Content uploaded by Tamara A Franz-Odendaal
Author content
All content in this area was uploaded by Tamara A Franz-Odendaal
Content may be subject to copyright.
675
GEODIVERSITAS • 2004 • 26 (4) © Publications Scientifiques du Muséum national d’Histoire naturelle, Paris. www.geodiversitas.com
Comparative dietary evaluations of an extinct
giraffid (Sivatherium hendeyi)
(Mammalia, Giraffidae, Sivatheriinae) from
Langebaanweg, South Africa (early Pliocene)
Tamara A. FRANZ-ODENDAAL
Department of Zoology, University of Cape Town,
Rondebosch, 7700, Cape Town (South Africa)
Current address: Department of Biology, Dalhousie University,
1355 Oxford Street, Halifax, B3H 4J1 (Canada)
tfranzod@dal.ca
Nikos SOLOUNIAS
Department of Anatomy, New York College of Osteopathic Medicine,
Old Westbury, NY 11568-8000 (USA)
nsolouni@nyit.edu
Franz-Odendaal T. A. & Solounias N. 2004. — Comparative dietary evaluations of an
extinct giraffid (Sivatherium hendeyi) (Mammalia, Giraffidae, Sivatheriinae) from
Langebaanweg, South Africa (early Pliocene). Geodiversitas 26 (4) : 675-685.
ABSTRACT
The dietary preference of Sivatherium hendeyi (Harris, 1976), an extinct giraf-
fid from the early Pliocene of South Africa, was investigated by applying three
dietary reconstruction tools – hypsodonty, mesowear and microwear. The
hypsodonty index for S.hendeyi is 1.51 ± 0.06, which is within the brachyo-
dont category as in most ruminant browsers. The mesowear signature of
S.hendeyi is most similar to the mixed feeders (the seasonal mixed feeders).
Microwear investigations also support a mixed diet for S.hendeyi.Taken
together, results indicate that the dietary preference of this extinct giraffid is
most similar to that of seasonal mixed feeders and show no similarities with
grazers. The slight differences in the type of mixed feeding are discussed and
highlight the constraints of each method for the interpretation of diets of
fossil herbivores. The importance of the results in terms of the evolution of
dietary strategies amongst African Sivatheriinae are also discussed.
KEY WORDS
Mammalia,
Giraffidae,
Sivatheriinae,
South Africa,
microwear,
mesowear,
paleodiet.
INTRODUCTION
The dietary preference of an early Pliocene giraf-
fid, Sivatherium hendeyi (Giraffidae Gray, 1821,
Sivatheriinae Zittel, 1993), is investigated by
means of three dietary reconstruction tools –
hypsodonty index, mesowear and microwear.
The determination from unworn teeth (used for
hypsodonty index) provides a broad indication of
diet, with more pronounced hypsodonty indicat-
ing a larger wear (attrition and abrasion). The
mesowear method is based on a combined evalu-
ation of the relative amounts of attritive and
abrasive wear on occlusal dental enamel and gives
a reasonably good estimation of the diet of an
animal throughout its life. Microwear investiga-
tions are also applied to provide insight of the
“last” meals of the animal prior to death.
The diet of fossil Giraffidae was until recently
thought to be similar to that of extant giraffes
(i.e. that they were committed browsers). In
other words, all Giraffidae were described tradi-
tionally as browsers. Such notion changed when
Solounias et al. (1988) showed that the extinct
giraffid Samotherium boissieri Major, 1888 from
the Miocene of Samos (Greece) was a mixed
feeder-grazer. Solounias and co-workers (Solou-
nias et al. 1988, 2000; Solounias & Moelleken
1993) showed, using tooth microwear analyses
and premaxillary shape, that the diets of fossil
Giraffidae are highly heterogeneous. For exam-
ple, the Sivatheriinae Bramatherium mega-
cephalum (Lydekker, 1878) and Sivatherium
giganteum Falconer & Cautley, 1835 were proba-
bly grazers. Of the two Sivatheriinae s.l.,
Giraffokeryx punjabensis Pilgrim, 1910, was a
mixed feeder whereas “Palaeotragus”primaevus
Churcher, 1970 was a browser. The Sivatheriinae,
Helladotherium duvemoyi Gaudry, 1860, was also
a browser. Among the Giraffinae and Palaeo-
traginae Pilgrim, 1911 there are also browsing,
grazing and mixed feeding taxa. In addition,
Solounias & Semprebon (2002) found that
the okapi (Okapia johnstoni (Sclater, 1901)), the
Franz-Odendaal T. A. & Solounias N.
