SOUTH AMERICAN EXTINCT HORSES. 7 



three or four inches in length. In Hippidium (as in Onohip- 

 pldmm, lig. 4), on the other hand^ these slits are about 10^ in. 

 long, while the nasal bones themselves are proportionately long 

 and slender. This clearly indicates that these extinct American 

 Horses had extremely elongated noses, not improbably forming a 

 kind of short trunk comparable to that of the Saiga Antelope. 



In that animal, as well as in its relative the Chiru Antelope of 

 Tibet, the increased size of the nasal chamber has been brought 

 about by a shortening instead of an elongation of the nasal bones, 

 but it is probable that in these two Antelopes and in the Hippid- 

 ium the purpose of the modification is the same. It has been 

 generally supposed that in the case of the Chiru the large size of 

 the nasal chamber is an adaptation to the respiratory needs of an 

 animal living at a very high elevation. In the case of the Saiga 

 such an explanation would, however, obviously not hold good; 

 and the real explanation in all three cases may perhaps be found 

 in a special adaptation to a desert life, the long nose serving as a 

 filter to prevent particles of sand reaching the organ of smell. 



As regards the rest of its skeleton, Hippidium is remarkable for 

 its short and stout limbs ; this being chiefly due to the excessive 

 shortness of the cannon-bones, which are also unusually wide, 

 with very stout splint-bones. Each limb terminates in a single 

 toe. These short limbs, coupled with the huge unwieldy head, 

 indicate that Hippidium had less speed than ordinary ponies. 

 There are only five lumbar, or ribless trunk, vertebras, as in the 

 Arabian Horse. 



Two other points of interest in connection with these peculiar 

 equine animals deserve brief reference. From the conformation 

 of the bones of the nasal region it seems certain that neither 

 Hippidium or Onohippidium can be derivatives from the genus 

 Equus, while it is still more evident that Equus cannot be 

 descended from Hippidium. Consequently, the reduction of the 

 digits from three in the ancestral Horses to a single one on each 

 foot has taken place independently in the two genera. The second 

 point is that if the wild Horses alleged to have been seen by Cabot 

 in Argentina in the year 1530 really were, as some suppose, 

 indigenous, they must have been either Hippidium or Onohippidium^ 

 and not Horses of the Old World type. With the evidence 

 afforded by the skins of the Patagonian Ground-Sloth as to the 



