taxonID	type	description	language	source
D417CA60F57CFF996785E3329061FF6E.taxon	materials_examined	Type species: Thryonomys swinderianus (Temninck, 1827) Synonym: Aulacodus Temninck, 1827; Triaulacodus Lydekker, 1896; Choeromys Thomas, 1922	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F57DFF9864ECE1C49003FD82.taxon	materials_examined	Type species: Paraphiomys pigotti Andrews, 1914	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F57DFF9864E3E25B9169FA7C.taxon	description	Paraphiomys pigotti is clearly distinct from all other Thryonomyidae. It has the metalophulid II on the lower molars, contrary to the condition in Gaudeamus aegyptius, Neosciuromys africanus, Apodecter stromeri, Kochalia geespei, Paraulacodus johanesi, Paraulacodus indicus, Paraphiomys orangeus, P. renelavocati sp. nov., P. roessneri, P. australis, P. shipmani and P. afarensis. Furthermore, P. pigotti differs from P. occidentalis in having a metaloph on the upper molars and from P. hopwoodi, Paraphiomys sp. nov. from Saudi Arabia, and Epiphiomys coryndoni, in being much larger. Paraphiomys pigotti differs from Sacaresia moyaeponsi in having a mesoloph in the upper molars.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F57DFF9D6402E615958AFF6F.taxon	description	H B D E G I J K 45 – 46) distinguished this taxon from P. pigotti by the presence in the former of a vestigial mesolophid (metalophulid II). According to Wood (1968), the metalophulid II is very short and vestigial in both forms and therefore this taxon should be considered a junior synonym of P. pigotti. However, according to Lavocat (1973: 39, 45) the upper teeth are very different from those of P. pigotti because of the absence of a metaloph. We agree with Lavocat (1973) in considering P. occidentalis a valid taxon. Paraphiomys occidentalis is much larger than nearly all thryonomyids, except for Neosciuromys africanus, P. simonsi and P. pigotti. Paraphiomys occidentalis is distinct from Neosciuromys africanus and P. simonsi in having a metalophulid II on the lower molars.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F578FF9D67B3E4C49686FE81.taxon	description	The holotype of P. hopwoodi (Lavocat, 1973: 44) is a left mandibular fragment (KNM SO 627) with d 4 - m 2 and an erupting m 3 from the Lower Miocene of Songhor. There are many differences between P. hopwoodi and Gaudeamus aegyptius, Neosciuromys africanus, Apodecter stromeri, Kochalia geespei, Paraulacodus johanesi, Paraulacodus indicus, Paraphiomys renelavocati sp. nov., P. orangeus, P. simonsi, P. roessneri, P. australis, P. shipmani and P. afarensis. One of these is the presence of the metalophulid II on the lower molars of the former. Paraphiomys hopwoodi differs from Paraphiomys sp. nov. from Saudi Arabia in being larger, having the upper molars less antero-posteriorly compressed and in showing a less labially displaced hypocone. Paraphiomys hopwoodi is distinct from P. occidentalis and P. pigotti in being much smaller. It differs from Epiphiomys coryndoni, e. g. in having the metalophid II much less developed, and from Sacaresia moyaeponsi in having an anterolabial cuspid on the lower molars.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F578FF9C6440E58F9442FD30.taxon	description	In a preliminary study of the late Miocene micromammals from the Harasib deposits of Namibia, Senut et al. (1992: 730) mentioned cf. Apodecter. A subsequent detailed study of the material, which consists of 399 complete isolated teeth, has shown that it represents a new species that Mein, Pickford & Senut (2000) named P. roessneri (Fig. 3 D, E). The holotype is an upper left molar (GSW Ari 2). Paraphiomys roessneri differs most importantly from Kochalia geespei in the absence of dental replacement, in lacking a welldeveloped anteroconid, and in having a, generally complete, metaloph. Paraphiomys roessneri is different from P. shipmani in being larger and in lacking the prominent labial cuspid on the lower molars. Paraphiomys roessneri mainly differs from Apodecter stromeri in having the m 3 less reduced and the cheek teeth appreciably longer than wide (whereas in Apodecter stromeri they are relatively broad). Paraphiomys roessneri is distinct from P. renelavocati sp. nov. in having the anterolabial cingulum of the p 4 less developed than in the latter, in having higher crowned teeth, and in the absence of a mesoloph in the P 4 and M 1. Paraphiomys roessneri differs from P. orangeus in lacking mesoloph in the P 4 and M 1. Paraphiomys roessneri is distinct from P. hopwoodi and from Paraphiomys sp. nov. from Saudi Arabia mainly in lacking metalophulid II on the lower molars. One difference of P. roessneri from Neosciuromys africanus, P. occidentalis, P. pigotti and P. simonsi is its much smaller size.