A trigonid of a lower molar of a primitive, large hippopotamus from the upper lower Miocene of Mfwangano Island in southwestern Kenya is described. The molar trigonid is similar in size to that of living hippopotamuses, and is comparable in morphology to that of kenyapotamine hippopotamids (Mammalia, Cetartiodactyla) in having a brachyodont crown, bunodont cusps, an M-like structure on the distal trigonid wall, a single-ridged premetacristid, and a buccolingually bifurcate mesial root, and in lacking a paraconid. On the basis of its size and morphology, the specimen appears to be assignable to Kulutherium, which is a putative kenyapotamine previously known from the upper lower Miocene of Kenya and is so far represented only by the upper dentition. The present specimen provides additional evidence that a hippopotamus-sized, large hippopotamid was already living during the early Miocene. If it proves to be Kulutherium, it provides additional evidence that Kulutherium should be assigned to the Kenyapotaminae.
Introduction
The origin of the Hippopotamidae (Mammalia, Cetartiodactyla) is currently debated hotly by several researchers (e.g. Pickford, 2007b, 2008, 2011; Boisserie and Lihoreau, 2006; Boisserie et al., 2010, 2011; Orliac et al., 2010; Tsubamoto et al., 2011; Alloing-Séguier et al., 2014; Lihoreau et al., 2015; and references cited therein). The undoubtedly oldest record of the family is from the middle Miocene of Kenya, which has yielded a kenyapotamine hippopotamid, Palaeopotamus ternani (= Kenyapotamus ternani) (Pickford, 1983, 2007b, 2008, 2011; Boisserie and Lihoreau, 2006; Boisserie, 2007; Boisserie et al., 2010; Weston and Boisserie, 2010). A few years ago, Orliac et al. (2010) suggested that Morotochoerus from the lower Miocene of Uganda and Kulutherium from the lower Miocene of Kenya were representatives of primitive hippopotamids and tentatively assigned them to the Kenyapotaminae. Tsubamoto et al. (2011) also suggested a possible hippopotamid affinity of Kulutherium. In contrast, Pickford (2011) suggested an anthracotheriid affinity of Morotochoerus and Kulutherium. However, the fossil material of these two early Miocene genera is very limited (Pickford, 1998, 2007a).
This article describes a trigonid of a left lower molar of a primitive, large hippopotamus discovered in the upper lower Miocene Rusinga Group of Mfwangano (= Mfangano) Island in Lake Victoria, southwestern Kenya. The specimen has long been stored in the National Museums of Kenya, Nairobi, Kenya. It likely belongs to Kulutherium on the basis of its size, morphology, and geographical and geological evidence. Until now, Kulutherium has been represented only by the upper dentition, and its lower dental morphology has been unknown (Pickford, 2007a). The present study provides a possible lower molar morphology of Kulutherium.
Dental terminology.—Mostly follows Boisserie et al. (2010) and Orliac et al. (2010). Some dental terms of the lower molars used in this paper are shown in Figure 1B.
Dental abbreviations.—M/m, upper/lower molars.
Institutional abbreviations.—KNM, National Museums of Kenya, Nairobi, Kenya; NMMP-KU, specimens catalogued by the Myanmar-Japan (Kyoto University) Joint Fossil Expedition Team (Tsubamoto et al., 2006), which are currently stored in the Department of Archaeology (Ministry of Culture), Yangon, Myanmar; NMMP, National Museum, Myanmar, Paleontology; KU, Kyoto University, Japan.
Figure 1.
Cf. Kulutherium sp., KNM-MW 13143, a trigonid of a left lower molar, from the lower Miocene of Kenya. A, occlusal view; B, schematic drawing of occlusal view; C, lingual view; D, buccal view; E, ventral view; F, mesial view; G, distal view; H, distodorsal view. For A: 1 and 2, stereo pair.
Other abbreviations.—MOR, Moroto locality, Uganda (Pickford, 1998); MW, Mfwangano (= Mfangano) Island, Kenya; OCO, Orrorin Community Organisation, Baringo, Kenya; SH, Samburu Hills, Kenya.
Figure 2.
