610 ME. C. FORSTER COOPER ON MIOCENE 



Lydekker's variety pcdceindica, which was founded chiefly on a 

 third lower molar and which he states is within the range of 

 variation of var. typica. The specimens now to be described 

 tend, if anything, to reverse Schlesinger s view and in part to 

 confirm Lydekker's. 



The material in the collection consists of two palates of 

 B. angustidens of slightly different age, part of a lower jaw 

 belonging to the older of the two palates, and some separate upper 

 and lower molar and premolar teeth. 



The younger of the two palates (text-fig. 1) has on each side 

 the last premolar and first molar in moderate wear, and the 

 second molar erupting and unused. In front are the broken 

 roots of the third premolar. Beyond this point and behind the 

 second molars the specimen ends. The fourth premolars are 

 weathered, but show a square outline and four subequal cusps a 

 little elongated in the transverse direction. There is a shelf in 

 front and low talon behind, and there are no intermediate cusps. 

 They are very similar in size and shape to a correspoding tooth 

 figured by Schlesinger for B. longirostre. There is no sign of a 

 successional tooth underneath, so that it is not likely that they 

 are imperfect milk teeth. 



The first molars are three-ridged, with a very small talon. The 

 first two ridges are worn, the third as yet untouched. They 

 agree more with those figured by Schlesinger as var. suhtapi- 

 roidea * than as var. typica t. 



The second molars, again, agree best with the figures of suh- 

 tapiroidea J, but here the cusps seem more separate from each other, 

 a diffei-ence which may be more apparent than real, and due in 

 large measure to the tooth being unworn; an internal accessory- 

 cusp on the third ridge is more prominent than his figures show. 



The other palate is i-ather older in that the second molar is 

 now in wear on the first two ridges. The specimen on its right 

 side has an open socket, from which the third premolar has fallen 

 out; the fourth premolar and first molar are well worn, the second 

 is partly in wear, and the front ridge of the third is preserved in 

 the alveolus. It is clear that this animal had two premolars and 

 two molars in simultaneous wear, which is a longer time than 

 Andrews § suggests for this species. The present specimen 

 appears to show that these four teeth i-emain until the third molar 

 erupts, when it finally pushes the premolars and first molar 

 forward and out. This is consonant with the view, supported by 

 other reasons, that the stage of 'angustidens represented by these 

 Indian forms is earlier than that of the more typical French 

 forms from Sansan etc., and is possibly as early as any yet 

 described. 



The structure of the teeth in this second palate is essentially 

 the same as in the other, except that in the second molars the 



* Schlesinger, Mitt. Geol. Geselt. Wien, Bd. xi. pi. iii. fig. 2. 



t Loc. cit. pl.'ii. fig. 2. J Loc. cit. pi. vii. fig. 3. 



§ Andrews, Phil. Trans. R. S. vol. xciv. p. 108 (1903). 



