hipparionine horse Hippotherium primigenium Meyer, 1829 and of two populations of the recent zebra Equus burchelli Gray, 1824 are employed as models. Subsequently, all single cheek tooth positions and all possible combinations of these tooth positions are tested for their consistency in classification of the mesowear variables compared to the M2, the model tooth of Fortelius, Solounias (2000). As the most consistent model for the proposed "extended" mesowear method, the combination of four tooth positions P4, M1, M2, and M3 is identified, which allows to include the largest number of isolated tooth specimens from a given assemblage, and fulfills the demand of being consistent in the dietary mesowear classification with the "original" mesowear method. We propose the "extended" mesowear method to be particularly well suited for the reconstruction of paleodiets in hypsodont equids. © Publications Scientifiques du Muséum national d'Histoire naturelle, Paris.
The mesowear method evaluates the wear patterns of herbivore cheek teeth by visually evaluating the facet development of the occlusal surfaces. It thus allows classification of most herbivorous ungulates into browsers, grazers or intermediate feeders, due to the fact that in grazers, tooth wear is characterized by a comparatively high degree of abrasion, most probably due to the presence of silicacious phytoliths in grasses, a higher amount of dust and grit adhering to their forage, or both. It has been suggested that excessive tooth wear could be a particularly limiting factor in the husbandry of captive large browsing species, and major tooth wear was demonstrated in captive as compared to free-ranging giraffe. If this increased tooth wear in captivity was an effect of feeding type and diets fed, then it would be expected that other browsing species are affected in a similar manner. In order to test this hypothesis, we investigated the dental mesowear pattern in captive individuals of 19 ruminant species and compared the results to data on free-ranging animals. Compared to free-ranging populations, captive browsers show a significantly more abrasion-dominated tooth wear signal. The reverse applies to captive grazers, which tend to show a less abrasion-dominated wear in captivity. Captive ruminants were generally more homogenous in their wear signature than free-ranging ruminants. If grit contamination in the natural habitat is a major cause of dental wear in grazers, then diets in captivity, although similar in botanical composition, most likely contain less abrasives due to feeding hygiene. If dental wear is one of the major factors limiting longevity, then captive grazers should achieve longer lifespans than both captive browsers and free-ranging grazers. In particular with respect to browsers, the results suggest that captive feeding regimes could be improved.