4.24 THE CAUSES OF THE XI 
it is quite certain that the hybrids are often abso- 
lutely infertile one with another. 
Here is a feature, then, great or small as it may 
be, which distinguishes coined species of animals. 
Can we find any approximation to this in the 
different races known to be produced by selective 
breeding from a common stock? Up to the 
present time the answer to that question is abso- 
lutely a negative one. As far as we know at 
present, there is nothing approximating to this 
check. In crossing the breeds between the Fan- 
tail and the Pouter, the Carrier and the Tumbler, 
or any other variety or race you may name—-so far 
as we know at present—there is no difficulty in 
breeding together the mongrels. Take the Carrier 
and the Fantail, for instance, and let them repre- 
sent the Horse and the Ass in the case of distinct 
species; then you have, as the result of their breed- 
ing, the Carrier-Fantail mongrel,—we will say the 
male and female mongrel,—and, as far as we know, 
these two when crossed would not be less fertile 
than the original cross, or than Carrier with Car- | 
rier. Here, you see, is a physiological contrast 
between the races produced by selective modifica- 
tion and natural species. I shall inquire into the 
value of this fact, and of some modifying cireum- 
stances by and by; for the ren I merely put 
it broadly before you. 
But while considering this question of the limi- 
tations of species, a word must be said about what 
