:> 



78 THE THEORIES OF EVOLUTION 



fact that one thing is developing in one direction or 

 another. 



No tendency can be inherited any more than any 

 other abstraction can be inherited; wljat can be in- 

 herited is a certain chemical combination, a certain 

 morphological structure, and such characters are ac- 

 quirable as they are and not as they might be. 



The workings of natural selection can therefore ex- 

 plain the persistency o f a use ful character" through 

 several generations but they fail to explain the grad- 

 ual development of that character. For this purpose 

 other factors must be "called into play ; heredity of the 

 effects of use and disuse and the uninterrupted action 

 of the environment working steadily in one and the 

 same direction. 



The preceding objections bear only upon details; 

 there are a number of others of a more general char- 

 acter, formulated from different points of view and 

 very different in their importance. We can divide 

 them up into two classes : Those relative to the nature 

 of variations upon which natural selection can exert 

 its influence, and those relative to a comparison be- 

 tween natural and artificial selection and to the close 

 bonds which unite natural selection and sexual selec- 

 tion. 



If natural selection is to favour an animal or a 

 plant, owing to its possessing certain characters, these 

 characters must fulfil a certain condition; they; must 



