J. Geol. Soc. Japan, Vol. 126, No. 4, p. 167–181, April 2020 doi: 10.5575/geosoc.2020.0002 A new cane rat (Rodentia, Thryonomyidae) from the Upper Miocene Nakali Formation, northern Kenya Abstract 1, 2 2 Yoshiki Tanabe , Mayu Onodera , Masato Nakatsukasa3, Yutaka Kunimatsu4 and Hideo Nakaya2 Received June 1, 2019 Accepted January 22, 2020 1 2 3 4 Tottori Prefectural Museum, 2-124 Higashimachi, Tottori-shi, Tottori 680-0011, Japan Faculty of Science, Kagoshima University, 1-21-35 Korimoto, Kagoshima 890-0065, Japan Graduate School of Science, Kyoto University, Kitashirakawa Oiwake-cho, Sakyo-ku, Kyoto 606-8502, Japan Faculty of Business Administration, Ryukoku University, 67 Tsukamoto-cho, Fukakusa Fushimi-ku, Kyoto 612-8577, Japan Cane rats are represented by two extant species inhabiting SubSaharan Africa. Previous studies have established that the earliest occurrence of the extant genus Thryonomys was the latest Miocene in eastern Africa. A Japan–Kenya Expedition team has recovered a new species of Thryonomys, T. kamulai, sp. nov., from the base of the upper Miocene Nakali Formation (~10 Ma) of northern Kenya. We describe the new species on the basis of dental specimens. The new species is characterized by features such as small dentition size, shallower and wider sinuses on the upper cheek teeth, and a well-developed crescent-shaped metalophulid I on dp4. The present findings extend the first-appearance datum of the genus to the early late Miocene (~10 Ma), suggesting that the Thryonomys lineage inhabited Africa prior to ~10 Ma. Key words: Africa, Late Miocene, Rodent, Fossil, New species Corresponding author: Y. Tanabe, tanabey@pref.tottori.lg.jp nabe et al. (2013) reported additional specimens of Thryonomys sp. from the Nakali Formation. We re-examined the specimens and determined that they belong to a new Thryonomys species. Here we describe the new species and discuss the temporal range of the genus Thryonomys. Introduction The extant cane rats (Rodentia, Thryonomyidae) are represented by two species, Thryonomys swinderianus (greater cane rat) and T. gregorianus (lesser cane rat), in Sub-Saharan Africa (Woods and Kilpatrick, 2005; Monadjem et al., 2015; López-Antoñanzas, 2016). Thryonomyids were widely distributed over continental AfroEurasia during the Miocene but their distribution has been limited to Sub-Saharan since the Pliocene (Winkler, 1994; Flynn and Winkler, 1994; Winkler et al., 2010). In previous studies, it has been considered that the extant genus Thryonomys appeared during the Late Miocene (Winkler et al., 2010; Kraatz et al., 2013), based on the oldest fossil from Lemudong o, Kenya (Thryonomys sp., 〜6 Ma; Manthi, 2007). The Japan-Kenya Expedition team has conducted paleontological field-work in Nakali since 2002 and has remarkably expanded Late Miocene (〜10 Ma) mammalian fossil collections (Kunimatsu et al., 2007; Nakatsukasa et al., 2010; Handa et al. 2015, 2017a, b; Tsubamoto et al., 2015, 2017, 2020). Of the current 〜 3,500 specimens, approximately 45% are fossil rodents (Tsubamoto et al., 2020). Onodera et al. (2011) and Ta©The Geological Society of Japan 2020 Geologic setting The Nakali Formation is distributed in the Nakali area along the eastern shoulder of the central Kenyan Rift (Fig. 1). The Nakali Formation is unconformably covered by the Nasorut Formation and characterized by lacustrine, fluvio-lacustrine, and pyroclastic flow deposits (Kunimatsu et al., 2007; Sakai et al., 2013). The majority of fossils have been unearthed from the volcanic mud flow deposits of the Upper Member, e.g., at site NA39 (1°12′6.6″N, 36°22′23.2″E), which yielded Nakalipithecus, and site NA60 (1°11′57″N, 36°23′14.9″E), where the bulk of rodent fossils have been recovered. The 40 Ar/39Ar ages of the uppermost part of the Lower Member are 9.82 ± 0.09 and 9.90 ± 0.09 Ma (Kunimatsu et al., 2007). The paleomagnetic stratigraphy of the uppermost part of the Lower Member and the lowermost part of the Upper Member are correlated with Chron C5n. 1r 167 168 2020―4 Yoshiki Tanabe et al. Fig. 1. Sketch map of eastern-central Africa, showing the location of Nakali (modified from Kunimatsu et al., 2007). (9.88–9.92 Ma) (Kunimatsu et al., 2007). Materials and methods The Nakali Thryonomys fossils are housed at the National Museums of Kenya, Nairobi. A total of 17 specimens were collected from the Nakali Formation from 2003 to 2012. Most were recovered by