676 GEODIVERSITAS • 2004 • 26 (4)
RÉSUMÉ
Caractérisations comparatives de l’alimentation d’un giraffidé éteint
(Sivatherium hendeyi) (Mammalia, Giraffidae, Sivatheriinae) de Lange-
baanweg, Afrique du Sud (Pliocène inférieur).
Les préférences alimentaires de Sivatherium hendeyi (Harris, 1976), un giraffidé
éteint du Pliocène inférieur d’Afrique du Sud, ont été étudiées en appliquant
trois outils de reconstitution du régime alimentaire : hypsondontie, méso-
usure (mesowear) et micro-traces d’usures (microwear). L’indice d’hypso-
dontie pour S.hendeyi est de 1,51 ± 0,06. Cet indice le place dans la catégorie
des ruminants herbivores brachyodontes. Le type de méso-usure observé pour
S.hendeyi est similaire à celui des espèces ayant une alimentation mixte (mixte
saisonnière). L’analyse des micro-traces d’usure confirme aussi une stratégie
de type alimentation mixte pour S.hendeyi. Considérés simultanément, les
résultats indiquent que la préférence alimentaire de ce giraffidé éteint est assez
similaire à celle des espèces ayant une alimentation mixte saisonnière et ne
montre aucune similarité avec les paisseurs. Les légères différences dans le type
d’alimentation mixte sont discutées et soulignent les contraintes de chaque
méthode pour l’interprétation de l’alimentation des herbivores fossiles.
L’importance de ces résultats en terme d’évolution des stratégies alimentaires
parmi les Sivatheriinae africains est également discutée.
MOTS CLÉS
Mammalia,
Giraffidae,
Sivatheriinae,
Afrique du Sud,
micro-traces,
méso-usure,
paléorégime.
second and rare extant species of Giraffidae, is
not a browser but a fruit-dominated browser while
the giraffe (Giraffa camelopardalis (Linnaeus,
1758)) can be redefined as a leaf-dominated browser.
Such dietary data suggest that giraffid grazing was
taking place before the expansion of C4grass-
lands. Some giraffids were mixed feeders around
6 to 8 My ago (Cerling et al. 1997), and as such,
they were feeding on C3 grasses, which are domi-
nant in wooded environments (in open meadows
and near the margins of water). Since, Harris &
Cerling (1998), report that African sivatheres
became grazers during the late Pliocene, when
the C4expansion took place, we were interested
to investigate the diet of the South African
sivathere S.hendeyi.
The dietary preferences of S.hendeyi are not only
important to help elucidate dental defects such as
enamel hypoplasia, which are particularly abun-
dant in this animal (Franz-Odendaal et al. 2003,
2004), but also because it should provide clarity
on the evolution of the diets of African
Sivatheriinae, especially in Southern Africa where
no data are currently available. None of these
methods have previously been applied to any of
the fauna of Langebaanweg. Previous dietary
interpretations of herbivores from this locality
were largely based on general comparisons with
extant animals (e.g., by Hendey 1976, 1981,
1983, 1984). Using this approach, Hendey
assumed that the short-necked giraffid,
S.hendeyi, was a browser similar to extant
giraffes. In the last 20 years, however, several
methods have been developed for determining
the diets of extinct mammals. Stable isotope
analysis were unable to determine the diet of
S.hendeyi and other herbivores because of a
strong C3dominant signature at Langebaanweg
which could be grazing on C3grasses or browsing
(Franz-Odendaal et al. 2002).
ABBREVIATIONS
HI hypsodonty index;
m2 mandibular second molar;
M2 maxillary second molar;
PQL Palaeontology Quaternary Langebaanweg
collection;
SAM South African Museum, Cape Town.
MATERIAL AND METHODS
S.hendeyi teeth from the Pelletal Phosphate
Member at Langebaanweg (18°9’E, 32°58’S),
South Africa, housed at the South African
Museum, Iziko Museums of Cape Town, were
examined. The HI was determined based on the
measurements of two completely unworn m3s.