Comparisons of cf. Kulutherium sp. (KNM-MW 13143) with Kulutherium and Morotochoerus (cf. Kenyapotaminae [or Morotochoerinae]), Kenyapotamus and Palaeopotamus (Kenyapotaminae), and Myaingtherium (Anthracotheriidae). A, cf. Kulutherium sp., KNM-MW 13143, occlusal view of the trigonid of the left lower molar; B, Kulutherium kenyense Pickford, a cast of left M2 of R 773′49 (= holotype, of which original specimen is now missing and of which cast is stored in Natural History Museum, London, UK), occlusal view; C, Kenyapotamus coryndonae Pickford, a trigonid of left ml of KNM-SH 14792, occlusal view; D, Palaeopotamus ternani (Pickford) (= Kenyapotamus ternani Pickford), a trigonid of left ml of Bar 1186′99, occlusal view; E, Kenyapotamus coryndonae, a cross section of a left mandible at ml trigonid of KNM-SH 14789, mesial view, showing the buccolingually separated anterior root of m1; F, Morotochoerus ugandensis Pickford, a trigonid of right m3 (reversed) of a cast of MOR 177 (= holotype, stored in Uganda Museum, Kampala, Uganda), occlusal view; G, Myaingtherium kenyapotamoides Tsubamoto et al., a trigonid of left m3 of a cast of NMMP-KU 2245 (= holotype, currently stored in Department of Archaeology, Ministry of Culture, Yangon, Myanmar), occlusal view.
Figure 3.
Correlation of the general Miocene stratigraphy of Mfwangano and Rusinga islands in southwestern Kenya, with their ages and the occurrences of cf. Kulutherium sp. from Mfwangano Island and Kulutherium kenyense from Rusinga Island (after Drake et al., 1988; Pickford, 2007a; Peppe et al., 2009, 2011). These two islands are approximately 10 km apart.
Systematic paleontology
Family Hippopotamidae Gray, 1821
Subfamily cf. Kenyapotaminae Pickford, 1983
(sensu Boisserie et al., 2011)
Genus
Kulutherium
Pickford, 2007a
Type and only known species.—Kulutherium kenyense Pickford, 2007a (= Kulutherium kenyensis Pickford, 2007a [sic] = Kulutherium rusingense [sic] and Kulutherium rusingensis [sic] in Orliac et al., 2010). The gender of the genus is neuter; and the specific name of the type species is an adjective that comes from the place name (Pickford, 2007a). In this case, the gender of the specific name should be the same as that of the genus name according to Articles 31.2, 32.4, and 32.5 of the International Code of Zoological Nomenclature (Ride et al., 1999). Therefore, the specific name, “kenyensis” (masculine/feminine), originated by Pickford (2007a) should be corrected to “kenyense” (neuter). Orliac et al. (2010) wrote the specific name of this species as “rusingense” and “rusingensis” by mistake.
Comments.—The genus has been only represented by the upper dentition until now (Pickford, 2007a). It was firstly attributed to the Anthracotheriidae (Cetartiodactyla) by Pickford (2007a). Later, it was assigned to the Hippopotamidae and tentatively to the Kenyapotaminae by Orliac et al. (2010) and Boisserie et al. (2011). More recently, Pickford (2011) has stated that Kulutherium is an anthracotheriid and provisionally included the genus in the new subfamily Morotochoerinae (Anthracotheriidae) along with Morotochoerus.
cf.
Kulutherium
sp.
Figures 1, 2A
Material.—KNM-MW 13143, a trigonid of a left lower molar.
Repository.—Palaeontology Section, National Museums of Kenya, Nairobi, Kenya.
Locality.—Mfwangano (= Mfangano) Island (approx. 0°28′S, 34°01′E) in Lake Victoria, southwestern Kenya (Drake et al., 1988; Peppe et al., 2009).
Horizon and age.—Rusinga Group (Formation is unknown) (Figure 3; Drake et al., 1988; Peppe et al., 2009); ca. 17.9 Ma according to Drake et al. (1988) and Werdelin (2010), but a recent analysis (Peppe et al., 2011) indicates ca. 20–18 Ma; Burdigalian (late early Miocene).
