1907.] EARS OF AFRICAN ELEPHANTS. 403 



Apart from the fact that all the other skulls of African 

 Elephants (some old and some young) in the British Museum are 

 of the same general type, such differences cannot be attributed 

 either to age or to individual variation. Indeed, if the Albert 

 Nyanza skull had been found in a fossil condition without the 

 teeth, there would be little hesitation on the part of naturalists 

 in regarding it as specifically distinct from the African Elephant. 



The Albert Nyanza skull, it may be added, is quite different 

 from that of the Rhodesian Elephant referred to E. a knochen- 

 haueri, and apparently also differs as markedly from Di-. Matschie's 

 description of that of the type specimen of the latter. The ear, 

 moreover, in Mr. Tomkins's photograph is certainly different from 

 that of the Rhodesian Elej^hant in the British Museum. A 

 small telephotograph of an elephant killed by Major Powell 

 Cotton on the Semliki side of the Albert Nyanza, recently pub- 

 lished in the ' Illustrated London News,' may belong to this race, 

 and appears to have an altogether peculiar type of ear. Under 

 these circumstances the name of Elephas africanus albertensis, 

 which I suggested in the ' Field ' for the Albert Nyanza Elephant, 

 may stand. 



The most noteworthy feature about the skull of E. a. cdbertensis 

 is its striking resemblance in contour to that of E. 2)l(f'n'\frons 

 of the Pliocene of the Siwalik Hills of Northern India, as will be 

 apparent by a comparison of the figure here given (text- fig. 121) 

 with the one of the fossil skull re]3resented in plates ix. and x. of 

 Falconer and Cautley's ' Fauna Antiqua Sivalensis.' E. plmiifrons 

 is a member of the " loxodont " group, with a ridge-formula very 

 similar to (although slightl}^ higher than) that of E. africamtSy 

 and one disjolaying primitive features in the development of suc- 

 cessional premolars. 



Although accepting the view of the African origin of the Pro- 

 boscidea, I am of opinion (from the occurrence in those countries 

 of the intermediate " stegodont " group) that the evolution of the 

 Mastodons into Elephants probably took place somewhere in 

 South-eastern Asia ; and the similarity presented by the skull of 

 E. africanus albertensis to that of E. planifrons seems to suggest 

 that the African Elephant may be the descendant of the fossil 

 Indian species. The many remarkable instances of affinity 

 between the Pliocene mainmals of India and the modern mammals 

 of Africa (among which the relationship of Canis curvipalatus to 

 Otocyon is one of the most noteworthy), indicate that there is no 

 inherent improbability in such an ancestry for Elephas africanus, 

 of which, on this view, E. a. albertensis is the most generalised 

 representative. 



[Postscript. — A second Elephant's head in Mr. Rothschild's 

 museum, from some part of East Africa, appears to indicate an 

 animal nearly allied to the one from Swaziland of which the head 

 is represented in text-fig. 110 (p. 389).] 



