ELEPHANT AND MASTODON. 103 



America, some of which have been figured and described by 

 Peter Collinson and William Hunter.' 



The ridge formula, in the successive molars of M. Andium., 



may therefore be safely expressed as [-g ! ^3+ 3 ^^ ^^ TaJiSk 

 molars, and 3 ;); 3 | ^.^ in the true molars, the only difference 



from M. Ohioticus being an additional ridge in the second 

 milk molar. The constancy of the ternary division in the last 

 milk molar and the first and second true molars in both jaws 

 of these species, andof ilf. angustidens, defines a weU-marked 

 group of mastodons, for which we propose the sectional 

 name of Trilophodon. Of these M. Ohioticus was the most 

 colossal form, next M. Andium, and last M. angustidens, 

 which appears to be the smallest known species among the 

 elephantine Proboscidea. While the first of all the species 

 indicates the nearest affinities to Dinotherium, the molar teeth 

 of the two last present close analogies to those of Eiiypopota- 

 mus, to gigantic forms of which genus the earher palaeon- 

 tologists were led to refer them. To the same section M. 

 Tapiro'ides appears also to belong, so far as the limited infor- 

 mation regarding the dentition of the species will warrant 

 an inference respecting it. 



M. Tapir o'ides. — Cuvier founded this nominal species upon 

 a single tooth from Montabusard, near Orleans,^ the crown 

 of which, divided into three lobes, is not bisected along the 

 axis by a longitudinal furrow, as in M. angustidens and the 

 other species, these eminences being continued across, and 

 their edges simply cremulated with small regular dentictila- 

 tions, as in the teeth of Dinotherium. It has also a small 

 talon, which exhibits the same cremulated character. This 

 tooth would correspond in size with the second milk molar 

 of M, Andium. Regarded per se, it does not furnish suffi- 

 cient evidence to establish a distinct species of Mastodon. 



Buffbn, in 1778, had described two large grinders, the one 

 stated to have been found in Little Tartary, and the other in 

 Siberia, both of which Cuvier referred to M. Ohioticus, at 

 the same time that he questioned the accuracy of the locali- 

 ties attributed to them ; having restricted the geographical 

 range of this species to the ISTorth American continent. 

 Doubts were also expressed by him regarding a three-ridged 

 grinder, figured by Pallas,^ and described by the latter as 

 belonging to the species of the Ohio, although found in the 

 Ural mountains. In the additions to the last edition of the 

 'Ossemens Possiles,' Cuvier relinquished his doubts regarding 



• Phil. Trans, vol. Ivii. | ^ Oss. Foss. torn. iii. addit. p. 375. 



2 Oss. Toss. p. 267, tab, iii. fig. 6. | 