sieving using a 1.0 × 1.0 mm mesh screen at locality NA60. The taxonomic determination was completed based on cheek tooth morphology because cheek teeth are relatively well preserved and morphologically diversified in rodent taxa. The thryonomyid dental terminology and measurements used here are based on Winkler (1992), Marivaux et al. (2002), López-Antoñanzas et al. (2004), López-Antoñanzas and Sen (2005) and Kraatz et al. (2013). Occlusal measurements were taken using photogrammetry. Abbreviations −maxi. maxilla; mandi. mandible; frag. fragment; I incisor; DP/dp upper/lower deciduous premolar; M/m upper/lower molar; w/ with; KNM the National Museums of Kenya, Nairobi, Kenya; NA Nakali; SH Samburu Hills; OM specimens of the Osteology section in the KNM; BM the Natural History Museum, London, England (formerly the British Museum [Natural History]); EP specimens collected by the Eyasi Plateau Expedition, Tanzania; F., L. and P. specimens collected by the Omo Expedition, Ethiopia (housed in the National Museum of Ethiopia); LAET specimens from Laetoli collected by Mary Leakey; ALA-VP-2 the Alayla Vertebrate Locality 2 (housed in the National Museum of Ethiopia); and NHM Natural History Museum, London, England. Systematics Order Rodentia Bowdich, 1821 Suborder Hystricomorpha Brandt, 1855 Infraorder Hystricognathi Brandt, 1855 Fig. 2. Thryonomyid dental terminology and measurements used for upper (A, KNM-NA 50198) and lower (B, KNM-NA 54957) teeth, based on Winkler (1992), Marivaux et al. (2002), López-Antoñanzas et al. (2004), López-Antoñanzas and Sen (2005), and Kraatz et al. (2013). Anterior is to the left. Al Anteroloph, Prl Protoloph, Pol Posteroloph, Mu Mure, Pr Protocone, Pa Paracone, Hy Hypocone, Mt Metecone, Si Sinus, ASi Anterosinus, Psi Posterosinus, Mld I Metalophulid I, Mld II Metalophulid II, Hyld Hypolophid, Pold Posterolophid, Ecd Ectolophid, Prd Protoconid, Mtd Metaconid, Hyd Hypoconid, End Entoconid, Sid Sinusid, ASid Anterosinusid, MSid Mesosinusid, PSid Posterosinusid, A Anterior, L Labial. J. Geol. Soc. Japan 126（ 4 ） A new Thryonomys from the Nakali Formation Family Thryonomyidae Pocock, 1922 Genus Thryonomys Fitzinger, 1867 Thryonomys kamulai, sp. nov. (Figs. 3–6) Paraphiomys sp., Kawamura and Nakaya, 1984. p.133– 139. Etymology.−In commemoration of the late Jackson Kamula, who was a veteran field assistant at Nakali and other fossil localities in Kenya. Holotype.−KNM-NA54957, Rt. mand. frag. w/ dp4– m1 (Figs. 5f, 6f) Paratypes.−Lt. maxi. frag. w/ DP4–M2 (KNM-NA 50312), Lt. maxi. frag. w/ DP4-M1 (Field No. 2668 11), Rt. maxi. frag. w/ DP4–M1 (KNM-NA 52402), Lt. maxi frag. w/ M2–3 (KNM-NA 52431), Lt. mand. frag. w/ I1(broken)–dp4–m2 (KNM-NA 50313), Rt. mand. frag. w/ Rt. m1–3 (KNM-NA 50311), Rt. mand. frag. w/ dP4–m1 (KNM-NA 46295), Lt. mand. frag. w/ m1–2 (KNM-NA 52480), Lt. isolated M1 or M2 (KNM-NA 50198, 52657), Rt. isolated M1 or M2 (KNM-NA 50199, 52410, NA60-2646 11), Lt. isolated m1 or m2 (KNM-NA 50197), Rt. isolated m1 or m2 (NA60-2778 11, NA60-2815 11) Referred material. −Lt. mand. frag w/ dp4–m2 (KNM-SH 10524 in Kawamura and Nakaya, 1984) Locality and horizon.−Nakali Formation in Nakali and Namurungule Formation in the Samburu Hills, northern Kenya; both early late Miocene (〜10 Ma) in 169 age. Diagnosis.−Smaller than the other species of the genus. Upper cheek teeth lacking mesoloph and metaloph and with shallower and wider sinuses. Lower cheek teeth lacking anterolabial cuspid. dp4 longer than that of the lower molars. dp4 with four lophids (metalophulid I, metalophulid II, hypolophid, and posterolophid) and well developed crescent-shaped metalophulid I with spur on the posterior side. Anterosinusid on dp4 closed lingually closed. m1–3 with relatively straight meta lophulid I and hypolophid. Differential diagnosis.