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F578FF9D6794E13B958AFB02.taxon	description	The holotype and single published specimen of this species (CGM 26908, Fig. 4 E) is a badly damaged lower jaw with d 4 - m 3 and the incisor. It comes from the Oligocene Jebel el Qatrani Formation at ‘ Yale Quarry I’, Fayum Province, Egypt (Wood, 1968). The casts of three unpublished specimens, a fragment of maxilla with P 4 - M 2 (CGM 80 - 17) and two mandibular fragments, one of them with d 4 - m 2 (CGM 40428) and the other with m 1 - m 2 (CGM 40478) are available in the Muséum National d’Histoire Naturelle (Paris). CGM 80 - 17 shows that the upper molars of this species have a three-lophed morphology. According to Patterson & Wood (1982: 520), this species should be included in the genus Neosciuromys because of the high hypsodonty, the simplification of the occlusal pattern, and the large size of the m 2 and m 3. However, according to the cladistic analysis conducted below, the high hypsodonty of Neosciuromys africanus and P. simonsi is homoplastic. Paraphiomys simonsi is much bigger than all other extinct species of thryonomyids, except Neosciuromys africanus, P. occidentalis and P. pigotti. It differs from P. occidentalis and P. pigotti in lacking the metalophulid II on the lower molars and from Neosciuromys africanus in the absence of the mesoloph in the upper molars.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F579FF9C67B9E6FD90F2FAF1.taxon	description	The morphology of the lower teeth of P. afarensis is close to that of P. shipmani. However, P. shipmani is smaller and its hypolophid is slightly more posteriorly directed. Extrinsic differences such as geographical distribution and age substantiate their distinctiveness, which remains unquestioned to date. One of the differences between P. afarensis and the two species of Paraulacodus rests in the ungrooved upper incisor of the former. The three-lophed lower molars of P. afarensis clearly differ from the four-lophed lower cheek teeth of P. pigotti, P. hopwoodi, Paraphiomys sp. nov. from Saudi Arabia, P. occidentalis, Epiphiomys coryndoni and Sacaresia moyaeponsi. According to Geraads (1998: 210), P. afarensis retains the mesoloph in the upper molars, whereas this structure is absent in the Harasib species P. roessneri and P. australis. Paraphiomys afarensis mainly differs from Apodecter stromeri and P. renelavocati sp. nov. in having the m 1 smaller than the m 2 and in being more lophodont. It is distinct from Neosciuromys africanus and P. simonsi in being smaller and less hypsodont. Paraphiomys afarensis is larger than the Arrisdrift species P. orangeus.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F579FF9C67FDE2E994ADF92C.taxon	description	In their preliminary study of the micromammals from the Upper Miocene Harasib deposits of Namibia, Senut et al. (1992) reported Ch’orora sp. 2., which Mein et al. (2000) named P. australis on the basis of 317 isolated teeth and two maxillae (Fig. 3 F, G). The holotype of this species (GSW Ari 13) is a right p 4 (Mein et al., 2000: fig. 1, no. 13). This species differs from P. roessneri in being larger, in having a better developed anterolabial cingulum on the p 4, and in having often accessory crest in the upper molars (Mein et al., 2000). This species clearly differs from P. pigotti, P. hopwoodi, Paraphiomys sp. nov. from Saudi Arabia, P. occidentalis, Epiphiomys coryndoni and Sacaresia moyaeponsi, notably in the absence of metalophulid II on the lower molars. Paraphiomys australis is distinct from Neosciuromys