ACCUMULATION OF DIFFERENCES. 93 



selective breeding do not tend to specific 

 variation, neither does their accumulation 

 go on indefinitely, for experience tells us 

 that the maximum accumulation of bene- 

 ficial differences may be attained within 

 three generations, and that retrogression 

 then takes place either the beneficial dif- 

 ferences disappear, or sterility bars further 

 accumulation. 



Galton, in his work on Hereditary Genius, 

 seems to have proved, by carefully collected 

 statistics, that eminent intellectual and also 

 great physical ability run in families, and 

 are therefore more or less hereditary ; but 

 we do not think his statistics present any ex- 

 ample of a father, son, and grandson possess- 

 ing abilities above the average of the family, 

 and in every case the eminence subsequently 

 disappears. 



"Ability," says Galton (p. 76), "in the 

 long-run does not suddenly start into exist- 

 ence and disappear with equal abruptness, 

 but rather it rises in gradual and regular 

 curve, out of the ordinary level of family 

 life. The statistics show that there is a 

 regular average increase of ability in the 

 generations that precede its culmination, 

 and as regular decrease in those that succeed 

 it. After three successive dilutions of blood. 



