No. 3.1 SKXUAL SELECTION TN SPIDERS. 151 



is no stronger than the weakest link in the chain. If males 

 have greater vitality than females, and if the males that have 

 the greatest vitality have also the greatest ornamentation ; and, 

 further, if the superior ornamentation of these males is due to 

 their greater activity, then the great vitality is the cause of high 

 color. "We have found the weak point in his argument was 

 the small amount of evidence that he was able to offer in sup- 

 port of each of the three propositions, so that the successive 

 steps in the argument grew weaker and weaker. Indeed, it 

 seemed to us that although many of his arguments were strik- 

 ingly ingenious, they all appeared to rest on very slender evi- 

 dence, or to admit of another interpretation. Even yet he had 

 not explained the distribution of the color over certain limited 

 areas of the body, and he was obliged to adopt the exceed- 

 ingly unsatisfactory hypothesis of Mr. Tylor that color and 

 ornament are a sort of by-product from the surfaces where 

 there is a great deal of activity. The most weighty objection to 

 this view is the almost entire lack of evidence in its support ; 

 perhaps it is as great an objection that it places so many strik- 

 ing and apparently important structures in the category of 

 useless organs.* "Tropical Nature " was published, and Mr. 

 Wallace argued at great length and with much ability in favor 

 of his views of sexual color, during Darwin's life, and it is of 

 the highest interest to learn what impression this reasoning 

 had made upon his mind. We close with a quotation from a 

 paper read at a meeting of the Zoological Society only a few 

 hours before his death : 



" I may, perhaps, be here permitted to say that, after hav- 

 ing carefully weighed, to the best of my ability, the various 

 arguments which have been advanced against the principle of 

 sexual selection, I remain firmly convinced of its truth." 



* We do not deny that some specific characters may he useless, but the demand that 

 Mr. Wallace's theory makes in this direction seems to us extravagant and beyond the 

 limits of reasonable probability. 



