258 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. [Mar. 11, 



antero-posteriorly, lower, thinner transversely, and more trenchant, 

 than in the tyipe-AnopIot/ieria (whence the name Xiphodon^ or Sword- 

 tooth). The feet are didactyle, with metacarpals and metatarsals 

 distinct. The tail is short. The lower true molars have two pairs 

 of crescentic lobes with the convexity turned outwards. But per- 

 haps the most important proof of generic distinction from Dicho- 

 biine and Dichodon, e. g., would be afforded by the modifications of 

 the crown of the true molars of the upper jaw. AVith these I am 

 acquainted only by the figures and descriptions given by Cuvier and 

 M. Gervais. 



An oblique view of the grinding surface of the true molars of one 

 side of the upper jaw of Xiphodon gracilis is given by Cuvier in the 

 pi. 14, Supplement, of the Original Memoir above quoted (pi. 52. 

 ed. 1822) ; in which the inner cone of the front half of the tooth 

 appears, in the second molar, to be separated from the outer cone 

 by an intermediate ridge or cone of equal size with the outer one, and 

 not of half that size, as in Anoplotherium proper. 



The same tooth, m 2, in the figure of the five last grinders of the 

 upper jaw of the Xiphodon gracilis, from Debruge (Paleontologie 

 Fran9aise, pi. 34, fig. 2), shows also three cones on the anterior 

 moiety of the crow^n, and M. Gervais, in his description of that figure, 

 expressly says, it is given " pour montre les trois pointes ou pyra- 

 mides de la colline ante'rieure et les deux pyramides de la coUine 

 posterieure." (PI. 34. p. 1.) But in his characters of the genus 

 Xiphodon (op. cit. p. 90) he cites " les arrieres molaires a deux col- 

 lines, formees superieurement de deux pointes chacune, subarques et 

 rappelant celles des ruminants." 



Now", this character would apply to Dichodon, but it does not ac- 

 cord with the published figures of the upper true molars, or with his 

 own description of one of those figures, of the Xiphodon gracilis. 



The first and second lower molars have two pairs of lobes, the 

 third having an additional hinder lobe ; but that lobe appears to be 

 simple, and the internal lobes of the normal pairs have not the 

 accessory basal cusps, as in Dichodon. M. Gervais adds the fol- 

 lowing remark with reference to the aflftnities of Xiphodon : — 

 '* Ces dents (de V Hyopotamiis porcinus) sont a deux collines trans- 

 verses, dont I'anterieure a trois pointes ou pyramides, et la seconde 

 deux seulement. Elles indiquent une espece intermediaire a VAn- 

 thracotherium magnum et a V Hyopotamus velaunus, et montrent 

 aussi que les Xiphodons sont des animaux tres-voisins des Hyopo- 

 tames." (Paleontologie Fran^aise, Description of pi. 31. p. 1.) 



But the genera Anthracotherium and Hyopotamus differ, in the 

 insulation of ^ 1 from both the canine and from p 2, from the genera 

 Anoplotherium, Xiphodon, and Dichodon, in which the dental series 

 is continuous. 



Genus Dichobune. 



The genus Dichobune (from ^/x"» hipartito ; ^owos, collis) was 

 proposed by Cuvier, in the second edition of his * Ossemens Fossiles,' 