africanus and P. simonsi in being much smaller and lower-crowned. Paraphiomys australis is larger than P. orangeus, P. renelavocati sp. nov., P. shipmani Apodecter stromeri and Kochalia geespei, and lacks the anterior cusp on the lower molars (present in the five latter taxa). One of the differences between P. australis and the two species of Paraulacodus is its lack of the anterolabial cusp on lower molars. According to Mein et al. (2000: 379), P. chororensis (renamed as P. afarensis) and P. australis are quite close. However, the taxa show some small differences, e. g. P. australis has the m 2 smaller than the m 1, the m 3 shorter than the m 2 and it lacks the anterolabial cusp on the lower molars.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F576FF936413E5B19189F9B9.taxon	materials_examined	Type species: Neosciuromys africanus Stromer, 1922 Synonymy: Phthinylla Hopwood, 1929	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F576FF9367DAE1C49447F8E4.taxon	description	The material of this species, which is currently housed in the MNHN, comes from the Early Miocene locality of As-Sarrar, Dam Formation, Saudi Arabia. Thomas et al. (1982: 129) provisionally assigned it to Paraphiomys sp. They briefly described this taxon as pentalophodont with well-developed mesolophs and mesolophids. Flynn, Jacobs & Sen (1983) described and illustrated the right lower molar AS 8 - 1100, announcing on the basis of this the presence of a plesiomorphic thryonomyid in the As-Sarrar fauna. Later, Winkler (1992: 244) considered this taxon to be possibly new, at the species or even genus level. Finally, a detailed study of this thryonomyid material (López-Antoñanzas & Sen, in press) has proved this species to be a new representative of the genus Paraphiomys. It will be referred to in the present work as Paraphiomys sp. nov. from Saudi Arabia (Fig. 3 J, K). Its holotype, AS 8 - 1100, is a right m 1 - m 2 (López-Antoñanzas & Sen, in press). The Paraphiomys sp. nov. from Saudi Arabia has been also possibly recorded in the coeval Nari Formation in Pakistan (as ‘ Kochalia sp. ’ by de Bruijn & Hussain, 1985). In addition, the Early Miocene thryonomyid material from Jebel Zelten (Libya) described by Wessels et al. (2003) is very close in morphology and size to Paraphiomys sp. nov. from Saudi Arabia and therefore could belong to this species. Paraphiomys sp. nov. from Saudi Arabia, differs from Gaudeamus aegyptius, Neosciuromys africanus, Apodecter stromeri, P. orangeus, P. renelavocati sp. nov., P. simonsi, P. shipmani, P. afarensis, P. roessneri, P. australis, Kochalia geespei, Paraulacodus indicus, Paraulacodus johanesi and the two living species of Thryonomys notably in having metalophulid II on the lower molars. It is distinct from P. pigotti and P. occidentalis in being much smaller than these and from Epiphiomys coryndoni in having a much shorter metalophulid II. Paraphiomys sp. nov. from Saudi Arabia differs from P. hopwoodi in being smaller, having the upper molars more antero-posteriorly compressed and in showing a more labially displaced hypocone. Finally, this species is distinct from Sacaresia moyaeponsi in having the teeth semibunodont whereas the latter species is hyperlophodont with slender lophs.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F576FF936417E0C891B6FA18.taxon	description	We name this species P. renelavocati sp. nov. to honour the extensive work of René Lavocat on rodent systematics. The mandible with d 4 - m 3 from the Lower Miocene deposits of Rusinga depicted by Lavocat (1973: pl. 26, fig. 9) is chosen as holotype (Fig. 3 H). It is housed in the National Museums of Kenya (Nairobi), where it is available for study under the reference number KNM RU 2208. Paraphiomys renelavocati is diagnosed as a Paraphiomys species of middle size with upper molars with mesoloph and lower molars lacking the metalophulid II. Paraphiomys renelavocati is bigger and less high crowned than Apodecter stromeri. It is distinct from P. pigotti, P. hopwoodi, Paraphiomys sp. nov. from Saudi Arabia, P. occidentalis, Epiphiomys coryndoni and Sacaresia moyaeponsi in the absence of metalophulid II on the lower