1902.] THE UPPER ILI AND YANA VALLEYS. 83 



be identified with certainty. I accordingly propose to regard it as 

 a local race of that species, with the title of 0. sairensis littledalei. 

 The white muzzle and dark forehead serve to distinguish 0. sairensis 

 from both 0. jyoli and 0. ctmmori ; and the typical Saiar race of the 

 former species may be dift'erentiated from the Hi race by its 

 inferior size and the much closer spiral formed by the horns. I 

 have elsewhere suggested that 0. sairensis should be known in 

 English as Littledale's Sheep ; and the two races of the species 

 may be severally distinguished as Littledale's Saiar Sheep and 

 Littledale's Hi Sheep. 



The description of a new subspecies is of but little interest 

 unless some deduction can be drawn with regard to the habits or 

 distribution of the group to which it belongs. In this case 

 something of this nature can, I think, be suggested. A glance at 

 the map of Central Asia will show that the Saiar and Ili Sheep 

 occur a]3proximately on the same great line of watershed ; and 

 that to the south-west 0. poll harelini, of the western Thian-Shan, 

 comes in on the same line, and thus continues the chain to the 

 Pamirs, the home of the typical 0. poli. Now all these four types 

 of Sheep have horns with a longer spiral than that of 0. amnion 

 and its races, and their habitat appears to form a kind of wedge 

 driven into that of the latter group. It is further noteworthy 

 that among the Argalis the length and openness of the horn-spiral 

 decrease from west to east, as exemplified by the occurrence of 

 0. 2ioli on the western and 0. ammon Jwdgsoni on the eastern 

 frontiers of the group. Nor is this all, for in the three species 

 0. foli^ 0. ammon, and 0. sairensis, the further east or the further 

 south they go it is noticeable that the less open becomes the spiral 

 of the horns. This is apparent when we compare 0. poli with 

 0. p. karelinij 0. a7nmon with 0. a. jv,hata and 0. a. hodgsoni, 

 and 0. sairensis with 0. s. littledalei. What may be the reason 

 for this feature, I am unable to conjecture. 



Admitting that the three species of Argali just mentioned are 

 very closely allied, the question may be legitimately asked, why 

 they and their subspecies are not all classed as phases of a single 

 species, as is done in the case of the Wild Goats of the same 

 region. To this it may be replied, that if such a course were 

 adopted it would be necessary to employ quadrinomialism, in order 

 to expi-ess adeqiiately the mutual relationships of the forms hej^e 

 regarded as local races of the three species mentioned. 



I may add that I have been told by more than one sportsman 

 that all the Central Asian A.rgalis pass more or less completely 

 into one another. Without denying the possibility that sucla 

 may be the case, it is certain that no such transition is exhibited 

 by the series of specimens in the British Museum. 



I will now proceed to deal with a Wild Sheep recently brought 

 by Mr. J. Talbot Clifton from Northern Siberia, at a point 

 distant about 40 miles from the mouth of the Yana River. 



6* 



