EVIDENCE FROM PALAEONTOLOGY 107 



side-branches of the equine stock, which, after a 

 longer or shorter duration, died out without leaving 

 descendants. One such branch was that of the 

 browsing horses, which were abundant through the 

 Miocene and early Pliocene of North America; they 

 retained the low-crowned teeth and the three-toed 

 foot until the end. The peculiar South American 

 forms were likewise side-branches, given off from the 

 main stem at a later date; one of these was a small 

 mountain horse, with very short feet, especially 

 adapted to climbing, like several of the existing 

 mountain antelopes. A very early branch, which 

 probably originated in Asia, diverged so widely from 

 the equine series that it is included in a different 

 family, is that of the Palaeotheres, which was abun- 

 dant in the Eocene of Europe, but did not extend its 

 range to the western hemisphere. 



We have thus traced the horses back to little 

 animals, whose structure is but remotely equine and 

 which are in all respects so unlike the modern species, 

 that hardly any palaeontologist would be bold enough 

 to connect them, were it not for the many inter- 

 mediate stages, which join these extremes in almost 

 unbroken continuity. No more illuminating example 

 of evolutionary development can be found among the 

 geological records of mammalian life than the history 

 of the horses, just because well-preserved specimens, 

 exemplifying all the important steps of change, have 

 been discovered. In turn, this was probably due to 

 the fact that, from the beginning to the end of their 