−Thryonomys kamulai differs from any known species of Miocene thryonomyids such as Paraphiomys, Paraulacodus, and Protohummus in having the upper molars without mesoloph, lower molars lacking the anterolabial cuspid, and the presence of four transverse lophids on dp4. T. kamulai is different from the two extant thryonomyids, T. swinderianus and T. gregorianus, in being smaller, in having lower hypsodonty, less transversely expanded M1–2, and in the presence of a well developed metalophulid I on dp4. T. kamulai is distinct from T. asakomae in being its smaller, in having shallower and wider sinuses on the upper cheek teeth, relatively oblique anteroloph, in lacking the metaloph on M1–2, in having a well developed crescent-shaped metalophulid I with a spur and a closed anterosinusid on dp4, a slightly straight hypolophid and an inclined posterolophid on the lower molars. T. kamulai differs from T. wesselmani in being smaller, shallower and wider sinuses on the upper cheek teeth with straight anteroloph, in lacking a metaloph on M1–2, and in hav- Table 1. Measurements (in millimeters) of cheek teeth of Thryonomys kamulai, sp. nov., collected from the Nakali Formation. * = where tooth identification is uncertain, the measurement is repeated, in parentheses. 170 Yoshiki Tanabe et al. 2020―4 Fig. 3. Schematic drawings of the upper cheek teeth occlusal surfaces of Thryonomys kamulai, sp. nov., from the Nakali Formation. a. KNM-NA 50198, Lt. M1 or M2. b. KNM-NA 50199, Rt. M1 or M2. c. KNM-NA 50312, Lt. maxilla fragments with DP4–M2. d. KNM-NA 52402, Rt. maxilla with DP4–M1. e. KNM-NA 52410, Rt. M1 or M2. f. KNM-NA 52431, Lt. maxilla fragment with M2–3. g. KNM-NA 52657, Lt. M1 or M2. h. NA60-2646 11, Rt. M1 or M2. i. NA602668 11, Lt. DP4–M1. The scale bar is 1 mm; anterior (A) is denoted by the arrow. ing well developed crescent-shaped metalophulid I and closed anterosinusid on dp4. Measurements.−See Table 1. Description.−Tooth terminology is shown in Fig. 2A (upper molars) and Fig. 2B (lower molars). Upper dentition (See Figs. 3–4) DP4 is the smallest tooth in the upper cheek tooth row. The outline is rectangular with round corners, with the anterior margin shorter than the posterior margin. Three transverse lophs (anteroloph, protoloph, and posteroloph) are developed and buccally run from the entoloph. The anteroloph obliquely runs toward the molar anterior margin. The tip of the anteroloph becomes sharp. The anteroloph is narrower than the other two lophs. The protoloph is nearly straight, reaching the buccal margin of the tooth. The metaloph is lacking. There is no mesoloph, which projects buccally from an- terior arm of the hypocone. The protoloph is slightly nearer the anteroloph than the posteroloph. The posteroloph is oblique along the posterior margin and it meets the posterior arm of the hypocone. The posteroloph is the thickest of the three lophs, swelling midway. There is no metaloph, which bucco-anteriorly projects from the midpoint of the posteroloph. There are four main cusps (protocone, paracone, hypocone and metacone). The crown is lower hypsodonty. The protocone and hypocone are connected to one another by the mure on the lingual side. These cusps are large. The buccal ends of the protoloph and posteroloph meet the paracone and metacone, respectively. The paracone and metacone are smaller than the protocone and hypocone. The sinuses are shallow and wide, and the anterosinus is smaller than the posterosinus. M1–2 resemble the DP4 in the morphology of the oc- J. Geol. Soc. Japan 126（ 4 ） A new Thryonomys from the Nakali Formation 171 Fig. 4. Occlusal view of the upper cheek teeth of Thryonomys kamulai, sp. nov., from the Nakali Formation. a. KNM-NA 50198, Lt. M1 or M2. b. KNM-NA 50199, Rt. M1 or M2. c. KNM-NA 50312, Lt. maxilla fragments with DP4–M2. d. KNM-NA 52402, Rt. maxilla with DP4–M1. e. KNM-NA 52410, Rt. M1 or M2. f. KNM-NA 52431, Lt. maxilla fragment with M2–3. g. KNM-NA 52657, Lt. M1 or M2. h. NA60-2646 11, Rt. M1 or M2. i. NA60-2668 11, Lt. DP4–M1. The scale bar is 1 mm; anterior is to the left. clusal surface but they are much wider. The occlusal outline is rectangular with rounded corners. M2 is wider than M1. The three transverse lophs are relatively more transversely directed and longer than those of DP4. In early wear (KNM-NA50199, Figs. 3b, 4b; KNMNA52410, Figs. 3e, 4e; KNM-NA52657, Figs. 3g, 4g), the wear facet of the posterior arm of the protocone is absent or weak, but after moderate wear (the other specimens) the wear facet is developed. Although the occlusal pattern of the M3 resembles that of the other upper molars, the occlusal outline is triangular. The posteroloph obliquely runs toward the posterior margin of the tooth, and the crown posteriorly narrows. Lower dentition (See Figs. 5–6) dp4 is rectangular in occlusal outline, anteriorly narrowing. The crown is lower hypsodonty. dp4 has four distinct transverse lophids (metalophulid I, metalophulid II, hypolophid, and posterolophid). The crescentic metalophulid I has a posterior