820 MR. R. I. POCOCK ON THE 



frequented by typical Stoats, Weasels, and Polecats. If nudipes, 

 which ranges from the Malay Peninsula to Borneo, were alone 

 concerned, that view would be more defensible than, in my 

 opinion, it is. But strigidorsa occurs much farther to the 

 north, in Sikhim. Again, it must be remembered that the 

 tropical American species hrasiliensis, which may be taken as the 

 type of Gray's genus Neogale, if it be resuscitated, has hairy feet. 

 I incline, therefore, to the opinion that in Plesiogale we have the 

 most primitive type of foot met with in the subfamily Mustelinae 

 as here restricted — a foot which differs mainly from that of 

 G-rison and Grisonella in the complete suppression of the meta- 

 tarsal pads. 



The feet of Gulo * resemble in all essential respects those of 

 JIartes, except that they are relatively broader. 



The feet of Pmcilogcde are very like those of Plesiogale, but the 

 claws of the fore feet are a little larger and the third and fourth 

 digits of the hind foot less widely separated. In both these 

 particulars Pcecilogcde, so far as the feet are concerned, connects 

 Plesiogale with Ictonyx, which in the length of the claws belongs 

 to the fossorial group of Mustelidis — Poncilogale being one of 

 those genera in which the distinction between fossorial and 

 cursorial Mustelidse breaks down. (Text-fig. 32, A, B.) 



In the remaining genera the feet are fossorial and characterised 

 by long and blunt claws, which are especially long on the fore 

 feet. The only other point they have in common is close union, 

 sometimes amounting to basal fusion, between the digital pads of 

 the third and fourth digits of the hind feet — a phenomenon 

 foreshadowed in the feet of Grison and Tayra and Ptecilogale. 



Fossorial feet which come nearest to the feet of the Weasels, 

 Polecats, and Martens, although differing in the characters 

 mentioned in the last paragraph, are found in Ictonyx and 

 Piecilictis. The feet, closely resembling those of Pcecilogale, are 

 tolerably narrow, and the digits, apart from the third and fourth 

 of the hind foot which are close together, are widely separable ; 

 the metatarsal area is without trace of pads and covered with 

 hair down to or nearly down to the plantar pad ; the digital pads 

 a.re oval, well defined, and coarsel}^ striate — at least apically ; the 

 plantar pad is narrow, deeply emarginate behind, and very 

 distinctly four-lobed, and the lobes are coarsely striate and are 

 in contact to about the same extent as in Martes foina ; there 

 are two striate carpal pads, a smaller inner and a larger outer, 

 separated from each otlier and from the plantar pad by about 

 the same distances as in Ilartes foina, although the sizes of these 

 pa.ds and the distances above mentioned vary according to the 

 spesies. The carpal pads are largest and the distances in question 

 smallest in the species in which the area round them and the 

 area between the plantar and digital pads are naked. In other 



* See my paper on the external characters of this genus (Proc. Zool. Soc. 1920, 

 pp. 179-187). 



