48 Mr. O. Thomas on the 



causing the muzzle in that sex to be conspicuously more 

 developed than in tlie female. This difference has been 

 taken again and again for a racial character, but is really 

 only a sexual one. 



As to size, those from the north of tlie area — Rhodesia — 

 averao-e larger than the more southern ones, but the difference 

 is not very great; while of otlier cranial characters tliere do 

 not seem to be any at all, and the teeth are alike throughout. 



Colour-characters are therefore the only means of sorting 

 the races, and on this account I should consider all to belong 

 to but a single species — G. crassicaudatus — witli several 

 local subspecies. 



The type-locality of crassicaudatus itself, not known at the 

 time of description, has first to be settled, and on this I 

 should accept the first authoritative identification of specimens 

 and statement of locality, which were made by Peters in 1852. 

 He says that Geoffroy's type-specimen " stammt oline Zweifel 

 ebenfals aus Mossambique her," and identifies with it his 

 own specimens from various places, of which Quelimane is 

 the first to be mentioned. I should therefore take that as the 

 type-locality. 



As a consequence, Gray's hirlcii from the same place 

 becomes an absolute synonym of crassicaudatus, and ins type- 

 specimen is a topotype of it. 



With regard to the next name on the list, garnetti of 

 Ogilby, commonly assigned to " Natal," I find that tiie skull 

 of the type — which is in the Museum, in spite of Elliot's 

 assertion to the contrary — is distinctly too small for any 

 S.-African Galugo at all, while it exactly agrees witii two 

 from Zanzibar Island (coll. Sir J. Kirk and C H. B. Grant), 

 representing G. agisymhanus, Coquerel. 



As the locality of the ty))e was nowhere recorded, and was 

 definitely stated in the MS. Catalogue of the Zoological 

 Society's Museum to be " unknown," Zanzibar is as likely a 

 locality as Natal, and I therefore propose to accept it for the 

 Galago of that island. It would thus be the first of the 

 E.-African series of names, and would antedate Coquerel's 

 agisgmJianus. 



The following are the four subspecies of crassicaudatus 

 which appear to me to be recognizable, taking them from 

 north to south : — 



1. G, crassicaudatus monteiri, Gray. 

 Wholly grey. 



Angola {Monteiro, Ansorge), N. Rhodesia and Angoniland 

 [Xeave, Melland, Mrs. Colville). 