spur and obliquely runs toward the anterior margin from the metaconid. The metalophulid II lingually runs straight from the protoconid. The hypolophid is straight, reaching the lingual margin of the tooth. The posterolophid obliquely runs toward the posterior margin of the tooth. The protoconid is at the junction of the labial end of the metalophulid I 172 Yoshiki Tanabe et al. 2020―4 Fig. 5. Schematic drawings of the lower molar occlusal surfaces of Thryonomys kamulai, sp. nov., from the Nakali Formation. a. KNM-NA 46295, Rt. mandible with dp4–m1. b. KNM-NA 50197, Lt. m1 or m2. c. KNM-NA 50311, Rt. mandible with m1–3. d. KNM-NA 50313, Lt. mandible with dp4–m2. e. KNM-NA 52480, Lt. mandible with m1–2. f. KNM-NA 54957, Rt. mandible with dp4–m1. g. NA60-2778 11, Rt. m1 or m2. h. NA60-2815 11, Rt. m1 or m2. The scale bar is 1 mm; anterior (A) is denoted by the arrow. and II, and the metaconid is at the junction of the lingual ends of these lophids. The hypoconid is at the labial end of the posterolophid, and the entoconid is at the lingual end of the hypolophid. The lingual cuspids are larger than the labial cuspids. The sinuses are shallow and the anterosinusid is lingually closed with metalophulid I and II, while the mesosinusid and posterosinusid lingually open. The occlusal pattern of the lower molars is similar to that of the upper molars. m1–2 generally resemble dp4, but they are rather square in outline and have only three lophids (metalophulid I, hypolophid, and posterolophid) without metalophulid II. m1 is longer than m2. The mesosinusid of m1–2 is slightly wider than the posterosinusid. Metalophulid I runs straight along the anterior margin of the tooth. The hypolophid runs straight toward the lingual side and is slightly postero-lingually directed. The posterolophid obliquely runs toward the posterior margin of the crown. m3 differs from the m1–2 in being posteriorly reduced. The posterolophid runs obliquely toward the posterior margin of the tooth and the lingual end of the J. Geol. Soc. Japan 126（ 4 ） A new Thryonomys from the Nakali Formation 173 Fig. 6. Occlusal view of the lower cheek teeth of Thryonomys kamulai, sp. nov., from the Nakali Formation. a. KNM-NA 46295, Rt. mandible with dp4–m1. b. KNM-NA 50197, Lt. m1 or m2. c. KNM-NA 50311, Rt. mandible with m1–3. d. KNM-NA 50313, Lt. mandible with dp4–m2. e. KNM-NA 52480, Lt. mandible with m1–2. f. KNM-NA 54957, Rt. mandible with dp4–m1. g. NA60-2778 11, Rt. m1 or m2. h. NA60-2815 11, Rt. m1 or m2. The scale bar is 1 mm; anterior is to the left. lophid strongly antero-lingually inclines. Remarks.−The Nakali specimens are assigned to the Thryonomyidae because they have well developed lophs/lophids (Winkler et al., 2010), upper cheek teeth with two labial and one lingual enamel infoldings (=sinus, anterosinus, posterosinus), and lower cheek teeth with the opposite pattern (=sinusid, mesosinusid, posterosinusid) except for dp4 which has an extra lingual infolding (=anterosinusid) (Woods, 1984; Ungar, 2010). Among other genera of Thryonomyidae, they closely resemble Paraphiomys, Paraulacodus, Protohummus, and Thryonomys. Paraphiomys differs from the Nakali 174 2020―4 Yoshiki Tanabe et al. specimens in having the mesoloph on M1–2, in the absence of the metalophulid II on dp4 and in the presence of the anterolabial cuspid on the lower molars (LópezAntoñanzas, 2004; López-Antoñanzas and Sen, 2005; Kraatz et al., 2013). Paraulacodus differs from the Nakali specimens in the absence of the metalophulid II on dp4 and presence of the anterolabial cuspid on the lower molars (López-Antoñanzas et al., 2004; López-Antoñanzas and Sen, 2005; Kraatz et al., 2013). Protohummus differs from the Nakali specimens in the absence of the metaloph on the M1–2 and the metalophulid II on dp4 (Kraatz et al., 2013). The Nakali specimens most closely resemble the genus Thryonomys in having upper molars without a mesoloph, lower molars lacking the anterolabial cuspid, and the presence of four transverse lophs on dp4 (Winkler et al., 2010; Kraatz et al., 2013). Thryonomys is characterized by the presence of two or three-grooved upper incisor (Winkler et al., 2010; Kraatz et al., 2013); however, we have not found a grooved upper incisor in T. kamulai from the Nakali Formation. The genus Thryonomys contains four species as previously named. Two of them (T. swinderianus and T. gregorianus) are extant (Musser and Carleton, 