XXIV.] 



NATURAL SELECTION. 311 



individuals are born every year than can possibly come to 

 maturity. Consequently there will be a "struggle for 

 existence." If auy spontaneous variation takes place Favour- 

 which gives an advantage in that struggle to its pos- ations'^wVu 

 sessor, the individual so favoured will have the best ije pre- 



1 -, ■ on ■ 1 i.1 !• served and 



chance of surviving and leaving ottsprmg, and the lavour- inherited. 

 able variation wUl probably be inherited. Thus, by pre- 

 serving the best and weeding out the inferior ones, the 

 race will be gradually improved. This process is exactly 

 the same in principle as that by which the breeds of 

 domestic plants and animals have been improved ; the 

 best are preserved, and the others are destroyed or not 

 permitted to breed. 



This short explanation will make it obvious how selec- 

 tion, whether by natural or by human agency, can improve 

 and perfect a breed. But it must be further explained 

 how selection can produce, from the same parent stock, 

 two or more races differing not only from the parent stock, 

 but from each other. For this piu-pose, however, it is only 

 necessary that selection should act on two different and 

 divergent lines of descent, and that the individuals selected 

 in the different lines should be selected for different qualities. 

 The various breeds of the dog, for instance, have chiefly Diver- 

 arisen by selection under domestication ; and they are f "a^acter, 

 distinct, because selection has been applied for different how x>ro- 

 purposes in the different breeds : in some for fidelity and seiectiou 

 sagacity ; in others, for power of hunting by scent ; in iu^do-^ 

 others, again, for fleetness. I believe this account of the paces, 

 origin of the breeds of dogs is historically true; it is 

 certainly possible ; and there cannot be the slightest doubt 

 that such has been the origin of the domestic breeds of the 

 pigeon,^ which vary from each other more than those of 

 any other domestic animal. Selection, constantly appKed 

 for different qualities, must necessarily give origin to dis- 

 tinct breeds, because the different qualities can be but 

 seldom found together in any unusual degree. Thus, 

 among dogs, if the qualities of keenness of scent and fleet- 

 ness are not correlated, which they certainly are not ; and ^ 



1 Darwin's Origin of Species, p. 2J. 