The HI for S.hendeyi was determined according
to Janis (1988) where HI = unworn m3 height/
m3 width. Thirty M2 and 30 m2 were selected
for mesowear analyses according to Fortelius &
Solounias (2000). The mesowear method deter-
mines average diet based on two variables – cusp
relief and cusp shape. Both variables were deter-
mined by direct observation and the percentage
of teeth with high/low cusps and sharp/round/
blunt cusps was calculated for the species. These
variables were then plotted against HI, as recom-
mended by Fortelius & Solounias (2000). The
new light microscope method for examining
microwear scar topography established by
Solounias & Semprebon (2002) was applied to
52 lower third molars (Appendix A). Clear epoxy
casts were first made and then examined for
microwear scarring. The average number of pits
and scratches per cast was calculated, as recom-
mended by Solounias & Semprebon (2002), and
used to obtain an average for the species. Pits
were also classified as small or large and scratches
were qualitatively scored by determining whether
1) only fine scratches were present; 2) only coarse
scratches were present; or 3) whether a mixture of
fine and coarse scratches were present within the
counting area (see Solounias & Semprebon 2002
for details). The presence of cross scratches and
gouges was also noted. Statistical analyses were
performed using Statistica (version 6.0).
RESULTS
HYPSODONTY INDEX
S.hendeyi is brachyodont. However, it is slightly
more hypsodont than the giraffe. The published
HI for the extant giraffe, G.camelopardalis, is 1.2
(Janis 1988). The HI for S.hendeyi determined
Dietary evaluation of Sivatherium hendeyi (Giraffidae)
677
GEODIVERSITAS • 2004 • 26 (4)
here is 1.51 ± 0.06 (n = 2). By including nine
slightly worn m3s, a recalculation of HI gives
1.33 ± 0.1 (n = 11). HI is used here primarily to
evaluate mesowear results and to compare it to
extant giraffids (Giraffa camelopardalis and
Okapia johnstoni). Since there is no difference in
the mesowear results obtained when using 1.3 or
1.5 as the HI, the HI of 1.3 is used for S.hendeyi
in all graphical representations. Accession num-
bers as well as crown height and width measure-
ments are given in Appendix B.
MESOWEAR ANALYSIS
The mesowear of S.hendeyi is similar to both
browsers and mixed feeders (browsers such as the
gerenuk (Litocranius walleri (Brooke, 1879)) and
mixed feeders such as the lama (Lama glama
(Linnaeus, 1758)) and other mixed feeders). The
raw data for each mesowear variable for each sec-
ond molar is provided in Appendix C and D and
the absolute and relative scorings for upper and
lower teeth are provided in Table 1. The average
occlusal relief for upper teeth is very similar to
that in lower teeth with all S.hendeyi teeth having
high relief. In terms of cusp shape, maxillary
teeth have 63% round cusps, 37% sharp cusps
and no blunt cusps, whereas mandibular teeth
have slightly more rounded (77%) and fewer
sharp (23%) cusps than upper teeth (Fig. 1). No
blunt cusps were observed in the either dental
sample. Mesowear results for S.hendeyi were
plotted against mesowear variables obtained by
Fortelius & Solounias (2000) for 64 extant ungu-
late species in Figure 2. Bivariate plots, for the
three dietary classes (browsers, grazers, mixed
feeders), of percentage high occlusal relief against
HI indicates that S.hendeyi falls within the range
for browsers and mixed feeders. S.hendeyi is not
similar to grazers (no blunt cusps). The same
result is obtained when cusp shape is plotted
against HI (Fig. 3). That is, no similarities with
grazers were found.
MICROWEAR ANALYSIS
The microwear shows that S.hendeyi was a mixed
feeder. Examination of individual microwear
scoring for each specimen shows that approxi-
mately half of the teeth have microwear similar to
that of browsers and the other half similar to that
of grazers. This is a bimodal distribution of
microwear as in mixed feeders (Solounias &
Semprebon 2002). Average values for each
microwear variable are given in Table 2. Average
number of pits and scratches for S.hendeyi are
20.5% and 15.8% respectively, per unit area.
The range of pit counts per tooth was 3.5 to 73.
Franz-Odendaal T. A. & Solounias N.
678 GEODIVERSITAS • 2004 • 26 (4)
37
63
0
100
23
77
0
100
0
10
20
30
40
50
60
70
80
90
100
high relief sharp round blunt
M2
m2
FIG. 1. — Mesowear cusp relief and cusp shape for S.hendeyi
(Harris, 1976) lower (m2) and upper (M2) molars. Cusp relief
(100% high) and cusp shape (sharp and round) are shown. No
blunt teeth or teeth with low relief were observed.
