CONTROL OF HEREDITY: EUGENICS 385 



supposed that the character in question was 

 "built up" by continued selection in one direc- 

 tion, and that the average development of the 

 character in all the offspring was thus in- 

 creased in successive generations to an indefi- 

 nite extent. It was this view as to the sup- 

 posed action of artificial selection which formed 

 the basis of Darwin's theory of natural 

 selection. 



On the other hand it has been known for a 

 long time that the limits of the possible im- 

 provement of any character by artificial selec- 

 tion are soon reached and that thereafter selec- 

 tion serves only to maintain the character at its 

 high level but not to advance it. The probable 

 explanation of this fact has been found only in 

 recent years. The researches of deVries, 

 Johannsen, Jennings, Tower, Pearl, and oth- 

 ers have shown that in some cases at least 

 selection merely isolates mutants or distinct 

 hereditary lines which are already present in a 

 mixed population but that it does not "build 

 up" characters nor produce new mutations; 

 in short it does not create the variations on 

 which it acts. 



Johannsen found that from a single species 



