XXIV.] NATURAL SELECTION. 313 



Here occurs a difficulty wliich Darwin appears not to How will 

 see, but wliicli H. Spencer thinks fatal to the theory as "aces^be 

 applied to the highest animals. Among animals that kept 

 imite for each birth and wander much, will not variations, 

 no matter how favourable, be lost and merged in the second 

 generation, by the union of the individual possessing the 

 favourable variation with others that are without it ? ^ It 

 is quite obvious that the domestic breeds of animals could 

 not be kept distinct, if free inter-breeding between different 

 varieties were permitted.^ I am inclined, however, to think 

 that this objection is sufficiently answered by the fact, that 

 most varieties in a state of nature are local varieties : so By being 

 that while natural selection is forming a variety, suited to °°^ ' 

 the locality, by preserving the individuals suited to it and 

 destroying the rest, this action will not be much interfered 

 with by the race getting crossed with other varieties. 

 Dissimilar habits of life will also probably have con- llow wild 

 siderable effect in keeping the individuals of different ^^^^ ^^^ 

 races apart : and, as we have seen, races are kept apart by distinct, 

 their instincts among the higher animals, and probably 

 among all animals. 



I believe that natural selection will account for much 

 which self-adaptation will not account for.^ To take one 

 very curious instance : so far as I can see, natural selec- 

 tion is capable of giving origin to the bat's wing. Once Origin of 

 the wing is formed, self-adaptation will no doubt give the *^.® ^^^ ® 

 necessary power to the wing-miiscles ; but self-adaptation 

 cannot, I think, have any tendency to produce that ex- 



1 Spencer's Principles of Biolog}', vol. i. p. 454. 



2 " The prevention of free crossing, and the intentional matching of indi- 

 vidual animals, are the corner-stones of the breeder's art. No man in his 

 senses would expect to improve or modify a breed in any particular 

 manner, or keep an old breed tnie and distinct, unless he separated his 

 animals." (Darwin's Variation under Domestication, vol. ii. p. 85.) 

 " Cats, which from their nocturnal habits cannot be selected for breeding, 

 do not yield distinct races in the same country." (Ibid. vol. ii. p. 236.) 



* I may here mention Darwin's belief, that what is called the acclima- 

 tization of a race of plants is really, in some cases, the production of a 

 new and hardier race by selection. It is obvious that natural selection may 

 operate among plants gi-own in a garden but not under glass, by killing the 

 least hardy and preserving the most so. (Origin of Species, p. 169.) 



