Vol. 70.] VERTEBRATES FROM BRITISH EAST AFRICA. 175 



beyond the articulation and is rounded off, not forming a right 

 angle as in Brachyodus. The ectocondyle (ect., fig. 2, A) is pro- 

 minent, but the supinator ridge is not well marked. On the whole, 

 this bone may be regarded as belonging to a large Anthraeothere, 

 probably allied closely to Brachyodus. 



A second less perfect humerus (specimen b), of smaller size, 

 though otherwise similar, was collected ; and from beds of the same 

 age at Moghara I obtained another similar but still smaller humerus. 

 Probably, therefore, the fauna of this horizon in Africa will be 

 found to include a great variety of Anthracotheroid Ungulates, 

 differing much one from the other in size and in other respects, just 

 as is the case with the contemporary fauna of the Bugti Hills in 

 British Baluchistan. 



The dimensions (in centimetres) of these specimens are : — 



Specimen a. Specimen b. 



Length 53'0 43*9 (approx.) 



Greatest anteroposterior width of proximal 



end 18-2 



Greatest antero-posterior width of articular 



surface 11-0 10-0 



Antero-posterior diameter of middle of shaft. 9*1 



Lateral diameter of middle of shaft 6*5 



Greatest width of distal end 13'6 1 1*4 



Width of distal articular surface 8"5 7'3 



A very imperfect femur of an animal, at least as big as that to 

 which the largest of the above humeri belonged, probably should 

 be included here. It is imperfect at both ends ; but it can be seen 

 that the great trochanter was very large, with a deep digital fossa. 

 The shaft was oval in section, its long diameter measuring about 

 8*5 centimetres ; its posterior face bears a very strongly- marked 

 linea aspera. The distal end is too imperfect for description. 



A distal end of a right tibia (fig. 1, B & C, p. 166) indicates the 

 existence of an Anthraeothere still bigger than that to which the 

 larger humerus above described belonged. The distal articulation is 

 typically Artiodactyl in form ; but the internal malleolus is broken 

 away, and the rim of the articular surface is incomplete in some 

 places. The two concave surfaces for the reception of the astra- 

 galus are approximately equal in size ; on the inner border of the 

 inner astragalar surface, opposite the anterior end of the internal 

 malleolus, the surface is notched by a well-marked pit — this is also 

 obscurely indicated in a very closely similar tibia from Moghara, 

 but is not seen in Diplopus, Brachyodus yom'nyei, or Sus. On 

 the other hand, in the tibia of Hippopotamus there is a large 

 notch with a deep fossa at this point. This comparatively un- 

 important character, like the absence of the notch in the acetabulum 

 of Brachyodus, may point to the close relationship of these 

 African Anthracotheres with Hippopotamus. The outer angle of 

 the distal end of the bone is truncated by a roughly-triangular 