TABLE 1. — Absolute and relative mesowear scorings of S.hendeyi (Harris, 1976) upper (M2) and lower (m2) second molars following
the scoring convention of Fortelius & Solounias (2000). Percentages of high, low, sharp, round and blunt cusps were calculated for
each dental sample. Abbreviation: n, number of specimens.
Counts Percentages
Cusp relief Cusp shape Cusp relief Cusp shape
nLow High Sharp Round Blunt % Low % High % Sharp % Round % Blunt
M2 30 0 30 11 19 0 0 100 37 63 0
m2 30 0 30 7 23 0 0 100 23 77 0
Average scratch counts were slightly more vari-
able ranging from 4 to 38. In S.hendeyi large pits
were present in 48.1% of teeth. The average pit
and scratch counts for browsing phase is 19.4 pits
and 10.9 scratches, compared to 22.1 pits and
23 scratches for the grazing phase.
Microwear data were compared to data obtained
by Solounias & Semprebon (2002) for 50 extant
ungulate species (Fig. 4). Taking both average
pits and average scratch counts into account,
S.hendeyi falls on the browser side of a division
between traditional grazers (more than 17 scratches)
and traditional browsers (less than 17 scratches)
(Fig. 4) (see Discussion below).
DISCUSSION
HYPSODONTY INDEX
The HI for S.hendeyi (1.51) falls within the cate-
gory of brachyodont as defined by both Janis (1988)
and Fortelius & Solounias (2000). The tooth mor-
phology is similar to that of other brachyodont
species. Using Janis’ (1988) interpretations of hyp-
sodonty, S.hendeyi also falls within the range for
mixed feeders in closed habitats, and all types of
browsers (regular, selective and high-level). In ad-
dition the HI of S.hendeyi is most similar to the av-
erage for selective browsers (1.5 ± 0.08). By
comparing the HI of S.hendeyi to the published
Dietary evaluation of Sivatherium hendeyi (Giraffidae)
679
GEODIVERSITAS • 2004 • 26 (4)
1
2
3
4
5
6
020406080100
1
2
3
4
5
6
020406080100
1
2
3
4
5
6
020406080100
HI
HIHI
Occlusal relief
Occlusal relief
Occlusal relief
A
B
C
FIG. 2. — Bivariate plots of percentage high occlusal relief against hypsodonty index (HI); A, browsers; B, grazers; C, mixed feeders.
Data from Fortelius & Solounias (2000) (O) and S.hendeyi (■) (Harris 1976).
Franz-Odendaal T. A. & Solounias N.
680 GEODIVERSITAS • 2004 • 26 (4)
SHARP ROUND BLUNT
A
1
2
3
4
5
6
020406080100 020406080100
020406080100
020406080100
020406080100
020406080100
02040
60 80 100
020406080100
020406080
100
B
1
2
3
4
5
6C
1
2
3
4
5
6
BROWSERS
D
1
2
3
4
5
6E
1
2
3
4
5
6F
1
2
3
4
5
6
GRAZERS
G
1
2
3
4
5
6H
1
2
3
4
5
6I
0
1
2
3
4
5
6
MIXED FEEDERS
FIG. 3. — Mesowear analyses of cusp shape in S.hendeyi (Harris, 1976). S.hendeyi data for lower () and upper () teeth are plotted
against 64 ungulate species () from Fortelius & Solounias (2000); A-Cbrowsers; D-F, grazers; G-I, mixed feeders. In all
graphs, y-axis represents hypsodonty index and x-axis represents one of the three mesowear variables (percentage sharp cusps in
A, D, G; percentage round cusps in B, E, H, and percentage blunt cusps in C, F and I).
HI of 127 species of living ungulates (Janis 1988),
S.hendeyi HI is most similar to that of two cervids,
Capreolus capreolus (Linnaeus, 1758) (roe deer)
and Blastocerus dichotomus Wagner, 1844 (marsh
deer) (Janis 1988). These two animals were classi-
fied by Janis (1988) as mixed feeders in closed
habitats. However, both of these animals have
body weights of less than 150 kg, which is signifi-
cantly smaller than that estimated for S.hendeyi.
Janis (1988) suggests that HI of fossil taxa should
be compared to animals of similar body size.
However there are very few living ungulates with
body masses equivalent to that of S.hendeyi.