molars. Paraphiomys renelavocati is much smaller and less hypsodont than Neosciuromys africanus and P. simonsi. It differs from Gaudeamus aegyptius, among many characters, in having an anterolabial cuspid on the lower molars and in having the hypolophid much less obliquely directed. Paraphiomys renelavocati shows the hypolophid on the d 4 anteriorly directed whereas in P. orangeus it is transverse. Paraphiomys renelavocati is distinct from P. roessneri in having a mesoloph on the P 4 and M 1 (Mein et al., 2000: 378) and from P. australis and both species of Paraulacodus in being much smaller. Paraphiomys renelavocati differs from P. afarensis mostly in having the m 1 and the m 2 nearly equal in length and an anterolabial cuspid on the lower molars, and the hypolophid nearly transverse.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F576FF926438E66F9050FDDC.taxon	description	This species (Fig. 4 D) has been described on the basis of material from the Lower Miocene of Langental and Plant 4 borehole, south of Lüderitz Bay (Namibia). Stromer (1922,1924: 263; 1926: 135 – 136) described and illustrated as N. africanus two mandibular fragments with p 4 - m 2 and a fragmentary left maxilla. Interestingly enough, the latter specimen is actually a right mandibular fragment with m 2 - m 3 of the bathyergoidid Bathyergoides neotertiarius and the abnormally narrow M 1 figured by Stromer (1926: pl. 42, fig. 24) as cf. Phiomys andrewsi belongs in fact to N. africanus.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F576FF926438E66F9050FDDC.taxon	description	Lavocat (1973: 39) considered Phthinylla fracta as a junior synonym of P. pigotti, which was acknowledged by Hendey (1978). Flynn & Winkler (1994: 230) and Winkler (1994: 181) suggest that Phthinylla fracta represents the upper dentition of Neosciuromys. In effect, the type maxilla of Phthinylla fracta and the type mandible of N. africanus match: they are com- mensurate and show an equally strong hypsodonty (R. López Antoñanzas, S. Sen & P. Mein, pers. observ.). These two taxa have close (if not identical) stratigraphic and geographical origins and, interestingly enough, new excavations at Elisabethfeld carried out by one of us (P. M.) have resulted in the discovery of Phthinylla – like maxillary teeth along with Neosciuromys - like mandibles. The recurrence of this association in addition to the intrinsic correspondence between Phthinylla fracta and N. africanus strongly suggest that the former taxon is a junior synonym of the latter. They should definitely be considered as such until evidence to the contrary comes to light.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F577FF9264F2E27C9036FD2A.taxon	materials_examined	Type species: Apodecter stromeri Hopwood, 1929	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F577FF9164E4E2E39409FC2F.taxon	description	The holotype of A. stromeri (AMNH 22538; Fig. 4 H) is a right mandibular ramus with m 1 - m 3 from an uncertain Lower Miocene locality of the Namib desert, south of Lüderitz Bay (Hopwood, 1929). Lavocat (1973: 42) reallocated this species to the genus Paraphiomys. Lavocat (1973): 44) noted several differences between the specimens of Paraphiomys stromeri found at Rusinga (Kenya, Lower Miocene) and those from Songhor (Kenya, Lower Miocene): the former are smaller and lack the metalophulid II, whereas the lower molars from Songhor frequently show it. Somewhat unusually, Lavocat (1973) considered these differences insufficient to justify a specific distinction, but nonetheless reliable indications for recognizing two subspecies of P. stromeri. Therefore, he coined P. stromeri stromeri to accommodate both the material from Rusinga and the holotype specimen of Apodecter stromeri, and P. stromeri hopwoodi for the sample from Songhor. We concur with Winkler (1992: 244) in considering that the morphological differences between these two ‘ subspecies’ are perfectly adequate to elevate them to the species rank. However, there is no reason to reallocate A. stromeri to Paraphiomys, as Lavocat (1973) did. The Namibian holotype of A. stromeri, which may be 2 Myr older (from rocks deposited 19 Mya), is smaller and