330 'DESCENT OF MAN '—EXPRESSION. [187L 



that I know is that I did my best. With familiarity I think 

 naturalists will accept sexual selection to a greater extent 

 than they now seem inclined to do. I should very much like 

 to publish your letter, but I do not see how it could be made 

 intelligible, without numerous coloured illustrations, but I will 

 consult Mr. Wallace on this head. I earnestly hope that you 

 keep notes of all your letters, and that some day you will 

 publish a book : * Notes of a Naturalist in S. Brazil,' or some 

 such title. Wallace will hardly admit the possibility of 

 sexual selection with Lepidoptera, and no doubt it is very 

 improbable. Therefore, I am very glad to hear of your cases 

 (which I will quote in the next edition) of the two sets of 

 Hesperiadae, which display their wings differently, according 

 to which surface is coloured. I cannot believe that such dis- 

 play is accidental and purposeless. . . . 



No fact of your letter has interested me more than that 

 about mimicry. It is a capital fact about the males pursuing 

 the wrong females. You put the difficulty of the first steps in 

 imitation in a most striking and co?ivi7icing manner. Your 

 idea of sexual selection having aided protective imitation 

 interests me greatly, for the same idea had occurred to me in 

 quite different cases, viz. the dulness of all animals in the 

 Galapagos Islands, Patagonia, (S:c., and in some other cases ; 

 but I was afraid even to hint at such an idea. W^ould you 

 object to my giving some such sentence as follows : '' F. 

 Miiller suspects that sexual selection may have come into 

 play, in aid of protective imitation, in a very peculiar manner, 

 which will appear extremely improbable to those who do not 

 fully believe in sexual selection. It is that the appreciation 

 of certain colour is developed in those species which fre- 

 quently behold other species thus ornamented." Again let 

 me thank you cordially for your most interesting letter. . . . 