2005; Monadjem et al., 2015; López-Antoñanzas, 2016), and two others are extinct: T. asakomae from the Adu-Asa Formation, Middle Awash, Ethiopia (5.7–5.6 Ma; Wesselman et al., 2009) and T. wesselmani from the Upper Ndolanya Beds of Laetoli, Tanzania (2.66 Ma; Denys, 1987, 2011). The only two extant species are similar in terms of molar morphology; however, T. gregorianus generally has a smaller dentition than that of T. swinderianus (Figs. 7, 8 and Tables 2, 3). In comparison with these species, the Nakali specimens are smaller yet (Figs. 7, 8 and Tables 2, 3), with lower hypsodonty and a rather antero-posteriorly short M3. Both extant species have a straighter metalophulid I on the dp4 than that of T. kamulai, and the outline of the occlusal surface of m3 is not reduced. Thryonomys asakomae is intermediate in dental size. It is smaller than the extant species but is larger than the Nakali specimens (Figs. 7, 8 and Tables 2, 3). The sinuses on the upper cheek teeth of T. asakomae are deeper and narrower than those of T. kamulai. M1–2 have a metaloph on the midpoint of the posteroloph. dp4 has a spur on metalophulid I, which is not well developed and crescent-shaped, and an open anterosinusid. The lower molars have a slightly inclined hypolophid and a relatively straight posterolophid. These characteristics are not found in the present specimens. T. asakomae has an upper incisor with two grooves in the labial enamel surface (Wesselman et al., 2009). This characteristic is also seen in the Miocene thryonomyids Paraulacodus and Protohummus and is considered a primitive characteristic (Kraatz et al., 2013). T. asakomae is considered morphologically intermediate between Paraulacodus or Protohummus (having an upper incisor with two grooves) and the other Thryonomys (having three grooves) (Kraatz et al., 2013). However, as no upper incisor of Thryonomys is known from the Nakali Formation, comparisons are not possible. Thryonomys wesselmani is large in dental size, as large as T. gregorianus and larger than the Nakali specimens (Figs. 7, 8 and Tables 2, 3). The sinuses on the upper cheek teeth of T. wesselmani are deeper and narrower than those of T. kamulai. The buccal end of the anteroloph obliquely runs toward the posterior on the upper cheek teeth, and M1–2 have a metaloph on the midpoint of the posteroloph. The occlusal surface of the lower molars of T. wesselmani resembles that of T. kamulai; however dp4 has the metalophulid I, which is neither well developed nor well developed and crescentshaped, and an open anterosinusid. Kawamura and Nakaya (1984) described a lower tooth row (KNM-SH 10524, left mandibular fragment with dp4 [broken]–m2) from the Namurungule Formation of the Samburu Hills in Kenya ( 〜9.5 Ma) as Paraphiomys sp. An important characteristic of the specimen is the absence of the anterolabial cuspid (as anteroconid in Kawamura and Nakaya, 1984). Furthermore, it has three transverse lophids and the metalophulid I ( metalophid in Kawamura and Nakaya, 1984) and hypolophid run straight toward the lingual side and are slightly postero-lingually directed. The posterolophid extends from the hypoconid along the posterior margin of the crown. The specimen is slightly larger than the Nakali specimens. However, based on morphological similarities, we consider that the Samburu Hills specimen (KNM-SH 10524) is a T. kamulai. Discussion According to previous studies, Thryonomys fossil records are reported from the uppermost Miocene of SubSaharan Africa. Fossils assigned to Thryonomys sp. (Manthi, 2007) were collected from Lemudong o, Kenya (6.12–6.08 Ma; Deino and Ambrose, 2007); this material has been considered the oldest representative of the genus (Winkler et al., 2010). However, the present findings from Nakali change the first appearance datum of Thryonomys to 〜10 Ma. Some authors claim that Thryonomys was derived J. Geol. Soc. Japan 126（ 4 ） Table 2. Measurements (in millimeters) of upper cheek teeth of Thryonomys species. A new Thryonomys from the Nakali Formation 175 176 Table 3. Measurements (in millimeters) of lower cheek teeth of Thryonomys species. Yoshiki Tanabe et al. 2020―4 J. Geol. Soc. Japan 126（ 4 ） A new Thryonomys from the Nakali Formation 177 Fig. 7. Bivariate plot of width versus length of upper cheek teeth for Thryonomys species. Measurements taken from previous studies are shown as maximum ranges in rectangles. Measurements