254 
fragment has been detected, though the number of species is so small 
that no importance can be attached to this fact. 
Seeing that these mammals all belong to existing genera, and 
that, with the exception of the elephant, they differ but comparatively 
little from Recent Ethiopian species, it is obvious that the fossils can- 
not have any very great geological antiquity, and probably they should 
be referred to the later Pliocene. So far, however, as the mammals 
themselves are concerned, they might almost equally well be regarded 
as early Pleistocene. Unfortunately they give us little information 
concerning the origin of the modern South African fauna, or the 
former relations of the region to other continents. 
ARTIODACTYLA. 
HIPPOPOTAMUS. 
Remains of this genus are the most numerous objects in the 
collection, and of one individual a considerable part of the skeleton 
is represented. This specimen comprises several thoracic, lumbar, 
and sacral vertebrae, and a number of ribs, the pelvis entire, fragments 
of several bones of the fore and hind limbs and part of the carpus. 
All of these agree closely with the corresponding parts of the recent 
H. amphibius in size and characteristics, and may very well have 
belonged to that species. A number of isolated teeth, however, indi- 
cate a species nearly allied to, but different from the modern one. 
Unfortunately there is no method at present of determining whether 
the bones and teeth belong to the same species, and thus it remains 
uncertain whether the collection contains one species of Hippo- 
potamus or two. 
Hippopotamus ponderosus, sp. nov. 
(Plate XVI., Figs. 1-3.) 
As the type of this species I have selected an isolated third lower 
molar of the left side ; the other teeth in the collection, which are less 
characteristic, may or may not have belonged to the same animal. 
This does not apply to a very much worn first lower molar, which 
evidently belonged to an older individual. 
The third lower molar (M. 3) is very large, being but slightly 
inferior in antero-posterior length to the corresponding tooth of the 
largest skull of H. amphibius, with which I have been able to compare 
it, and in proportion to that length is somewhat narrower. The tooth 
was evidently freshly erupted, and shows but slight sign of abrasion. 
In construction this tooth differs in several respects from that of the 
modern species, though the differences are not of sufficient importance 
to require a generic separation. As in H. amphibius, the four main 
cusps of the crown are arranged in two transverse pairs, but the two 
cusps of each pair are divided from each other by a deeper fissure. 
