122 p. N. B05E ON POSSIL CARKIVORA 



Blanford have made the age of the latter to hinge on that of the 

 former. After quoting M. Gaudry's authority to prove that the 

 Pikermi fauna is Pliocene and not Miocene, they conclude by saying 

 that " there can be, therefore, no reasonable doubt that the Pikermi 

 fossils, like the Sivalik, are of Pliocene age, and that the quotation 

 of them as Miocene is an error"*. But if it is an error to quote the 

 Pikermi fauna as Miocene, M. Gaudry himself, on whose authority 

 this conclusion is based, is guilty of it; for in his latest workt he 

 constantly cites the Pikermi fossils as Upper Miocene. Consider- 

 ing, therefore, the strong affinities of the Sivalik fauna, as a whole, 

 to the known Miocene fauna, and having regard to its close rela- 

 tionship with the Pikermi fauna, it would, I venture to suggest, be 

 advisable, for historical purposes, to place it on the same horizon as 

 the latter, and regard them as Upper Miocene. In the Sivalik 

 fauna, however, there is most probably an admixture of some 

 Pliocene forms. 



It will be seen from the list given above that there are 17 species 

 of Carnivora known to occur as fossil in the ossiferous Sivaliks*; but 

 there are indications of some more. In the collection of the British 

 Museum there are several mutilated skulls belonging, most probably, 

 to small forms of Carnivora, but the affinities of which, in the absence 

 of dentition, I have been unable to determine exactly. We may 

 therefore safely add 3 more species, making the total number 20. 

 In the Sivalik area, at the present day, there are altogether 17 or 

 18 species of Carnivora J, showing a decrease of 3. The Carnivore 

 fauna of India consists of about 40 species §, that is to say, double 

 the number known from the Sivaliks. This is owing to the increase 

 in the smaller forms, the number of the larger forms having decreased. 

 The numerical strength of the existing Carnivore fauna of India 

 presents a striking contrast to the extreme poverty of the Ungulate 

 fauna (including the Proboscidians), which does not consist of more 

 than 27 species ||. No less than 60 species of Ungulates (including 

 Proboscidians) are known from the Sivalik beds — that is to say, more 

 than double the number now inhabiting the whole of India. The 

 proportion of the fossil to the living Ungulates in the Sivalik area 

 itself, as pointed out by Messrs. Medlicott and Blanford, is as 5 to 2. 

 "We therefore come to the conclusion that if the Carnivore fauna 

 has been impoverished, the impoverishment has been considerably 

 less than that of the Ungulate fauna. 



Mach^eodus sivalen-sis, Pale, et Cautl. et nob. (PI. YI. fig. 5.) 



Two imperfect crania, six fragments of lower jaws, and two of 

 upper jaws, in the collection of the British Museum, are all labelled 

 Drepanodon {Machoerodus) sivcdensis. Of the six specimens of lower 



* Op. cit. p. 584. 



t ' Les enchainements du Monde Animal,' Paris, 1878, pp. 48, 264, 265, 271, 

 273, &c. 



X Medlicott and Blanford, op. cit. p. 589. 



Jerdon's ' Mammals of India,' pp. 68-154. 



Jerdon, op. cit. p. 228. 