are listed in Tables 1 and 2. from a Paraulacodus-like ancestor based on cladistic analysis of thryonomyid dental morphology (Flynn and Winkler, 1994; López-Antoñanzas et al., 2004; LópezAntoñanzas and Sen, 2005). Paraulacodus has a twogrooved upper incisor, lower cheek teeth with an anterolabial cuspid, no metalophulid II on dp4, and was known two taxa Pa. johanesi (Middle–Late Miocene, East Africa; i.e., from 〜8.5 to 〜7.5 Ma, Chorora, Ethiopia; Geraads, 1998; Suwa et al., 2015) and Pa. indicus (12.9–12.5 Ma, Potwar Plateau, Pakistan; Hinton, 1933; Black, 1977; Flynn and Winkler, 1994). Kraatz et al. (2013) reported a new genus and species of thryonomyids, Protohummus dango, from the upper Miocene (7.5– 6.5 Ma, Bibi et al., 2013) of the United Arab Emirates. Pr. dango has the two-grooved upper incisor, lower cheek teeth without the anterolabial cuspid, and no metalophulid II on dp4. They conducted a cladistic analysis of thryonomyids, including Pr. dango and recently de- 178 Yoshiki Tanabe et al. 2020―4 Fig. 8. Bivariate plot of width versus length of lower cheek teeth for Thryonomys species. Measurements taken from previous studies are shown as maximum ranges in rectangles. Measurements are listed in Tables 1 and 3. scribed thryonomyids (i.e., T. asakomae, Wesselman et al., 2009; T. wesselmani, Denys, 1987, 2011), and concluded that Pr. dango fills the phylogenetic and morphological gap between Paraulacodus (i.e., Pa. johanesi) and the fossil and extant species of Thryonomys (i.e., T. asakomae). Thus, Kraatz et al. (2013) claimed that the age of divergence between Pr. dango and T. asakomae predates 〜5.6 Ma, which is the age of T. asakomae. Ac- J. Geol. Soc. Japan 126（ 4 ） A new Thryonomys from the Nakali Formation tually, in previous studies, the oldest known fossil of Thryonomys was from Lemdong o, Kenya (Manthi, 2007), and the branch age of Pr. dango and Thryonomys would have been 〜6 Ma (instead of 〜5.6 Ma). Kraatz et al. (2013) claimed that Pr. dango is probably more closely related to Pa. johanesi than Pa. indicus. It is known that some dispersal events of Thryonomyidae between Africa and southern Asia occurred during the Early–Middle Miocene (e.g., Winkler and Flynn, 1994). Pr. dango, being phylogenetically intermediate between Paraulacodus (i.e., Pa. johanesi) and Thryonomys, indicates a dispersal event from Africa to Arabia before 〜7 Ma, which was followed by a return of its lineage to Africa at 〜6 Ma, evolving into the genus Thryonomys (Kraatz et al., 2013). The fossil records of Thryonomys have been limited to Sub-Saharan Africa since the Late Miocene (Winkler et al., 2010); no fossils of this genus are known from the Arabian Peninsula. Based on the aforementioned, the presence of T. kamulai at Nakali suggests that the divergence of Protohummus from a Paraulacodus-like ancestor occurred at least by 10 Ma, and most likely in Africa, and that Protohummus and Thryonomys subsequently diverged. Fossil records of Protohummus are unknown from Africa (Kraatz et al., 2013); however, two thryonomyid specimens from the Namurungule Formation of the Samburu Hills in Kenya (〜9.5 Ma) described by Kawamura and Nakaya (1987) could belong to a similar taxon of the base of the crown Thryonomys near Pr. dango or T. asakomae (Kraatz et al., 2013), allowing a better understanding of the paleobiogeography. Thus, T. kamulai and thryonomyid specimens from the Namurungule Formation favor the possibility that Protohummus and Thryonomys had diverged in Africa. Cladistic analysis is necessary to strengthen this hypothesis; it is ongoing and will be reported in a separate article in the future. Conclusion A new species of thryonomyid rodent, Thryonomys kamukai, sp. nov., was reported from the Upper Miocene Nakali Formation (〜10 Ma), Kenya. A specimen from the Samburu Hills, Kenya (Kawamura and Nakaya, 1984), was also assigned to T. kamulai. T. kamulai is the oldest species of the genus Thryonomys. This study suggests that Thryonomys inhabited Africa prior to 〜10 Ma. Acknowledgments The authors wish to thank the Government of Kenya and the National Museums of Kenya for research permission. We are very grateful to L. J. Flynn (Harvard 179 University, Cambridge, USA), A. J. Winkler (Southern Methodist University, Texas, USA), R. L. López-Antoñanzas (Université de Montpellier-CNRS, Montpellier, France) and Y. Kimura (National Museum of Nature and Science, Tokyo) for helpful