more hypsodont than the material from Rusinga (Mein et al., 2000: 377) assigned by Lavocat (1973) to P. stromeri stromeri. In conclusion, it is not justifiable to consider the holotype of A. stromeri and the remaining specimens of P. stromeri stromeri as representatives of the same species. Therefore, (1) the genus Apodecter is here resurrected, as envisaged by Flynn et al. (1983), (2) a new specific name is given to the material from Rusinga (P. renelavocati sp. nov.), and (3) the correct name for the material from Songhor is P. hopwoodi. Incidentally, de Bruijn (1986: 129) reallocated A. stromeri to Neosciuromys. Unfortunately, he did not indicate why, and we judge this combination unjustifiable. Apodecter stromeri differs from P. pigotti, P. hopwoodi, Paraphiomys sp. nov. from Saudi Arabia, P. occidentalis, Epiphiomys coryndoni and Sacaresia moyaeponsi in lacking the metalophulid II on the lower molars. Apodecter stromeri is much smaller and less hypsodont than N. africanus and P. simonsi. Among other features, it differs from Gaudeamus aegyptius in having an anterolabial cuspid on the lower molars and the hypolophid much less obliquely directed. Apodecter stromeri has the m 3 more reduced posteriorly than that of P. orangeus. It is distinct from P. roessneri in having a mesoloph on the P 4 and M 1 (Mein et al., 2000: 378). Apodecter stromeri is much smaller than P. australis, Paraulacodus indicus and Paraulacodus johanesi. Finally, A. stromeri mainly differs from P. afarensis in having the m 1 and the m 2 nearly equal in length, in having an anterolabial cuspid on the lower molars, and in having the hypolophid nearly transverse.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F574FF916780E40B97AAFBD1.taxon	materials_examined	Type species: Paraulacodus indicus Hinton, 1933	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F574FF906787E499979CFD75.taxon	description	There is definitely no reason to assume that these specimens where found associated and, furthermore, it is uncertain that they are from the same horizon. Interestingly enough, all three are believed to originate from the ‘ Upper Chinji Zone’ (Black, 1972), whereas specimens recently found in the Pakistani Siwaliks are from the Middle Chinji Formation (L. Flynn, pers. comm.). Be that as it may, G. S. I. D 281 – D 283 are almost certainly all middle Miocene in age. According to Jaeger et al. (1980: 369), the single lower molar G. S. I. D 282 does not belong to Paraulacodus. Without justification, de Bruijn & Hussain (1985) considered G. S. I. D 282 to be possibly an m 1. Since the m 1 are expected to be smaller than the m 2, they assumed that the lower molar would be too large to fit with the type. Consequently, they evoked the possibility that Hinton’s (1933) material comprises more than one species: G. S. I. D 282 would pertain to the genus Paraulacodus, but not to the species P. indicus. This option is rejected here because (1) the identification of isolated m 1 and m 2 is problematical (especially if the material is limited) and (2) the size variation in P. indicus is unknown. An isolated upper molar of Paraulacodus aff. indicus from the Chinji Formation (Middle Miocene) in Pakistan was reported by Wessels et al. (1982: 358, pl. 2, fig. 14). According to these authors, the morphology of this tooth is identical to that of the holotype of P. indicus, but is much smaller. Flynn et al. (1983: 357) mentioned minor morphological differences such as the stronger hypsodonty and the stronger ectoloph of this specimen with respect to the upper molars of P. indicus. Owing to the great difference in size and slight morphological dissimilarities between this tooth and equivalent teeth of P. indicus, de Bruijn & Hussain (1985) considered that it does not belong to this species and designated it as Paraulacodus sp., an opinion that we share. De Bruijn & Hussain (1984, 1985) reported new specimens from the Sindi (Pakistan) locality H-GSP. 82.14 (top of the Lower Manchar Formation, Middle Miocene) that they finally identified as Paraulacodus cf. P. indicus. Recently, the Pakistani Siwaliks have yielded further specimens of this species (Flynn & Winkler, 1994). This material consists of one right dentary fragment with m 2 - m 3 and incisor (Y-GSP 31380, from locality 691), one left dentary fragment with m 2 - m 3 (Y-GSP 31809, from locality 750), one upper incisor fragment (Y-GSP 33105, from locality 698), and six isolated teeth (Y-GSP 40066 – 40071, from locality 714). According to Jaeger et al. (1980), P. indicus differs from P. johanesi notably in having smaller upper incisors compared to the Ethiopian form. Yet, the size of incisors depends on the age of the individual. In addition, Flynn et al. (1983: 363) argued that, owing to the scarcity of comparative material, i. e. in the absence of a thorough knowledge of the variation, it is not possible to differentiate these species from a dimensional point of view. Finally, Jacobs et al. (1989: 167) asserted that both taxa are similar in size. In the Asiatic species of Paraulacodus, the size discrepancy between the M 2 and the M 1 is greater than in P. johanesi. The anterolabial cusp on the lower molars is stronger in most specimens of P. indicus than in P. johanesi. Another difference is the tendency of the protoloph of P. indicus to bend forwards towards the labial end of the anteroloph (nearly closing the mesoflexus). Paraulacodus indicus differs from Paraphiomys pigotti, Paraphiomys hopwoodi, Paraphiomys sp. nov. from Saudi Arabia, Paraphiomys occidentalis, Epiphiomys coryndoni and Sacaresia moyaeponsi in lacking the metalophulid II on the lower molars, to cite just one among quite a few differences. Paraulacodus indicus differs from Neosciuromys africanus and Paraphiomys simonsi, for instance, in its smaller size and reduced hypsodonty. Paraulacodus indicus is much larger than Kochalia geespei, Paraphiomys shipmani, Paraphiomys orangeus and Apodecter stromeri. One of the main differences between P. indicus and Paraphiomys australis and Paraphiomys roessneri is the presence of an anterolabial cusp on the lower molars of the former.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F575FF906418E43E9022F8E7.taxon	description	One of the differences between E. coryndoni and Paraphiomys occidentalis and P. pigotti is the size, which is much smaller in E. coryndoni. Another dissimilarity is the more developed metalophulid II of E. coryndoni. Epiphiomys coryndoni has the metalophulid II much more developed than in P. hopwoodi or Paraphiomys sp. nov. from Saudi Arabia. Epiphiomys coryndoni is distinct from Sacaresia moyaeponsi, for example, in being less lophodont and in having four lophed upper molars. It differs from Gaudeamus aegyptius, Neosciuromys africanus, Apodecter stromeri, Kochalia geespei, P. renelavocati sp. nov., P. roessneri, P. australis, P. shipmani, P. orangeus, P. afarensis, Paraulacodus johanesi and Paraulacodus indicus, among many other characters, in the presence of a metalophulid II on the lower molars.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F575FF9064FEE3BE901CFC6C.taxon	materials_examined	Type species: Epiphiomys coryndoni Lavocat, 1973	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F575FF90679EE6FF916FFCEB.taxon	description	Two mandibular rami (one of them toothless) from Fayum were first attributed to Phiomys andrewsi by Schlosser (1911: 90 – 2, pl. 5, fig. 7). Later, Stehlin & Schaub (1951: 266 – 267) showed that the teeth were much more derived than those of Phiomys andrewsi. The taxon in question was considered new and named Gaudeamus aegyptius by Wood (1968: 68 – 73). The holotype (CGM 26920) is a lower jaw with d 4 - m 2 and unerupted premolar (Wood, 1968: fig. 14 C – E). The cheek tooth pattern of G. aegyptius is very similar to that of Thryonomys, but the teeth of G. aegyptius are much smaller, shows less developed crests, and were replaced. Gaudeamus aegyptius differs from Paraphiomys pigotti, P. hopwoodi, Paraphiomys sp. nov. from Saudi Arabia, P. occidentalis, Epiphiomys coryndoni and Sacaresia moyaeponsi in lacking the metalophulid II on the lower molars. It is smaller and less hypsodont than Neosciuromys africanus and P. simonsi. In contrast to the condition in Apodecter stromeri, P. orangeus, Paraulacodus indicus and Paraulacodus johanesi, G. aegyptius lacks an anterolabial cuspid on the lower molars and has the hypolophid much more obliquely directed.