Part I: Evolution, Taxonomy, and Domestication34
Results and Discussion
Figures 3.1, 3.2, and 3.3 illustrate the morphological changes in
third molars in the Nyanzachoerus–Notochoerus, Kolpochoerus–
Hylochoerus, and Metridiochoerus–Phacochoerus groups,
respectively, compared to the changes in stable carbon isotopes
of their enamel.
Most species of Nyanzachoerus display cranio-mandibular
and dental morphologies typical of omnivorous suids. Their
third molars align well with the omnivorous morphotypes
described earlier (Figure 3.1). Those species with omnivorous
morphotypes also display much enlarged third and fourth pre-
molars with a morphology convergent to that of bone-crushing
hyaenids. It has therefore been proposed that Nyanzachoerus
species were omnivorous but with a specialization to process
hard objects such as bones or nuts (e.g. Cooke and Wilkinson
1978). Nyanzachoerus jaegeri documents the transition with
Notochoerus around 4 Ma (this species is often placed in the latter
genus Notochoerus) with more elongated third molars. Species of
Notochoerus display cranio-mandibular and dental adaptations
to herbivory. They are extremely derived dentally with third
molars that are very elongated and very high-crowned (Figure
3.1). They also display reduced premolars and incisors, roughly
similar to those of the extant Hylochoerus and Phacochoerus
(Harris & White 1979). Those morphological changes are asso-
ciated with similarly dramatic increases in body size.
Stable carbon isotope values of early Nyanzachoerus indicate
diets dominated by C 3 plants up to around 5 Ma (Figure 3.1). After
5 Ma, most Nyanzachoerus and Notochoerus display δ^13 C values
indicating diets dominated by C 4 plants (Figure 3.1). However,
note that the proportion of C 3 plants consumed remains non-
negligible and even important in some cases. For example,
in Plio–Pleistocene sites of the Malawi Rift, the paleodiets of
Nyanzachoerus and Notochoerus specimens were strongly domi-
nated by C 3 plants (Lüdecke et al. 2016). The authors interpreted
those data as suggesting a more closed habitat than in con-
temporaneous sites of eastern Africa, but an equally plausible
hypothesis would be that those suids were consuming C 3 grasses.
Overall, the δ^13 C values of the Nyanzachoerus–Notochoerus
group are in good agreement with the ecomorphological recon-
structions, indicating omnivorous diets before 5 Ma and then a
shift towards herbivory after 5 Ma. However, we cannot rule out
the hypothesis that specimens older than 5 Ma were consuming
large amounts of C 3 grasses or forbs.
Species of Kolpochoerus from the first occurrence around
5.5 Ma to around 3 Ma are characterized by cranio-mandibular
and dental morphologies typical of omnivorous suids, nota-
bly with third molars very similar to those of Potamochoerus
(Figure 3.2). Around 3 Ma, there is a major cladogenetic split
with K. afarensis giving rise to two lineages. The first lineage,
Kolpochoerus limnetes (also called K. heseloni), displays a pro-
gressive anagenetic trend towards more and more elongated
third molars and gave rise around 1.5 Ma to much more derived
species with more elongated and especially higher-crowned
third molars (Figure 3.2). The premolars are reduced in the
younger specimens. The second lineage, starting with K. phil
lipi then continued by K. majus, is more conservative dentally
and displays a more moderate increase in length and heightof the third molars, and retains strongly developed premolars
up to its extinction at the end of the Pleistocene (Gilbert 2008;
Souron et al. 2015a). Kolpochoerus majus shares some cranio-
mandibular and dental features with the extant Hylochoerus but
transitional forms are unknown, the biggest morphological dif-
ference being the strong reduction of incisors and premolars in
Hylochoerus (Souron et al. 2015a).
δ^13 C values of Kolpochoerus older than 4 Ma indicate mixed
diets of C 3 and C 4 plants. As early as 3.8 Ma for some specimens
of K. afarensis, and after 3 Ma for most specimens, the inferred
diets are strongly dominated by C 4 plants, even for the den-
tally more conservative K. majus. Some populations of derived
Kolpochoerus from South Africa were consuming mostly C 3
plants, but those could have been C 3 grasses as all the plants
there were C 3 plants (Figure 3.2; Luyt et al. 2000). Note that the
C 3 plants-dominated diet in extant Hylochoerus does not mean
it is a browser (contra Harris & Cerling 2002; Cerling et al. 2015)
as the specimens analysed come from high-altitude areas where
grasses mostly use the C 3 photosynthetic pathway. Other stud-
ies indicate a mixed diet of various kinds of plants (including
grasses) in the extant giant forest hogs (e.g. d’Huart 1978).
Early members of Metridiochoerus from 3.8 Ma to 2.5 Ma
are characterized by cranio-mandibular and dental morpholo-
gies typical of omnivorous suids. Starting around 2.5 Ma, all the
species display third molars that align with herbivorous mor-
photypes (Figure 3.3). There is a dramatic increase in length and
height of third molars in the lineage M. andrewsi followed by
an even more derived M. compactus. There is another group of
Metridiochoerus with third molar morphologies very close to
that of Phacochoerus.
In Metridiochoerus, δ^13 C values as early as 3 Ma indicate diets
strongly dominated by C 4 plants. Again, a few South African
specimens and extant warthogs from high-altitude areas dis-
play diets with strong C 3 signals (Figure 3.3), but that does not
mean they were not consuming C 3 grasses in abundance.
A similar pattern is observed in Kolpochoerus and
Metridiochoerus with strong discrepancies between the diets
as expected from the dental morphology and the diets as
reconstructed from the δ^13 C values. Pliocene specimens with
third molars that would align with the omnivorous morpho-
types consumed large amounts of abrasive C 4 plants, presum-
ably grasses. In both groups, the strong morphological changes
(most notably the increase in length and height of third molars)
that take place during the Pleistocene are not correlated with an
increased consumption of C 4 plants. The key adaptations that
allowed those extinct suids to enter the herbivorous niches were
therefore likely not the dramatic changes of the third molars.Towards a New Scenario of Adaptations to
Herbivory in African Suids
As already observed by Harris and Cerling (2002) on more lim-
ited data, the review conducted here indicates that there is a
strong discrepancy between the changes in dental morphology
and the changes in diets as indicated by stable carbon isotopes.
To understand the herbivorous adaptations of the African suids,
it therefore seems crucial to use a more integrative approach.00512:31:27