comments that im proved the manuscript. We wish to thank all members of the Japan-Kenya Expedition team. The research in Nakali was supported by many local people. This study was supported in part by Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research (25257408, 16H02757) awards to Masato Nakatsukasa. References Bibi, F., Hill, A., Beech, M. and Yasin, W., 2013, Late Miocene fossils from the Baynunah Formation, United Arab Emirates: summary of a decade of new work. In Wang, X., Flynn, L. J. and Fortelius, M., eds., Neogene Terrestrial Mammalian Biostratigraphy and Chronology in Asia, Columbia Univ. Press, New York, 583–594. Deino, A. L. and Ambrose, S. H., 2007, 40Ar/39Ar dating of the Lemudong o late Miocene fossil assemblages, southern Kenya Rift. Kirtlandia, 56, 65–71. Denys, C., 1987, Fossil rodents (other than Pedetidae) from Laetoli. In Leakey, M. D. and Harris, J. M., eds., Laetoli: A Pliocene Site in Northern Tanzania, Oxford Univ. Press, New York, 118–170. Denys, C., 2011, Rodents. In Harris, T., ed., Paleontology and Geology of Laetoli: Human Evolution in Context. Volume 2: Fossil Hominins and the Associated Fauna. Springer, New York, 15–53. Flynn, L. J. and Winkler, A. J., 1994, Dispersalist implications of Paraulacodus indicus: a South Asian rodent of African affinity. Hist. Biol., 9, 223–235. Geraads, D., 1998, Rongeurs du Miocène supérieur de Chorora (Ethiopie): Cricetidae, Rhizomyidae, Phiomyidae, Thryonomyidae, Sciuridae. Palaeovertebrata, 27, 203–216.* Handa, N., Nakatsukasa, M., Kunimatsu, Y. and Nakaya, H., 2017a, A new Elasmotheriini (Perissodactyla, Rhinocerotidae) from the upper Miocene of Samburu Hills and Nakali, northern Kenya. Geobios, 50, 197–209. Handa, N., Nakatsukasa, M., Kunimatsu, Y. and Nakaya, H., 2017b, Additional specimens of Diceros (Perissodactyla, Rhinocerotidae) from the Upper Miocene Nakali Formation in Nakali, central Kenya. Hist. Biol., 31, 1–12. Handa, N., Nakatsukasa, M., Kunimatsu, Y., Tsubamoto, T. and Nakaya, H., 2015, New specimens of Chilotheridium (Perissodactyla, Rhinocerotidae) from the Upper Miocene Namurungule and Nakali Formations, northern Kenya. Paleontolog. Res., 19, 181–194. Kawamura, Y. and Nakaya, H., 1984, Thryonomyid rodent from the Late Miocene Namurungule Formation, Samburu Hills, northern Kenya. Afr. Study Monogr. Suppl., 2. 133–139. Kawamura, Y. and Nakaya, H., 1987, Additional materials of the Late Miocene rodents from the Namurungule Formation in the Samburu Hills, northern Kenya. Afr. Study Monogr. Suppl., 5, 131–139. Kraatz, B. P., Bibi, F., Hill, A. and Beech, M., 2013, A new fossil thryonomyid from the Late Miocene of the United Arab Emirates and the origin of African cane rats. Naturwissenschaften, 100, 437–449. Kunimatsu, Y., Nakatsukasa, M., Sawada, Y., Sakai, T., Hyodo, M., Itaya, T.,...Mbua, E., 2007, A new Late Miocene great ape from Kenya and its implications for the origins of African great apes and humans. Proc. Natl. Acad. Sci. U.S.A., 180 Yoshiki Tanabe et al. 104, 19220–19225. López-Antoñanzas, R. L., 2016, The family Thryonomyidae. In Wilson, D. E. and Mittermeier, R. A., eds., Handbook of Mammals of the World, Vol. 6, Rodents and Lagomorphas, Lynx Edicions, Barcelona, 330–335. López-Antoñanzas, R. L. and Sen, S., 2005, New species of Paraphiomys (Rodentia, Thryonomyidae) from the lower Miocene of As-Sarrar, Saudi Arabia. Palaeontology, 48, 223–233. López-Antoñanzas, R. L., Sen, S. and Mein, P., 2004, Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zool. J. Linn. Soc., 142, 423–444. Manthi, F. K., 2007, A preliminary review of the rodent fauna from Lemudong o, southwestern Kenya, and its implication to the Late Miocene paleoenvironments. Kirtlandia, 56, 92– 105. Marivaux, L., Vianey-Liaud, M., Welcomme, J. L. and Jaeger, J. J., 2002, The role of Asia in the origin and diversification of hystricognathous rodents. Zool. Scr., 31, 225–239. Monadjem, A., Taylor, P. J., Denys, C. and Cotterill, F. P. D., 2015, Rodents of Sub-Saharan Africa; A Biogeographic and Taxonomic Synthesis. Walter de Gruyter GmbH, Germany, 1092p. Nakatsukasa, M., Mbua, E., Sawada, Y., Sakai, T., Nakaya, H., Yano, W. and Kunimatsu Y., 2010, Earliest colobine skeletons From Nakali, Kenya. Am. J. Phys. Anthropol., 143, 365–382. Onodera, M., Kunimatsu, Y., Nakatsukasa, M., Yamada, E. and Nakaya, H., 2011, Late Miocene rodent fossils from the Nakali Formation, Kenya and its paleoenvironment. 