Measurement.—Trigonid width = 36.3 mm.
Characteristics.—The lower molar trigonid is bunodont and brachyodont with a buccolingually bifurcate mesial root and no paraconid. Its morphology is comparable to that of the kenyapotamines, Kenyapotamus and Palaeopotamus. However, it is much larger than that of the kenyapotamines and matches the upper molar of Kulutherium in size. The postprotocristid does not connect to the postmetacristid.
Description
KNM-MW 13143 (Figures 1, 2A) is a trigonid of a left lower molar, which is roughly comparable in size to that of living hippopotamuses. The crown is brachyodont and the cusps are bunodont. The root for the trigonid (= mesial root) is buccolingually bifurcate (Figure 1E–G). The protoconid and metaconid are nearly equal in size. There is no paraconid. The preprotocristid extends mesiolingually to a small bulge at the notch of the mesial trigonid wall. The premetacristid is single-ridged and extends mesiobuccally to the small bulge. The preprotocristid and premetacristid are continuous with each other with an interruption by the small bulge at the notch of the mesial trigonid wall, closing the central mesiodistal valley between the protoconid and the metaconid mesially. The endoprotocristid extends lingually. The endometacristid extends buccally and reaches the lingual face of the protoconid between the endoprotocristid and the postprotocristid (Figure 1A, H). The postprotocristid extends distolingually and then turns lingually, disappearing below the postmetacristid at the central mesiodistal valley between the protoconid and the metaconid (Figure 1A, H). In distal view (Figure 1G), the postprotocristid and endometacristid seems to make a notch at the distal trigonid wall. This notch does not interrupt the central mesiodistal valley between the protoconid and the metaconid. The postmetacristid extends distobuccally, connecting to the talonid at the median transverse valley between the trigonid and the talonid. The postectoprotocristid is weak and extends distobuccally. The postectometacristid is much stronger than the postectoprotocristid and extends distolingually. In the distal view (Figure 1G), the postectometacristid, postmetacristid, postprotocristid, and postectoprotocristid form an incomplete M-like structure. There is an ectostylid at the base of the hypoflexid. There is a precingulid. There are no buccal or lingual cingulids. The enamel is not thick proportionally to the tooth size (Alloing-Séguier et al., 2014); the maximum thickness is about 1.3 mm, judging from the natural breakage of the cross section (Figure 1G).
Comparisons and discussion
KNM-MW 13143 is assigned to the primitive hippopotamids and likely to the Kenyapotaminae on the basis of its trigonid morphology. It is comparable in morphology to the molar trigonid of the kenyapotamines in having a brachyodont crown, bunodont cusps, an M-like structure on the distal trigonid wall in distal view, a single-ridged premetacristid, and a buccolingually bifurcate mesial root, in lacking a paraconid, and in that the central mesiodistal valley between the protoconid and the metaconid is not closed at the distal trigonid wall by the cristids (Figures 1, 2A, C–E). It is distinguished from the derived hippopotamids, that is, the hippopotamines such as Archaeopotamus and Hexaprotodon, in having much thinner enamel, a much lower crown, and weaker cristids. It generally differs from the suoids in having more trenchant cusps and in lacking minor grooves (= Fürchen) that characterize the Suoidea.
KNM-MW 13143 shows a morphological similarity with the lower molar trigonid of primitive bunodont anthracotheriids such as Myaingtherium and Anthracotherium (Figure 2A, G). However, the former is distinguished from the latter in that the mesial root (a root for the trigonid) is bifurcate, that the central mesiodistal valley between the protoconid and the metaconid is not closed at the distal trigonid wall by the cristids, and that the postprotocristid does not connect to the postmetacristid at the distal trigonid notch (Figures 1, 2A). In the molar trigonid of primitive bunodont anthracotheriids, the mesial root is single and not buccolingually bifurcated, the central mesiodistal valley is closed at the distal trigonid wall by the cristids, and the postprotocristid connects to the postmetacristid at the distal trigonid notch (Figure 2G; Tsubamoto et al., 2002, 2011).
