386 Earwin, and after Larwin, 



battle, in fact, more nearly approaches the law of 

 natural selection-; seeing that it expresses the natural 

 advantages of brute force in the struggling of rival 

 animals, and so frequently results in death of the less 

 fitted, as distinguished from a mere failure to propa- 

 gate. Now against this doctrine of the law of battle, 

 and the consequences to which it leads in the superior 

 fighting powers of male animals, no objection has 

 been raised in any quarter. It is only with regard to 

 the other aspect of the theory of sexual selection 

 or that which is concerned with the superior em- 

 bellishment of male animals that any difference of 

 opinion obtains. I will now proceed to give the 

 main arguments on both sides of this question, 

 beginning with a resttme' oi the evidences in favour of 

 sexual selection. 



In the first place, the fact that secondary sexual 

 characters of the embellishing kind are so generally 

 restricted to the male sex in itself seems to constitute 

 very cogent proof that, in some way or another, such 

 characters are connected with the part which is played 

 by the male in the act of propagation. Moreover, 

 secondary sexual characters of this kind are of quite 

 as general occurrence as are those of the other kind 

 which have to do with rivalry in battle ; and the former 

 are usually of the more elaborate description. There- 

 fore, as there is no doubt that secondary sexual char- 

 acters of the one order have an immediate purpose to 

 serve in the act of propagation, we are by this close 

 analogy confirmed in our surmise that secondary sexual 

 characters of the other, and still more elaborate, order 

 are likewise so concerned. Moreover, this view of their 

 meaning becomes still further strengthened when we 



