1904.] MOUSE-llAUES OF TUB GEXUS OCHOTOXA. 219 



Skull. Palatal length 12 nini. ; zygomatic bi-eadtli 15 ; length of 

 molar series 6'5 ; length of nasals 10 ; interorbital breadth 4. 



Habitat. Kashmir ; Tibet ; Szechuen. 



This species is allied to the preceding, the dull pelage of which 

 it gi-eatly resembles ; its mnch smaller size will, however, enable 

 it to be easily recognised. 



I have seen no specimens from Kashmir, the type-locality of 

 this species, but there is a series of specimens in the Museum 

 from E. tSikkim, which agree so closely with Blyth's description, 

 that I have no hesitation in referring them to this species. I have 

 also examined specimens from Szechuen, which are indistinguish- 

 able from those of E. Sikkim, as well as a single indivi<lual from 

 Gannsu in N. Thibet. M. Milne-Edwards' speci&s from Moupin 

 is also, in my opinion, identical with 0. hodysoni, of ^vhich it 

 therefore becomes a synonym. 



Since the foregoing has been in the possession of the Society, 

 Mr. Marcus W. Lyon, jun., lias published* an exhaustive paper 

 on " The Classification of the Hares and their Allies," founded 

 almost entirel}^, as would naturally Ije the case, on their osteo- 

 logical characters. He divides the genus Ochotona (1. c. p. 4.38) 

 into thi-ee subgenera founded on cranial characters, and it is 

 gratifying to find that his subgenera correspond with the three 

 groups into which I have found it necessaiy to subdivide the 

 genus, as set out in this paper. For the names of his subgenera 

 Mr. Lyon makes use of two previously existing, viz. Ochotona and 

 Pika, and coins a new one, C'onothoa. The subgenera Ochotona, 

 Conothoa, and Pika represent respectively the rufescens, ctirzonice, 

 and laclacensis groups of this paper. 



With regard to the actual classification of some of the species, 

 there are several apparent discrepancies Ijetween Mr. Lyon's 

 results and my own. In considering these, however, it must be 

 borne in mind that Mr. Lyons was treating the subject fi-om a 

 lai'ger point of view and also almost entirely from the osteo- 

 logical side ; whereas in my work geographical distribution and 

 external characters were more especially studied, and I had the 

 additional advantage of superior series of skins, owing to many 

 of the actual types being in the Muserun, as well as a nearly 

 complete set from Biichner of the various species he had desciibed. 



The fii'st discrepancy is the grouping together of 0. ladacensis 

 and 0. koslovn. The figure given by Biichner of the skull of the 

 formei- does not agi-ee with the ty])Q skull, as I have already 

 pointed out {antea p. 209), Avhich undoubtedly belongs to the 

 same gi'oup as 0. alpina, and there can be but little douljt that 

 Biichner's figiu'e misled Mr. Lyon. 



The next discrepancy is with regard to 0. erythrotis, which 

 Mr. Lyon, on the strength of Biichner's figure, places in the 

 curzonice group. I have not been able to examine specimeixs of 

 0. erythrotis^ but there are in the Museum a series of 0. rntila 



* Smithsonian Miscc41. Coll. vol. xlv. p. 321 (1904), 



