THE EFFECTS OF CASTRATION 159 



CONCLUSIONS 



In conclusion it is evident that the secondary sexual 

 characters in four great groups, viz. mammals, birds, 

 Crustacea, and insects, are not on the same footing. 

 Their development depends on a different relation to 

 the reproductive organs in three of the groups, and is 

 independent of the reproductive organs hi the fourth. 

 It is not likely, therefore, that their evolution can be 

 explained by any one theory, even by one so broad in 

 its scope as that of sexual selection. 



If, for example, in the mammals a, mnrp vigorous 

 male, due to greater development of the testes, were 

 " selected " by a female, the chances are that his second- 

 ary sexual characters will be better developed than are 

 those of less vigorous males, but he is selected, not on this 

 account, but hppansp of his vigor. If a male bird were 

 "selected" on account of greater vigor, it does not 

 appear that his secondary sexual characters would 

 be more excessively developed than those of less vig- 

 orous males, provided that his vigor were due to the 

 early or greater development of the testes. If in 

 birds the male by selecting the female has brought 

 about the suppression of the male plumage, which is 

 their common inheritance, he must have done so by 

 selecting those females whose ovaries produced the 

 greatest amount of internal secretions which suppresses 

 male-feathering. Moreover, he must have selected, 

 not fluctuating variations, but germinal variations. 

 "Tn insects the development of the secondary sexual 

 characters is not connected with the condition of the 

 reproductive organs, but is determined by the complex 



