76 THE THEORIES OF EVOLUTION 



Delboeuf's law would only hold true if variations 

 were due to a permanent factor of modification whose 

 action would only bear upon certain individuals 

 which would remain affected forever. In fact, in- 

 stances of such a processus have hardly ever been ob- 

 served. 



The question at issue was then as to whether a spe- 

 cies would in the course of time give rise to more 

 and more variations or whether, on the contrary, con- 

 servative influences would predominate and preserve 

 the purity of the original type. Still, when a varia- 

 tion appears, it must, as it is at first a slight and in- 

 dividual one, increase gradually in the course of 

 several generations until it assumes the character of a 

 new species. This is the way Darwin conceived that 

 processus ; slight accidental variations accumulating in 

 the course of generations which transmit them to other 

 generations. This processus of accumulation is one of 

 the postulates of the selection theory and does not 

 seem ever to have been questioned by any of its oppo- 

 nents. And yet, why should a character, whatever 

 useful purpose it may serve, be more deeply marked 

 in the offspring than in the parent? Unless we ex- 

 plain the fact by the inheritance of the effects of use 

 or disuse and thus introduce a Lamarckian idea well 

 beyond the limits of natural selection, we cannot un- 

 derstand why it should be so. 



Let us suppose that among the short-necked an- 



