106 MOSTLY MAMMALS 
larger toes are marked by a longitudinal groove repre- 
senting the cleft of those of the latter; and as in both 
groups the middle toe is the largest, there is no reason 
why the ant-eaters should not trace their origin to these 
same pigmy ground-sloths or a closely allied type. In 
this case the specialisation has resulted in a lengthening 
of the skull and the loss of the teeth, the hind-foot having 
retained more or less of the primitive type. Here like- 
wise we must notice that the resemblance presented by 
the skull of the scelidotheres to that of the ant-eaters 
must be regarded as an instance of parallel development. 
From the structure of their teeth, the ground-sloths 
were evidently pure vegetarians; and the same may be said 
of the sloths, which are animals specially modified for the 
exigencies of an arboreal existence. On the other hand, 
the ant-eaters, as their name implies, have given up a 
vegetable diet and taken to living on ants, and to this may 
be attributed their total loss of teeth. Should germs of 
teeth ever be found in their jaws during an early stage of 
existence, I venture to predict they will approximate in. 
structure to the teeth of the ground-sloths, 
I cannot conclude without saying a few words as to the 
probable mode of life and external appearance of ground- 
sloths. The Patagonian specimens have shown that, like 
sloths and ant-eaters, they were clothed with a thick 
covering of coarse hair. Further, from their massive 
proportions, and also from their kinship to the sloths, 
it is most likely that ground-sloths were as slow and 
deliberate in their movements as the latter. That such 
monstrous creatures could not have existed in a treeless 
country like the Argentine Pampas has been already pointed 
out, and we may hence assume that in the days of the 
ground-sloths Argentina was much like what Brazil is at 