S.hendeyi with its robust metapodials is estimated
to have been heavier than G.camelopardalis
(c. 1075 kg) and more similar to the living rhino-
ceroses (see Janis 1988). The published HIs of these
animals are 1.59 for Rhinoceros unicornis Linnaeus,
1758, 1.72 for R.sondaicus Desmarest, 1822,
2.24 for Diceros bicornis Linnaeus, 1758 and 1.2
for G.camelopardalis (see Janis 1988). The extant
giraffe is a high-level browser whereas the rhinoc-
eroses are mixed feeders in closed habitats.
MESOWEAR
Although cusp relief in the extant giraffe,
G.camelopardalis,and S.hendeyi are similar, cusp
shape differs markedly. In terms of numbers,
G.camelopardalis has 74% sharp cusps and 26%
round cusps compared to the less marked pattern
in S.hendeyi (63% sharp, 37% round) (Fortelius
& Solounias 2000). Bivariate plots of mesowear
variables indicate that S.hendeyi had a different
diet from those of the extant giraffes. The
sivathere appears to be most similar to mixed
feeders. No evidence for pure grazing was found.
MICROWEAR
Solounias & Semprebon (2002: table 2) present-
ed five types of diets ranging from extreme
browsing to extreme grazing. The four columns
in between the two extremes have been termed
the transitional browsing-grazing group. The
bimodal microwear pattern of S.hendeyi is simi-
lar to the mixed feeders of column 2 and in par-
ticular to Gazella granti Brooke, 1872 (Grant’s
gazelle), Tragelaphus scriptus (Pallas, 1766) (bush
buck) and other seasonal-regional mixed feeders
(Solounias & Semprebon 2002: table 2). Micro-
wear shows similarities with Tragelaphus scriptus
(19.1% pits, 15.8% scratches) and Gazella granti
(20.5% pits, 14.7% scratches). Solounias &
Semprebon (2002: table 2, column 2) includes
additional dietary categories to the mixed feeders;
such as fruit-browsers and leaf browsers.
Comparison of S.hendeyi to the leaf-dominated
browser Odocoileus hermionus (Rafinesque, 1817)
(22.7% pits, 14.2% scratches) is of interest.
Solounias & Semprebon (2002) also show that
seasonal and regional mixed feeders show a
greater dispersion of scratch numbers and a high-
er number of pits on average than typically seen
in grazers. S.hendeyi has a wide range of scratch
counts (from 1 to 34 per counting area) and a
high number of pits (up to 70) indicating greatest
similarity to the seasonal mixed feeders.
Microwear can also distinguish browsing from
grazing based on scratch textures (widths of
scratches). Solounias & Semprebon (2002) sug-
gest that grazers have more coarse scratches and
browsers and mixed feeders have more fine
scratches. S.hendeyi teeth have all three scratch
textures (fine, coarse and mixed), however coarse
scratches are the least prevalent (Table 2) illus-
trating that S.hendeyi was more similar to
browsers and mixed feeders.
In summary, microwear data seems to indicate
that S.hendeyi was probably a seasonal mixed
feeder – it browsed about 50% of the time and
grazed about 50% of the time on grasses that do
Dietary evaluation of Sivatherium hendeyi (Giraffidae)
681
GEODIVERSITAS • 2004 • 26 (4)
TABLE 2. — Microwear data for S.hendeyi (Harris, 1976). Average number of pits and scratches are reported. The percentage of large
pits (LP), cross-scratches (CS), gouges (GOU), fine, coarse and mixed scratch textures are also reported. Abbreviation: n, number of
specimens.
nPits Scratches LP CS GOU Fine Coarse Mixed
52 20.5 15.8 48.1 67.3 48.1 54.9 11.8 33.3
0
10
20
30
40
50
60
70
80
90
010203040
grazers mixed feeders B dry habitat B leaf dominated
B fruit dominated elephants hippopotamus
G. camelopardalis
S. hendeyi
Average number of pits
Average number of scratches
Traditional browsers Traditional grazers
FIG. 4. — Microwear data for S.hendeyi (Harris, 1976) () compared to extant ungulates (from Solounias & Semprebon 2002).
Browsers in dry, leaf-dominated and fruit-dominated habitats are shown.
not form many wide scratches. It must be noted
that Solounias & Semprebon’s (2002) analysis
correctly classified all extant browsers in their
database but only 38% of mixed feeders. The
correct classification is less reliable for mixed
feeders because of the very nature of what it
means to be a mixed feeder. Mixed feeders incor-
porate a relatively high percentage of browse in
their diet and may therefore be misclassified as
browsers. Since both microwear counts and
scratch textures suggest mixed feeding, this is
probably the likely diet of S.hendeyi in Southern
Africa.
