Kenyapotamus, PICKFORD, 1983

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 : 344

publication ID 10.1111/j.1096-3642.2009.00548.x

persistent identifier

treatment provided by


scientific name




‘creature very close to the basal stock from which true hippopotamus was derived’; Coryndon, 1978a. ‘a genus of the Suidae similar to Bunolistriodon ’;

Coryndon, 1978: 291, fig. 18:7B. Palaeopotamus gen. nov.; Pickford, 2007b.

Emended diagnosis: Differs from hippopotamine genera notably by: P 3 protocone formed by a single bulge of the lingual cingulum; molar wear pattern generally less trefoliate; relatively marked paraconule on upper molars; frequent occurrence of two ectometafossules at least on M 1; more complex trigonid (more cristids/fossids); postprotocristid less reduced, extending more distally; M 3 ectohypocristulid often expressed as few ectoconulids. Differs from anthracotheriids, palaeochoerids, and suids notably in the following character association: lower incisors with rounded transverse sections, lacking marked fossids on the crown; P 4 lacking a clear parastyle and a metacone, retaining a complex paracone with developed paracristae and fossae; frequent occurrence of ectostyles on upper molars, with a parastyle tending to fuse to mesiostyle; occurrence of small, multiple distal conids on lower premolars; a lower molar postmetacristid elongating toward the centre of the crown; presence of a posthypofossid.

Discussion: The diagnosis proposed by Pickford (2007b: 99) for Palaeopotamus , a monospecific genus defined for K. ternani , listed features that are also found in the late Miocene specimens (see, for comparisons, Figs 5 View Figure 5 , 7 View Figure 7 ): upper molar protocone with an ‘oblique anterior crest’ (preprotocrista) joining the mesial ‘accessory cusplet in the centre line of the tooth behind the anterior cingulum’ (paraconule); deep sagittal valley, separating the labial and lingual cusps until quite deep wear; presence of labial and lingual ‘basal pillars rising from the cingulum’ at the transverse valley (ecto- and entostyles); presence of a postmetaconule (distal ‘hypoconule’); roots fused near the crown but separate root apices; deep mesial cingulum; on lower molars, mesial and distal cingulids well developed; the labial cingulid, although deep on the hypoconid, remains incomplete ‘with only a few pimples of enamel showing its course’, except at the transverse valley where a ‘basal pillar’ (ectostylid) occurs.

Wear may have played a role in shaping some of the proposed diagnostic features: this is the case for the paracone crests orientated directly mesially and distally from the apex (pre- and postparacristae). In the M 3 Bar 2171 ′ 01, direct observation indicated a position of the paracone cristae similar to that of other kenyapotamines, whereas wear has destroyed the salient ridges topping the cristae and formed artificial ridges. The attenuation of the lingual and labial cingula on the upper molars could also be related to wear. In addition, the holotype of K. ternani (KNM-FT 3934, Fig. 5F View Figure 5 ) exhibits well-developed, continuous cingula not limited to ‘small pustules’, a trait that featured prominently in the initial diagnosis of K. ternani , that was not revised by Pickford (2007b). Another proposed diagnostic trait is the ‘small median accessory cusplet’ (mesoconulid) not blocking the centre of lower molar median transverse valley in Palaeopotamus . The difference in mesoconulid position between the late Miocene lower molars and those from Kipsaramon or Maboko ( Fig. 7C View Figure 7 ) is however not obvious, all, including the M 1 Bar 1186 ′ 99 ( Pickford, 2007b: fig. 6 B2), having a similar central position on the crown. The mesoconulid of Bar 1186 ′ 99 actually appears more developed and differentiated than in other material. Consequently, none of the diagnostic features proposed for Palaeopotamus allow for a clear distinction from the late Miocene material attributed to K. coryndonae .

The comparative analysis performed here indicated discrete morphological differences between K. ternani and K. coryndonae . These differences are minor compared to those distinguishing the different genera of Hippopotaminae , essentially based on craniomandibular features (Boisserie, 2005), and do not involve marked differences of crown structures as between Kenyapotamus and Hippopotaminae . Moreover, our analyses support the monophyly of the Kenyapotaminae , and in our opinion taxonomy should reflect the phylogeny. Therefore, we prefer retaining a distinction of specific rank only for the two known species of Kenyapotaminae . Given the scarcity of material, we concede that future discoveries of middle Miocene specimens may alter this conclusion, indicating multiple lineages within kenyapotamines and, possibly, a closer link for some of these lineages with Hippopotaminae . For the time being, however, there is no clear evidence of multiple lineages, and it seems more cautious to retain only one genus and to treat Palaeopotamus as a junior synonym of Kenyapotamus .