662 MR. R. I. POCOCK ON SOME AFRICAN CATS [June 18, 



point between the shoulders to a transverse line in front of the 

 anterior rim of the ears on the crown of the head, groios forioards. 

 This area is defined on each side by a crest, and is marked 

 posteriorly by a single or double whorl on the shoulders. This is 

 a most noticeable character in all the skins I have examined. It 

 serves to distinguish the skins of F. aurata at a glance from those 

 of all the small or medium-sized African and Oriental species of 

 Felis. It is, however, paralleled in some South- American species 

 like F. 2Kcrdcdis and F. iigrina, in which its systematic value does 

 not appear to have been thoroughly worked out *. 



The second character is the shape of the mesopterygoid fossa 

 of the skull (text-fig. 175, p. 661), which is narrow and ellipti- 

 cally rounded in front, and thus differs from this fossa in all 

 other African species of Felis and approaches that of some of the 

 Oriental and South-American species. Indeed, I do not doubt 

 that F. aurata is to be classified in a group of which F. pardcdis 

 is one of the South- American, and F. temmincki one of the 

 Oriental representatives, and not with the groups exemplified in 

 Africa respectively by F. serval, F. caracal, and F. ocreata. 



As a woi-king hypothesis, it may be held that this group to 

 which F. aitrata belongs originated in the Europa?o- Asiatic 

 continent and migrated thence in a south-easterly direction as 

 far as Borneo. From the Oriental Region it passed into tropical 

 West Africa. Into America it made its way probably by the Alaskan 

 route and spread southwards through the continent as fai' as Chili 

 and Patagonia. There is as yet no reason to think that Felis 

 entered Africa at a sufiiciently early date to pass direct to 

 America by the transatlantic bridge, which is believed to have 

 joined these two continents together formerly. 



At the present time the distribution of this group is discon- 

 tinuous ; but thei'e are many known instances of afiinity between 

 the faunas of the Congo basin in Africa and of the Indo- 

 Malayan area of the Oriental Region; and also between the 

 faunas of the latter area and of the Neotropical Region. Hence 

 there is nothins: extravagant in the claim that Fells aurata is 

 nearly related to species now inhabiting those areas, despite its 

 distributional isolation. 



On Felis servalina Ogilhy. (Plate XXXVIII. figs. 3, 4.) 



Felis servalina Ogilby, P. Z. S. 1839, p. 4; Sclater, P. Z. S. 

 1874, p. 495, pi. Ixiii. ; Thomas, P. Z. S. 1888, p. 5; Lydekker, 

 Cats etc., Lloyd's Nat. Hist. pp. 135-136 (1896). 



Felis serval Elliot, Mon. Felidse, pi. xxvi., 1883 (in part). 



Trouessart (Cat. Mamm. 1906, p. 274), misled apparently by 

 Gray (P. Z. S. 1867, p. 272), cites F. servalina Ogilby as a synonym 

 of F. chrysothrix. This is an error. The type of F. servalina is 



* See 0. Thomas, Ann. Mag. Nat. Hist. (7) xii. pp. 235-237, 1903. This peculiarity 

 is verji- well shown in the figure of an example of F. tigrina which Schreber printed 

 as an illustration of T'ells onca Linn. (Saug. iii. pi. cii.). 