CONCLUSION
All three dietary assessments of the African
sivathere, S.hendeyi from the early Pliocene, indi-
cate that this animal was not a grazer and is best
categorized as being a mixed feeder. Mesowear,
which incorporates HI, indicates that S.hendeyi
is most similar to abrasion-dominated seasonal
mixed feeders. Microwear data also classifies
S.hendeyi as a mixed feeder but more likely an
attrition-dominated mixed feeder. S.hendeyi was
clearly not an exclusive browser or grazer. Some
possible reasons for the differences in abrasion
and attrition within the mixed feeding category
are outlined below.
The Langebaanweg assemblage represents a num-
ber of populations or cohorts of animals that
accumulated over 0.5 My (Hendey 1981, 1984)
and it is therefore highly likely that animals died
at different times of the year and hence had diets
that varied seasonally. This could account for the
large differences in scratch and pit counts for
individual S.hendeyi teeth. Alternatively, a transi-
tion in overall diet was made, possibly relating to
the changing environmental conditions (Franz-
Odendaal et al. 2003, 2004). The microwear
dietary evaluation may be slightly less accurate
than the mesowear one, which incorporates HI,
and which gives average diet for the species (over
geological time) rather than scoring individual
“last” meals and averaging them out (as is the case
with the microwear method).
Another source of complexity is that the south-
western region of South Africa is dominated by a
unique fynbos vegetation. Fynbos is an evergreen
species-rich sclerophyllous macchia vegetation
unique to the Western Cape, South Africa. Scott
(1995) showed that fynbos was beginning to
establish itself during the period of deposition of
the Pelletal Phosphate Member at Lange-
baanweg c. 5 My ago. The microwear features
that this vegetation would cause are not known,
and therefore interpretations of what is causing
the observed gouges, cross scratches, large pits,
coarse scratches etc. in S.hendeyi is not clear.
Determining the wear features that fynbos pro-
duces is not easily accomplished as few extant
ungulates live in the fynbos areas of South Africa
today.
Acknowledgements
We would like to thank the staff of the SAM for
access to the Langebaanweg material. We are
grateful to The National Research Foundation
(South Africa) and the University of Cape Town
(South Africa) for funding this study. We also
thank Thomas Kaiser (Greifswald University,
Germany) for discussions, Floret Rivals (New
York College of Osteopathic Medicine, USA) for
the French translation, and Mikael Fortelius
(University of Helsinki, Finland) and Raymond
L. Bernor (Howard University, USA) for their
valuable comments on our manuscript.
REFERENCES
CERLING T. E., HARRIS J. M., MACFADDEN B. J.,
LEAKEY M. G., QUADE J., EISENMANN V. &
EHLERINGER J. R. 1997. — Global vegetation
change through the Miocene/Pliocene boundary.
Nature 389: 152-157.
FORTELIUS M. & SOLOUNIAS N. 2000. — Functional
characterization of ungulate molars using the abra-
sion-attrition wear gradient: a new method for
reconstruction paleodiets. American Museum
Novitates 3301: 1-36.
FRANZ-ODENDAAL T. A., LEE-THORP J. A. &
CHINSAMY A. 2002. — New evidence for the lack
of C4grassland expansions during the Early
Franz-Odendaal T. A. & Solounias N.
682 GEODIVERSITAS • 2004 • 26 (4)
Pliocene at Langebaanweg, South Africa. Paleo-
biology 28: 378-388.
FRANZ-ODENDAAL T. A., LEE-THORP J. A. &
CHINSAMY A. 2003. — Insights from stable light
isotopes on enamel defects and weaning in Pliocene
herbivores. Journal of Bioscience 28: 101-109.
FRANZ-ODENDAAL T. A., CHINSAMY A. & LEE-THORP
J. A. 2004. — High prevalence of enamel hypopla-
sia in an Early Pliocene giraffid (Sivatherium
hendeyi) assemblage. Journal of Vertebrate Paleon-
tology 24 (1): 235-244.
HARRIS J. M. & CERLING T. E. 1998. — Isotopic
changes in the diet of Giraffids. Journal of Vertebrate
Paleontology 18, Supplement to 3: 49A.