KNM-MW 13143 appears to be assignable to Kulutherium (cf. Kenyapotaminae). The Kenyapotaminae consist of two genera, Kenyapotamus and Palaeopotamus (Pickford, 2007b; contra Boisserie et al., 2010, 2011, and Orliac et al., 2010), with two other genera, Kulutherium and Morotochoerus, tentatively included in the subfamily (Orliac et al., 2010; Boisserie et al., 2011; contra Pickford, 2011). KNM-MW 13143 is morphologically comparable to the molar trigonid of Morotochoerus and is distinguished from that of Kenyapotamus and Palaeopotamus in that the postprotocristid does not connect to the postmetacristid but disappears at the central mesiodistal valley between the protoconid and the metaconid (Figure 2A, C, D, F; Pickford, 1983, 2007b; Boisserie et al., 2010; Tsubamoto et al., 2015). This characteristic is rather comparable to the molar trigonid of the hippopotamines. An m3 specimen (specimen no. OCO 3′08) that was discovered in the Member D or E (upper middle Miocene) of the Ngorora Formation (Tugen Hills, Kenya) and was assigned to Kenyapotamus by Pickford (2011) has this characteristic. Judging from the figures by Pickford (2011, p. 534, fig. 11A), however, this specimen (OCO 3′08) differs from m3 of Kenyapotamus in having this characteristic, thicker enamel, and stronger cristids. Therefore, it may be better assigned to a new genus of the Hippopotamidae that is more derived than Kenyapotamus. In the molar trigonid of Kenyapotamus and Palaeopotamus, the postprotocristid connects to the postmetacristid, making a notch at the distal trigonid wall (Figure 2C, D). KNM-MW 13143 differs from the molar trigonid of Morotochoerus in that the preprotocristid is continuous with the premetacristid (Figure 2A). In the molar trigonid of Morotochoerus, the two cristids are not continuous (Figure 2F). In dental size, on the other hand, KNM-MW 13143 matches the upper molars of Kulutherium (Figure 2A, B), which has so far been known only from the upper dentition (Pickford, 2007a). It is much larger in molar trigonid size than Kenyapotamus, Palaeopotamus, and Morotochoerus (Figure 2A, C, D, F). Also, it morphologically matches the upper molars of Kulutherium at least in having a low crown and bunodont cusps (Figure 2A, B).
The identification of KNM-MW 13143 as Kulutherium is also supported indirectly by the geographical and geological evidence. Kulutherium was discovered in the Kulu Formation of the Rusinga Group at Rusinga Island in Lake Victoria, southeastern Kenya (Pickford, 2007a). The Kulu Formation is the topmost formation of the Rusinga Group in Rusinga Island and its age is estimated to be ca. 17–15.5 Ma (Figure 3; Peppe et al., 2009). Mfwangano Island, which is the locality of KNM-MW 13143, is close to Rusinga Island. The former is approximately 10 km southwest from the latter in Lake Victoria. Although the precise horizon of KNM-MW 13143 within the Rusinga Group is unclear, the Rusinga Group in Mfwangano Island is stratigraphically lower than the Kulu Formation of Rusinga Island (Figure 3; Drake et al., 1988) and has been recently estimated to be ca. 20– 18 Ma (Peppe et al., 2011). The age of KNM-MW 13143 is slightly older than that of K. kenyense, but the former is not very different from the latter. Therefore, KNM-MW 13143 and K. kenyense are close in geographical and stratigraphical distances and are close also in geological age.
In conclusion, although the evidence is still poor, the present specimen provides additional evidence that a hippopotamus-sized, large hippopotamid was already living during the late early Miocene in East Africa. If the present specimen belongs to Kulutherium, it provides additional evidence that Kulutherium should be assigned not to the Anthracotheriidae but to the Kenyapotaminae (Hippopotamidae).
Acknowledgments
We are grateful to the Government of Kenya for research permission at the National Museums of Kenya (Nairobi, Kenya). We also thank Emma Mbua, Fredrick Kyalo Manthi, Mary Muungu, and Francis Ndiritu (National Museums of Kenya) for graciously providing access to the specimens examined. This research was supported by JSPS KAKENHI Grant Number 25257408 (to M. Nakatsukasa).
