412 Mr. O. Thomas — Xotes on 



— the genotype — from tlie great mass of the species usually 

 assigned to that genus, and proposed to recognize the two 

 groups so distinguished as subgenera, accepting for the 

 second subgenus the name Cryptomys from Dr. Gray^s paper 

 of 1864 *. _ 



This division appears to me thoroughly sound, especially 

 as, in addition to the tooth-characters mentioned by de VVinton, 

 there is a material difference in the structure of the posterior 

 palate, Cryptoniys having this quite normal, while in Geo- 

 rychim (s. s.) the capsules of the incisor-roots are extended 

 backwards into the pterygoids in a manner approaching that 

 found in Helioplwbius. 



On the whole, therefore, it would appear to be advisable to 

 recognize Cryptomys as a full genus distinct from Geovychus, 

 the great majority of the described species falling into it. On 

 the other hand, Gray's Ccetomys is not worthy of any sort of 

 recognition. 



Restricted Georychus — the capensis group — inhabits the 

 extreme south of Africa, ranging northwards to Namaqualand 

 on the west and the Transvaal in the east. Cryptomys only 

 overlaps it along a comparatively narrow zone, but ranges 

 northwards from Knysna and Natal over the whole of the 

 Ethiopian region to Nigeria and Togoland. 



As regards species, the genus Cryptomys is extraordinarily 

 uniform both externally and cranially, the many forms 

 described only differing, apart from size and a limited 

 range in colour, by characters of but little importance, such 

 as slight differences in the shape of the nasals, the relative 

 lengths of premaxillary processes and the nasals, and the 

 shape of the anteorbital foramina. Moreover, these diffe- 

 rences, such as they are, are curiously inconstant in the 

 group, almost any good series from one locality being liable 

 to contain individuals with the characters supposed to be 

 peculiar to other species. This is notably the case with the 

 apparently important character of the anteorbital foramina, 

 which in some cases may be high and lunate, with slender 

 boundaries, or small and subcircular, with thick ones, in 

 specimens taken in the same localities and obviously of the 

 same species !• Young specimens tend to be of the former 



* P. Z. S. 1864, p. 123. 



t The type-skull of G. niimodi, de "Wiut., has a subcircular foramen 

 on oue side and a high one on the other. And anyone to whom the 

 British Museum collection is accessible should compare 4. 3. 1. 88 and 9:?, 

 both from Vredefort Eoad {Barrett-Hamilton ) , which have the extremes 

 of difference in the anteorbital foramina, and 5. 5. 1. 14 with 14. 5. 4. 20, 