HENDEY Q. B. 1976. — The Pliocene fossil occur-
rences in “E” quarry, Langebaanweg, South Africa.
Annals of the South African Museum 69: 215-247.
HENDEY Q. B. 1981. — Palaeoecology of the Late
Tertiary fossil occurrences in “E” quarry,
Langebaanweg, South Africa, and a reinterpretation
of their geological context. Annals of the South
African Museum 84: 1-104.
HENDEY Q. B. 1983. — Palaeoenvironmental impli-
cations of the Late Tertiary vertebrate fauna of the
fynbos region. South African National Science
Programmes Report 75: 100-116.
HENDEY Q. B. 1984. — Southern African Late
Tertiary vertebrates, in KLEIN R. G. (ed.), Southern
African Prehistory and Paleoenvironments. A.A.
Balkema, Rotterdam: 81-106.
JANIS C. M. 1988. — An estimation of tooth volume
and hypsodonty indices in ungulate mammals, and
the correlation of these factors with dietary prefer-
ence, in RUSSEL D. E., SANATORO J. P. &
SIGOGNEAU-RUSSEL D. (eds), Teeth Revisited.
Proceedings of the VIIth International Symposium
on Dental Morphology, Paris, May 1986. Mémoires
du Muséum national d’Histoire naturelle, sér. C,
Sciences de la Terre 53: 367-387.
SCOTT L. 1995. — Pollen evidence for vegetational
and climatic change in southern Africa during the
Neogene and Quaternary, in VRBA E. S., DENTON
G. H., PARTRIDGE T. C. & BURCKLE L. H. (eds),
Paleoclimate and Evolution with Emphasis on
Human Origins. Yale University Press, New York:
65-76.
SOLOUNIAS N. & MOELLEKEN S. M. C. 1993. —
Determination of dietary adaptations of some
extinct ruminants determined by premaxillary
shape. Journal of Mammology 74 (4): 1059-1071.
SOLOUNIAS N. & SEMPREBON G. 2002. — Advances
in the recontruction of ungulate ecomorphology
and application to early fossil equids. American
Museum Novitates 3366: 1-49.
SOLOUNIAS N., TEAFORD M. & WALKER A. 1988. —
Interpreting the diet of extinct ruminants: the
case of a non-browsing giraffid. Paleobiology 14:
287-300.
SOLOUNIAS N., MCGRAW W. S., HAYEK L. &
WERDELIN L. 2000. — The Paleodiet of Giraffidae,
in VRBA E. S. & SCHALLER G. B. (eds), Antelopes,
Deer and Relatives: Fossil Record, Behavioral Ecology,
Systematics, and Conservation. Yale University Press,
New Haven, Connecticut: 84-95.
Submitted on 6 November 2003;
accepted on 16 April 2004.
Dietary evaluation of Sivatherium hendeyi (Giraffidae)
683
GEODIVERSITAS • 2004 • 26 (4)
Franz-Odendaal T. A. & Solounias N.
684 GEODIVERSITAS • 2004 • 26 (4)
APPENDIX A
Accession numbers of mandibular m3s of Sivatherium hendeyi (Harris, 1976) used for microwear analysis. Abbreviations:
SAM,South African Museum, Cape Town; PQL,Palaeontology Quaternary Langebaanweg collection.
Accession number Accession number Accession number Accession number
SAM PQL 43964 SAM PQL 44024 SAM PQL 62737/2 SAM PQL 62737/27
SAM PQL 43966 SAM PQL 44034 SAM PQL 62737/4 SAM PQL 62737/28
SAM PQL 43976 SAM PQL 44945 SAM PQL 62737/5 SAM PQL 62737/32
SAM PQL 43994 SAM PQL 44948 SAM PQL 62737/11 SAM PQL 62737/34
SAM PQL 43995 SAM PQL 44950 SAM PQL 62737/12 SAM PQL 62737/36
SAM PQL 43999 SAM PQL 44952 SAM PQL 62737/15 SAM PQL 62737/36
SAM PQL 44003 SAM PQL 44956 SAM PQL 62737/16 SAM PQL 62737/39
SAM PQL 44004 SAM PQL 44959 SAM PQL 62737/19 SAM PQL 62737/43
SAM PQL 44008 SAM PQL 44970 SAM PQL 62737/22 SAM PQL 62737/44
SAM PQL 44011 SAM PQL 44971 SAM PQL 62737/23 SAM PQL 62737/45
SAM PQL 44012 SAM PQL 44972 SAM PQL 62737/24 SAM PQL 62737/49
SAM PQL 44020 SAM PQL 44976 SAM PQL 62737/25 SAM PQL 62737/50
SAM PQL 44021 SAM PQL 44977 SAM PQL 62737/26 SAM PQL 62738/38
APPENDIX B
Specimens used to calculate the hypsodonty index of Sivatherium hendeyi (Harris, 1976). Abbreviations: SAM, South African
Museum, Cape Town; PQL, Palaeontology Quaternary Langebaanweg collection.
