SCHWARZ — ^EUROPEAN FOSSIL HORSES 
155 
both by him and all those who have examined it, has six lumbar verte- 
brae. I should think that the original ‘‘ Celtic’^ pony was of northern’’ 
blood, and that the ‘‘southern” characters presented by a number 
of the present ponies are entirely due to mixture with horses imported 
from the Mediterranean region within historic times. 
The remarkable fact that in Europe alone both species of horse were 
found together, will be understood by the following zoogeographical 
deductions. In the later Tertiary, Europe had a land connection with 
North Africa as well as western Asia, immigration being possible both 
from the East and South. Thus Equus stenonis reached Europe from 
northern Africa as a representative of the western or Mediterranean, 
Equus caballus from central Asia with the eastern or Asiatic component 
of the Pliocene steppe fauna. During the Glacial Period the greater 
part of both faunse disappeared from central Europe. But when the 
ice had made its retreat, the Mediterranean landbridge was no longer 
in existence, and reimmigration was possible only from the East. In 
postglacial times Equus cabatlus ranged again all over northern and 
central Asia and the greater part of Europe and has given rise, through 
its various geographical subspecies, to the primitive domestic races 
of Russia and Asia and the heavy draught horses of western Europe. 
Equus stenonis, however, was held back in northern Africa where it 
has become the ancestor of the thoroughbred stock of Barbs and 
Arabs.^^ 
Frankfort a. M., Germany, 
2^ It is interesting to note here that a tooth from Wadi Haifa, Nubia, described 
by Lydekker (Quart. Journ. Geol. Soc., XLIII, pp. 161-63; (1887), as nearly re- 
lated to E. sivalensis, appears to me quite different. If conclusions may be 
drawn from a single specimen, it seems to approach E. quagga more than any- 
thing else. 
