248 EODENTIA. 



they appear to me to be referable to one species. Hodgson, who first described it, 

 referred to it all those Himalayan squirrels slightly larger than S. lokriah, and 

 which had the ventral surface either pale whitish or slightly washed with rufous, 

 the sides also being sometimes suffused with this tinge, especially on the anterior 

 haK of the outside of the thigh which in many is bright orange-red ; but this 

 colour is variable, and some squirrels have this portion of the body white, of which 

 S. hlytliii is an example, and others similar to it are before me from Bhutan and 

 Assam, which do not differ from S. lokroides, except in the presence of this white 

 area which is evidently only a variation on the red area, and probably a seasonal 

 change, as many show merely a faint rufous tinge in the inguinal region, that colour 

 being entirely absent on the outside of the thigh. 



It is, however, worthy of note, that those squirrels which have a rufous tinge 

 in the inguinal region, rarely or ever have the outside of the thigh bright red, 

 and that the squirrels distinguished by white on their thighs are from Bhutan, 

 Assam, and the Garo Hills. But I do not see that these latter differ in any other 

 respect from the squirrels sent by Hodgson as ^S'. lohroides, with and without red 

 thighs. Moreover, one of Hodgson's specimens of S. lokroides shows a tendency 

 in the thigh to become white. 



The ventral surface is seldom alike in two specimens. It varies from pure 

 white (young of Assam type) through grey to rufous grey generally suffused with 

 rufous on the groin and sometimes along the side and over the belly, but in 

 others the rufous is all but absent. In some there is a decided tendency to the 

 formation of a median, ventral, grizzled streak or line. The tail also is very variable, 

 being in some concolorous with the body, while in others it evinces a tendency to 

 become banded, black and rufous yellow, and to have a black tip ; but even this 

 wide difference of tail-colouring occurs in two animals which, in other respects, are 

 identical and from the same locality. The general tendency, however, for the tail is to 

 resemble the back. The upper surface is of variable intensity, the prevailing 

 colour being brownish-ohve, minutely punctulated with yellow or rufous. On the 

 character of the annulations depends the tinge of brown distinctive of the indivi- 

 dual. The feet in some tend to become white, but in the greater number they 

 are concolorous with the upper parts, or a little darker. 



In a specimen from Bhamo, where the species is not at all common, the under 

 surface is pale greyish, very faintly suffused with pale yellowish on the belly and 

 on the groin ; the upper surface is a rufous olive-brown, the tip of the tail showing 

 a tendency to black and white banding, but the rest of it is concolorous with the 

 back. 



The species may be described as rufous olive-brown, the base of the hairs 

 greyish-black, and their remainder banded with yellow, black and yellow, each hair 

 terminating in a dark-brown or blackish point. The fur is slightly coarser and more 

 broadly annulated than in S. lokriah. The chin, throat, and ventral surface are pale 

 greyish or sullied white, more or less tinged with rufous, or the rufous may be con- 

 fined to the inguinal region, with the rest of the under parts greyish, with occasionally 