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F575FF906668E67F97D4F92A.taxon	materials_examined	Type species: Gaudeamus aegyptius Wood, 1968	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F575FF9067ABE3DC94BCF9AA.taxon	description	The generic attribution of this species was rejected by de Bruijn (1986: 131) who reallocated it to Neosciuromys. Nevertheless, the phylogenetic analysis conducted by Winkler (1992) and Flynn & Winkler (1994) advocates that P. johanesi is the sister species of P. indicus, type species of the genus (Black, 1972), not Neosciuromys africanus. The differences between this species and P. indicus have been discussed above.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F572FF976790E0C897ACFEC0.taxon	materials_examined	Type species: Kochalia geespei de Bruijn et Hussain, 1985 Synonymy: Kirtharia de Bruijn et Hussain, 1985	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F572FF9767FAE193947FF8E6.taxon	description	This species was described, illustrated and originally coined as Kirtharia geespei by de Bruijn & Hussain (1985). Because the name Kirtharia de Bruijn et Hussain, 1985 was incorrectly believed to be a homonym of Khirtharia Pilgrim, 1940 (an artiodactyl), de Bruijn (1986: 125) reallocated this species to a new genus name: Kochalia. This name has been used by all subsequent authors, but, as noted by Mckenna & Bell (1997: 195), the act of de Bruijn, 1986) is a violation of the I. C. Z. N. rules. Pending a possible adjudication of the matter, we shall continue to use the name Kochalia in the present paper. The holotype of this species is a left d 4 or p 4 (p 4 according to de Bruijn & Hussain, 1985: pl. 1, fig. 7 a, b). According to de Bruijn & Hussain (1985), K. geespei differs from all known African thryonomyids, except for Gaudeamus aegyptius, in showing tooth replacement. However, the two teeth described by these authors as d 4 and p 4 show a similar morphology and the only difference between them rests in size. Therefore, further work might reveal that the two teeth correspond in fact to d 4. Kochalia geespei, as well as all Neogene thryonomyids, could have retained the deciduous premolar throughout their lifespan. As remarked by de Bruijn & Hussain (1985: 159), the material from the Chinji Formation described and figured by Flynn et al. (1983: fig. 2 a – f), as ‘ morphotype 3 ’ (that they considered represented a thryonomyid more derived than Paraphiomys and of a lineage distinct from Paraulacodus) seems to belong to this species. Kochalia geespei differs from Paraphiomys pigotti, P. hopwoodi, Paraphiomys sp. nov. from Saudi Arabia, P. occidentalis, Epiphiomys coryndoni and Sacaresia moyaeponsi, amongst other characters, in lacking a metalophulid II in the lower molars. It is distinct from Neosciuromys africanus and P. simonsi in being smaller and less hypsodont. Kochalia geespei mainly differs from P. roessneri and P. australis in having a prominent anteroconid, and from P. shipmani in having a smaller m 3 and a small hypoconulid. According to de Bruijn & Hussain (1985: 159), this taxon differs from Paraulacodus in having a bunodont d 4, in having the anteroloph of the D 4 not constricted near the protocone, and in being smaller. Kochalia geespei has the posterior part of the m 3 much more reduced than in P. orangeus.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F572FF97645CE0C896ACFEB1.taxon	materials_examined	Type species: Sacaresia moyaeponsi Hugueney et Adrover, 1991	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
D417CA60F573FF9667C9E29B96D9FD53.taxon	description	The software of phylogenetic reconstruction by parsimony analysis used was PAUP v. 3.1.1 (Swofford, 1993). The relatively high number of terminal taxa and characters precluded exact tree building, so a heuristic search by stepwise addition (‘ closest’ option) was performed.	en	Antoñanzas, Raquel López, Sen, Sevket, Mein, Pierre (2004): Systematics and phylogeny of the cane rats (Rodentia: Thryonomyidae). Zoological Journal of the Linnean Society 142: 423-444
