KENYAPOTAMINAE AND

Boisserie, Jean-Renaud, Lihoreau, Fabrice, Orliac, Maeva, Fisher, Rebecca E., Weston, Eleanor M. & Ducrocq, Stéphane, 2010, Morphology and phylogenetic relationships of the earliest known hippopotamids (Cetartiodactyla, Hippopotamidae, Kenyapotaminae), Zoological Journal of the Linnean Society 158 (2), pp. 325-366 : 347

KENYAPOTAMINAE AND HIPPOPOTAMINAE

The close relationship of all known kenyapotamines with hippopotamines was a particularly robust result of the present cladistic analysis, reinforcing the initial attribution of Kenyapotamus to Hippopotamidae . However, the relationship of this genus with later hippopotamids needs some clarification. Until now, kenyapotamines represented the only known hippopotamids prior to the terminal Miocene. It would thus be tempting to see Kenyapotamus as a forerunner of Hippopotaminae . This would imply a dental evolution following several, relatively simple, trends: reduction of conules/-ids and fusion into cristae/-ids; elongation of cristae/-ids; increase in crown height (already observed in KNM-NA 251, Fig. 5B ). Some trends apparently expressed between K. ternani and K. coryndonae could have led to early hippopotamine morphological features, such as the development of a deep median indentation of the P 3 labial cervix, the migration of the mesiolabial parastyle toward the mesiostyle reaching close to coalescence, and the disappearance of the postectohypocristid. The transition from a kenyapotamine trigonid (recalling to some extent the trigonid of Cebochoerus ) towards a hippopotamine trigonid appears a more complex issue. It would involve a complete reduction of a developed premetacristid and a parallel considerable elongation of the small kenyapotamine endometacristid. An alternative, seemingly more parsimonious interpretation would be the disappearance of the endometacristid and a marked labial shift of the premetacristid. In this case, the resulting hippopotamine premetacristid would often be disconnected from the mesial cingulid and would join the lingual wall of the preprotocristid, modifying the cuspid relations seen in the other studied groups. The latter interpretation was not favoured in our nomenclatural approach but cannot be excluded. However, it represents the same number of evolutionary steps, and therefore would not change the cladistic results.

Whatever the interpretation of the endometapremeta-preprotocristid complex evolution, it would have to be combined with a reduction of the postprotoconid to reach the condition seen in hippopotamines. Overall, this would represent quite significant modification of the trigonid in terms of dental morphofunctionality. For now, we do not have any concrete evidence of this evolution. If hippopotamines had derived from the late Miocene Kenyapotamus , such evidence should be looked for, together with the other modifications cited above, within a relatively short time interval, between about 9 and 7.5 Mya.

Our observations and cladistic analyses, supporting a monophyletic Kenyapotamus , suggest a different scenario. The late Miocene forms could have developed their ‘hippopotamine-like’ traits in parallel with a hypothetical hippopotamine lineage, whilst retaining or developing independently kenyapotamine features, such as the trigonid and the P 3 protocone morphologies. It is striking that KNM-NA 251 ( Fig. 5B ), the most advanced known specimen of Kenyapotamus in terms of crown height, also exhibits one of the most complex patterns of cristae/fossae and conules, i.e. most different from the somewhat younger holotype of K. coryndonae (KNM-BN 1321, Fig. 5A ) and from hippopotamines.

Regarding hippopotamine intragroup relationships obtained here, the paraphyly of Archaeopotamus agrees with Weston’s (2000) interpretation of the morphology of A. lothagamensis , advocating the basal position of this species within Hippopotaminae contra Boisserie (2005). This analysis, including mostly dental remains, did not take into account the mandibular morphological features on which Boisserie (2005) based the diagnosis of Archaeopotamus .

Figures

Figure 5. Maxilla and upper molars of Kenyapotaminae. A, Kenyapotamus coryndonae, KNM-BN 1321, right M3, holotype with KNM-BN 2075 (left, labial view; middle, mesial view; right, occlusal view). B, K. aff. coryndonae, KNM-NA 251, right M3 (left, labial view; middle, mesial view; right, occlusal view). C, K. coryndonae, KNM-BN 1494, fragmentary palate with left M1 and right M2 (occlusal view). D, K. coryndonae, KNM-NA 188, right M1 (left, labial view; middle, mesial view; right, occlusal view). E, cf. Kenyapotamus, KNM-BN 1618, upper molar fragment (top, labial view; bottom, occlusal view). F, Kenyapotamus ternani, KNM-FT 3934, right M1 - holotype (left, labial view; middle, mesial view; right, occlusal view). G, K. cf. coryndonae, KNM-BN 1601, molar fragment (labial view). Localities: BN, Ngeringerowa, except KNM-BN 1618: lower members of the Ngorora Formation; FT, Fort Ternan; NA, Nakali. Scale bars = 1 cm.