468 Mr. 0. Thomas on the 



sorted into their respective genera, in the modern restrictied 

 sense of the word, and the present paper is an attempt similarly 

 to arrange the African species. 



These latter are in a sense much more complicated and 

 difficult than their Asiatic allies, as the groups are less 

 obvious in general character, so that a careful study of the 

 teeth of every species has had to be made and the animals 

 sorted accordingly. 



It proves unfortunately that a larger number of genera 

 than is either pleasant or convenient demand recognition, if 

 we are to uphold the sound principle enunciated by Forsyth 

 Major that .squirrels should be classified by their dental and 

 cranial characters just as other rodents are, and do not rely 

 on such superficial characters as the presence or absence of 

 stripes, or similar external characters. I have tried to be as 

 conservative as possible, but there seems no stopping-place 

 between the present arrangement and the wholesale and 

 inconvenient lumping of all the forms in a single genus. 

 Such a lumping would conceal the natural relationships of 

 the different species and ignore all the iniportant structural 

 characters now dealt with. 



The basis of this work is the classical paper of 1893 *, by 

 Dr. Forsyth Major, adopted and modified by myself in 1897 t> 

 but now become more or less obsolete, and needing bringing 

 up to date by the examination of the skulls and teeth of all 

 the species known. 



The following is a synopsis of the genera which appear to 

 be recognizable : — 



I. Size not minute, skull-length at least 30 mm. Skull normal ; 

 anterior zygoma-root slanted, its anterior face looking downwai'ds 

 and forwards. (Sciurinse.) 

 A. Fur soft, not spinous. Palate not or little produced behind molars. 

 Postorbital processes Mell developed, directed outwards, near 

 middle of combined orbito-temporal fossa. 

 a. Size small or medium ; greatest skull-length less than 62 mm. 

 a^. Lower molars basin-shaped as in Sciurus. 

 a'. Skull normal, forehead flat. Anteorbital foramen in front 

 of level of premolars. 

 «*. Cheek-teeth -,% 1. Sciui-us. 



4 



♦ P. Z. S. 1893, p. 179 ; classification on p. 189. 



t P. Z. S. 1897, p. 933. 



J S. persicus, with only * cheek-teeth, but similar to true Sciurus in 

 all other character>i, should form a special subgenus, which might be 

 called Tenes. 



