90 FAUXA ANTIQUA SIVALENSIS. 



enter at some length upon tlie evidence on this point, more 

 especially, as the two latest authorities of weight, Prof. Owen 

 and M. de Blainville, do not admit a specific difference 

 between M. angustidens and M. longirostris. On the other 

 hand. Dr. Kaup has, in some instances (as in the case of the 

 Stellenhoff lower jaw found near Vienna * ) excluded from his 

 M. longirostris, specimens which assuredly belong to it; while 

 in others (viz. the Georgensmiind mastodon teeth described 

 by von Meyer) he has transferred to this species molars 

 which appear to pertain to M. angustidens. 



The first point to determine, under these circumstances, is 

 the form to which the specific name of M. angustidens is pro- 

 perly applicable. Cuvier's description of the species com- 

 mences with the Simorre tooth'^ which has the crown divided 

 into three ridges, with a back talon of two tubercles, measur- 

 ing 4*5 inches in leng-th by 2-35 in width. The next specimen 

 which he describes as belonging to it is the Dax fragment 

 (Oss. Foss. PI. III. fig. 2) containing two teeth implanted in the 

 palate on one side, the anterior of which is the unworn germ 

 of a premolar, and the posterior, nearly of the same size as 

 the Simorre tooth, like it consists of three ridges and a small 

 talon of two tubercles. A third tooth which he immediately 

 afterwards attributes to this species, is another Simorre speci- 

 men (Oss. Foss. PI. III. fig. 3) measuring 3*6 inches by 2-6, and 

 having its crown also divided into three ridges. It is therefore 

 to a species having the intermediate molars distinguished by 

 a ternary division of the crown, as in M. Ohioticus, that the 

 specific narae of M. angustidens is strictly apj)licable, so far as 

 priority of description and reference to original types can be 

 taken as the guides to a decision on the point. 



Of the other specimens referred by Cuvier to his M. angus- 

 tidens, and represented in the four plates devoted to ' Divers 

 Mastodontes,' the South American teeth (figs. 6 and 7 of PI. I. 

 and fig. 4 of PL III.) appear to belong to M. Andium, as has 

 been advanced by M. de Blainville, and nearly all the rest, 

 which are susceptible of determination, belong to M. Arver- 

 nensis (M. longh'ostris of Kaup), with the exception of figs. 1 

 and 2 of PL I., fig. 11 of PL II., and fig. 14 of PL III., which 

 are probably to be referred to M. angustidens. 



We have already stated the grounds^ upon which Von Meyer, 

 following up the observations of Croizet and Jobert, distin- 

 guished M. Arvernensis from M. angustidens, and that Kaup 

 was led by his researches to the same conclusion. It would 

 appear, from a communication in Bronn's ' Lethsea,' that the 

 ridge formula which Kaup attributed with doubt to 31. angus- 



' Cuv. Oss. Foss. edit. 1831, toni. ii. 

 p. 363, pi. ii. figs. 4 and 5. 



- Oss. Fossil, torn. i. p. 'loo ; Divers 



Mastodontes, pi. i. fig. 4. 

 " Ante, p. 60. 



