Vol. 70.] VERTEBRATES FROM BRITISH EAST AFRICA. 16? 



fossa immediately in front of the surface for the outer condyle oi 

 the femur, not seen in the recent tibia. The posterior face of the 

 upper end of the shaft is deeply concave, as in the African elephant ; 

 but about the middle of the outer rim of the concavity there is a 

 rugose surface, wanting in that species. The inner face of the 

 upper end of the shaft is, moreover, natter, and bears a deep 

 roughened pit for muscle-attachment. Only one tibia of JSle^has 

 africanus was available for comparison, so it is quite possible that 

 some of the above-noticed differences may be merely individual 

 variations. 



If the tibia ascribed by Graudry to Dinotherium really belongs 

 to that animal, as there seems no reason to doubt, then the tibia 

 now under discussion must belong to some other Proboscidean, the 

 lower end of the shaft showing none of the antero-posterior com- 

 pression seen in Gaudry's specimen. It is possible, therefore, that 

 this specimen may indicate the presence in these deposits of a 

 species,, probably a Tetrabelodont, at present otherwise unknown. 



The dimensions (in centimetres) of this specimen are: — 



Length without distal epiphysis 54*6 



Width of the proximal articular surface . . . 17*5 

 Width of the shaft at its narrowest 8*6 



Another portion of a much bigger tibia was collected in Bed 31. 

 This seems to indicate the presence of an animal considerably larger 

 than Dinotherium hobleyi, the fragment agreeing roughlv in size 

 with the corresponding bone of an African elephant standing about 

 8 feet at the shoulder. The inner side of the astragalar surface is 

 incomplete, but what remains shows that the form of the astragalus 

 was the same as that of the astragalus of Dinotherium described 

 below — its tibial surface being gently convex from before back- 

 wards, but with no median depression from side to side, a point 

 which distinguishes it from the astragalus of Tetrahelodon. The 

 fibular surface is preserved ; its line of union with the astragalar 

 surface is not directly antero-posterior as in Elephas, but is directed 

 somewhat inwards and backwards— in this the bone agrees with 

 the tibia of Dinotherium figured by Graudry. The fibular facet is 

 relatively smaller than in Elephas ; it is triangular in outline, and 

 looks downwards and outwards, making a more obtuse angle with 

 the fibular surface than in Elephas. On the whole, there does not 

 seem to be much doubt that this bone belongs to a species of 

 Dinotherium rather bigger than D. hobleyi. 



A left astragalus of a Proboscidean was collected from Bed 31 at 

 Nira. This specimen is much abraded, especially about its anterior 

 and posterior borders, and consequently the exact form of the 

 articular surfaces cannot be determined. The tibial surface is com- 

 paratively narrow from before backwards, and is gently convex in 

 the same direction ; it is not concave from side to side, and in this 

 respect is distinguished from the astragalus &h Tetrahelodon angusti- 

 dens, which might be expected to occur in these beds. This absence 



Q. J. G. S. No. 278. x 