160th Regular Meet. Palaeontol. Soc. Japan, Abstr., 19.** Sakai, T., Saneyoshi, M., Sawada, Y., Nakatsukasa, M., Kunimatsu, Y. and Mbua, E., 2013, Early continental rift basin stratigraphy, depositional facies and tectonics in volcaniclastic system: examples from the Miocene successions along the Japan Sea and in the East African Rift Valley (Kenya). In Itoh, Y., ed., Mechanism of Sedimentary Basin Formation−Multidisciplinary Approach on Active Plate Margin, InTech, Rijeka, Croatia, 83–108. Suwa, G., Beyene, Y., Nakaya, H., Bernor. R. L., Boisserie, J. R., Bibi, F.,...Asfaw, B., 2015, Newly described cercopithecid, equid and other mammalian fossils from the Chorora Formation, Ethiopia. Anthropol. Sci., 123, 19–39. Tanabe, Y., Onodera, M. and Nakaya, H., 2014, The Late Miocene rodent fauna from the Nakali Formation, Northern Kenya. 121st Annu. Meet. Geol. Soc. Japan, Abstr., 284.** Tsubamoto, T., Kunimatsu, Y., Nakaya, H., Sakai, T., Saneyoshi, M., Mbua, E. and Nakatsukasa, M., 2015, New specimens of a primitive hippopotamus, Kenyapotamus coryndonae, from the Upper Miocene Nakali Formation, Kenya. J. Geol. Soc. Japan, 121, 153–159. Tsubamoto, T., Kunimatsu, Y., Sakai, T., Saneyoshi, M., Shimizu, D., Morimoto, N.,...Nakatsukasa, M., 2017, Listri- 2020―4 odontine suid and tragulid artiodactyls (Mammalia) from the upper Miocene Nakali Formation, Kenya. Paleontol. Res., 21, 347–357. Tsubamoto, T., Kunimatsu, Y., Sakai, T., Saneyoshi, M., Shimizu, D., Morimoto, N.,...Nakatsukasa, M., 2020, A new species of Nyanzachoerus (Mammalia, Artiodactyla, Suidae, Tetraconodontinae) from the upper Miocene Nakali Formation, Kenya. Paleontol. Res., 24, 41–63. Unger, P. S., 2010, Mammal Teeth. Origin, Evolution, and Diversity. Johns Hopkins Univ. Press, Baltimore, MD, 304p. Wesselman, H. B., 1984, The Omo micromammals: systematics and paleoecology of early man sites from Ethiopia. Contrib. Vertebr. Evol., 7, 1–219. Wesselman, H. B., Black, M. T. and Asnake, M., 2009, Small mammals. In Haile-Selassie, Y. and Woldegabriel, G., eds., Ardipithecus kadabba: Miocene Evidence from the Middle Awash, Ethiopia. Univ. California Press, Berkeley, 105– 133. Winkler, A. J., 1992, Systematics and biogeography of middle Miocene rodents from the Muruyur Beds, Baringo District, Kenya. J. Vertebr. Paleontol., 12, 236–249. Winkler, A. J., 1994, The middle/upper Miocene dispersal of major rodent groups between southern Asia and Africa. Natl. Sci. Mus. Monogr., 8, 173–184. Winkler, A. J., 2003, Rodents and lagomorphs from the Miocene and Pliocene of Lothagam, northern Kenya. In Leakey, M. G. and Harris, J. M., eds., Lothogam, the Dawn of Humanity in Eastern Africa, Columbia Univ. Press, New York, 169–198. Winkler, A. J., Denys, C. and Avery, D. M. 2010, 17. Rodentia. In Werdelin, L. and Sanders, W. J., eds., Cenozoic Mammals of Africa, Univ. California Press, Berkeley, Los Angeles, London, 263–304. Woods, C. A. 1984, Hystricognath Rodents. In Anderson, S. and Jones, J. K., eds., Order and Families of Recent Mammals of the World, John Wiley and Sons, New York, 389– 446. Woods, C. A. and Kilpatrick, C. W., 2005, Infraorder Hystricognathi. In Wilson, D. E. and Reeder, D. M., eds., Mammal Species of the World, a Taxonomic and Geographic Reference, 3rd ed., Johns Hopkins Univ. Press, Baltimore, MD, 1538–1600. * in French ** in Japanese （著者の貢献） 田邉佳紀 化石の同定記載，原稿執筆を担当． 小野寺麻由 化石の同定記載を担当． 中務真人 現地での発掘調査の総括，本研究の議論を担当． 國松 豊 現地での発掘調査，本研究の議論を担当． 仲谷英夫 現地での発掘調査，本研究の議論を担当． J. Geol. Soc. Japan 126（ 4 ） （要 A new Thryonomys from the Nakali Formation 181 旨） Tanabe, Y., Onodera, M. Nakatsukasa, M. Kunimatsu, Y. and Nakaya, H., 2020, A new cane rat (Rodentia, Thryonomyidae) from the Upper Miocene Nakali （田邉佳紀・小野寺麻由・ Formation, northern Kenya. J. Geol. Soc. Japan, 126, 167–181． 中務真人・國松 豊・仲谷英夫，2020，ケニア中央部上部中新統ナカリ層から産出した新 種ヨシネズミ類．地質雑，126, 167–181.） アフリカヨシネズミ （齧歯目，ヨシネズミ科）は二種の現生種がアフリカ・サハラ以南に 生息している．先行研究では，現生のヨシネズミ属はその化石記録からアフリカの後期中 新世末に出現したと考えられていた．日本−ケニア調査隊はケニア北部に分布する上部中 からヨシネズミ属の新種 Thryonomys kamulai, sp. nov. を発見し 新統ナカリ層 （約 10 Ma） た．筆者らは頬歯化石を基に本種の記載を行い，ヨシネズミ属の中では小型で，また固有 の稜縁歯 （lophodonty）を有することが特徴づけられた．この発見により，ヨシネズミ属の 初産出記録が後期中新世の約 10 Ma に更新され，彼らは少なくとも 10 Ma 以前にアフリ カに生息していたことが考えられる． 科学論文では，学説の検証可能性を保証することが重要です．そのため，地質学雑誌掲載論文には，重 要な証拠となった試料がどこで得られたかを示しているものがあります．言うまでもないことですが， 見学や採取を行う場合，各自の責任において地権者や関係官庁への連絡と許可の取得の必要があること にご注意下さい．詳しくは，以下のサイトをご覧ください． http://www.geosociety.jp/publication/content0073.html 一般社団法人日本地質学会