62 TEETIAET VEETEBEATA OE THE EAT"DM. 



of Arsinoitherium, though in the absence of associated series there may be some 

 doubt about the exact proportions of the limb-bones. Text-fig. 37 is a view of the 

 pelvis and hind limb seen from the front, and shows the great width of the pelvis, 

 the proportions of the femur and tibia, and the structure of the tarsus. Both the 

 figures are about one- sixteenth natural size, the animal having stood about 1"76 

 metres (about 5 ft. 9 in.) at the withers and measured about 2-96 metres (9 ft. 9 in.) 

 from the snout to rump. 



The determination of the systematic position of the Barypoda is rendered specially 

 difficult from the circumstance that Arsinoitherium, the only member of the group 

 at present known, is a highly specialised terminal form of a series of which the 

 earlier terms are still undiscovered. Nevertheless some attempt may be made to refer 

 this animal to its place in the system, for although in many respects so specialised, in 

 others it retains primitive characters, and even where the degree of specialisation 

 is great, as in the case of the teeth, it may still be possible to arrive at a fairly clear 

 idea of the conditions that must have existed in the earlier types. 



From the account of the skeleton given above it will be seen that there is no doubt 

 that the Barypoda form a subdivision of the Ungulata, belonging to that section 

 of the order including the Elephants, Hyracoids, and Amblypoda, to which the 

 collective name Subungulates is sometimes applied. To determine the relations 

 of the Barypoda with the other Ungulates comparison must be made with (1) the 

 Proboscidea, (2) the Amblypoda, (3) the Hyracoidea, and (4) some of the South- 

 American Ungulates. The last of these comparisons is necessary because Ameghino 

 has suggested that the Ethiopian and S. -American regions were closely connected 

 in the later Cretaceous and early Tertiary times, and he considers that many of the 

 main groups of Ungulates (e. g. the Hyracoids and Proboscidea) originated in S. America 

 and subsequently migrated into Africa. It is not possible or desirable to discuss 

 these questions fully here, but some allusion must be made to them. 



Comparison of Arsinoitherium with the Proboscidea shows that in the structure of 

 the skull, in the form of the molars, and in the persistence of the complete series 

 of teeth without diastemata it differs entirely from any member of that suborder. 

 In the skeleton, on the other hand, there are some similarities, but these seem to be 

 merely parallel modifications due to great size and weight in the two groups, while, 

 on the other hand, differences of such fundamental importance exist {e. g. in the 

 structure of the tarsus) that any possibility of close relationship may be excluded. 



Comparison with the Amblypoda shows that many points of resemblance in the 

 skeleton exist, but, as in the case of the Proboscidea, these are probably nearly all 

 the results of parallel evolution and are modifications depending mainly on increasing 

 bulk and weight. On the other hand, the great similarity of the structure of the 

 tarsus in the two cases ^^cannot be so explained, but is probably a primitive character 

 derived from similar Condylarthrous ancestors in the two cases. In the skuU the 



