Tlie Evidence from Sexual Characters. 235 



It will be seen that the evidence from this eonrce is, 

 as far as it goes, very similar to the evidence from hy- 

 brids. A reciprocal cross between two species furnishes 

 a means of analvzinof the influence of the two sexes, and 

 of distinguishing, to some slight degree, the effect of 

 each sexual element in heredity. The study of sexual 

 character gives us another means of doing the same thing 

 on a more limited scale. 



As each cell of the body may throw off gemmnles, there 

 is no way of showing that a variation in a part which is 

 alike in both sexes, is due to the transyjission of gem- 

 mules from the cells of one parent rather than from those 

 of the other, but the case is different with a part which 

 is more developed in one sex than it is in tne other. In 

 this case we should, according to our theory of heredity, 

 expect it to throw off gemmules most frequently in the 

 sex in which it is of most functional importance, and 

 as w^e suppose that there is an especial arrangement for 

 the transmission to the Qgg of those gemmules which orig- 

 inate in the male body, we can see that* an organ which 

 is most important in the body of the male is much more 

 likely to give rise to hereditary modification than one 

 which is most important, and therefore most prolific of 

 gemmules, in the female body. 



The history of secondary sexual characters is, there- 

 fore, what our theory of heredity w^ould lead ns to 

 expect, and no other explanation which has ever been 

 proposed fully accounts for all the phenomena. 



Instances of Female Modification. 



"We should not expect, however, to find sccondaiy sex- 

 ual characters exclusively confined to males, but simply 

 more general than they are in females, and as a matter 

 of fact we do meet with many cases wdiere the female 

 has been more modified than the male. 