Accession number Crown height (mm) Width (mm) Hypsodonty index
SAM PQL 62732/33, unworn 46.89 30.33 1.55
SAM PQL 62733/63, unworn 43.50 29.72 1.46
SAM PQL 62733/11 39.18 32.10 1.22
SAM PQL 62732/12 42.10 35.08 1.20
SAM PQL 62733/20 40.63 30.28 1.34
SAM PQL 62732/26 47.88 36.03 1.33
SAM PQL 62732/28 47.52 31.28 1.52
SAM PQL 62732/42 45.70 36.52 1.25
SAM PQL 62733/43 44.78 35.42 1.26
SAM PQL 62732/45 45.26 35.87 1.26
SAM PQL 62732/49 40.04 32.83 1.22
Dietary evaluation of Sivatherium hendeyi (Giraffidae)
685
GEODIVERSITAS • 2004 • 26 (4)
APPENDIX C
Raw mesowear data of Sivatherium hendeyi (Harris, 1976) lower second molars. Cusp shape is described as round (R), sharp (S) and
blunt (B). Occlusal relief is described as high (H) or low (L) depending on how high the cusps rise above the valley between them.
Abbreviations: SAM, South African Museum, Cape Town; PQL, Palaeontology Quaternary Langebaanweg collection.
Accession number Cusp shape Occlusal relief Accession number Cusp shape Occlusal relief
SAM PQL 43967 R H SAM PQL 44917 R H
SAM PQL 43976 S H SAM PQL 44968 R H
SAM PQL 43997 R H SAM PQL 44978 S H
SAM PQL 44001 R H SAM PQL 44970 R H
SAM PQL 44004 R H SAM PQL 44967 S H
SAM PQL 44008 R H SAM PQL 44966 R H
SAM PQL 44024 R H SAM PQL 44985 R H
SAM PQL 43994 S H SAM PQL 45167 S H
SAM PQL 43978 R H SAM PQL 44032 R H
SAM PQL 43889 S H SAM PQL 44021 R H
SAM PQL 44895 R H SAM PQL 43995 R H
SAM PQL 44956 R H SAM PQL 43959 R H
SAM PQL 44943 R H SAM PQL 44028 R H
SAM PQL 44957 R H SAM PQL 43992 S H
SAM PQL 44921 R H SAM PQL 44035 R H
APPENDIX D
Raw mesowear data of Sivatherium hendeyi (Harris, 1976) upper second molars. Cusp shape is described as round (R), sharp (S)
and blunt (B). Occlusal relief is described as high (H) or low (L) depending on how high the cusps rise above the valley between them.
Abbreviations: SAM, South African Museum, Cape Town; PQL, Palaeontology Quaternary Langebaanweg collection.
Accession number Cusp shape Occlusal relief Accession number Cusp shape Occlusal relief
SAM PQL 44658 S H SAM PQL 44743 S H
SAM PQL 44659 S H SAM PQL 44745 R H
SAM PQL 44661 R H SAM PQL 44759 R H
SAM PQL 44663 R H SAM PQL 44760 R H
SAM PQL 44668 S H SAM PQL 44740 R H
SAM PQL 44665 R H SAM PQL 44761 S H
SAM PQL 44660 S H SAM PQL 44739 S H
SAM PQL 44664 R H SAM PQL 44734 R H
SAM PQL 44674 R H SAM PQL 44736 S H
SAM PQL 44671 S H SAM PQL 44937 S H
SAM PQL 44672 R H SAM PQL 44729 S H
SAM PQL 44673 R H SAM PQL 44731 R H
SAM PQL 44670 R H SAM PQL 44723 R H
SAM PQL 44751 R H SAM PQL 44752 R H
SAM PQL 44755 R H SAM PQL 44758 R H
