External Oharacters of the Palm- Civet. 159 



■of the pads, and the extent to which the lower sides of the 

 feet are overgrown with hair. Nothing unfortnnately can 

 be made out with regard to the scent-ghmd, but tlie affinities 

 between the tliree genera indicated by the many points of 

 resemblance and the fewness and nature of the differences 

 between them justify, from the analogy supplied by the 

 Viverrines and Paradoxurines, the conclusion that Diplogale 

 and Chrotogale possess scent-glands similar to, though per- 

 haps not absolutely structurally identical with, those of 

 Hemigalus. If this be so, the three genera in question form 

 a compact little group of the Viverridae, distinguishable ou 

 the one hand from the Viverrinse, and on the other from 

 the Paradoxurinae, exemplified particularly by Paradoxurus 

 and Paguma, and in some respects, especially as regards the 

 feet, intermediate between them. I think, therefore, that 

 Gray was right in isolating Hemigalus, and I propose to 

 adopt for the group it typiHes the name he proposed under 

 its modernised form, Heniigalinse, and to regard this group 

 as of subfamily rank. 



Possibly from want of properly preserved material, I am 

 not able to point out any external characters to distinguish 

 these three genera, apart from those mentioned by Thomas : 

 namely, the absence of pattern in Diplogale and the normal 

 direction of the hair on theneck in Chrotogale — characters 

 which in themselves are hardly of generic value. Reliance 

 must therefore be placed upon the cranial and dental 

 characters mentioned by Thomas. 



By these the three may be separated as follows : — 



fl. Premaxillfe (in immature form, at least) produced far 

 beyond canines, separated by a long median palatal 

 fissure, the incisors broad and forming a strongly 

 curved line ; direction of hair on the neck normal ; 



(pattern as in Hemigalus) Chrotogale. 



b. Premaxillfe normally produced, in contact on the palate 

 throughout their length ; the incisors narrow and 

 forming a gently curved line ; hair on the neck reversed 

 in direction. 

 a'. First upper premolar small, one-rooted, palate con- 

 siderably wider behind than in front ; pattern com- 

 posed of broad transverse stripes on a light ground . Hemigalus. 

 b'. First upper premolar large and two-rooted; palate 

 nearly as wide in front as behind ; no pattern, 

 colour " otter-like " Diplogale. 



In attempting to affiliate these genera, there are one or 

 two points to consider. First, the very specialised pattern 

 of Hemigalus and Chrotogale, and, second, the specialised 

 dentition of Chrotogale. Taking these two features in 

 conjunction, I think the assumption that Chrotogale is a 



