Alfred Russel Wallace 



of subsidence in the islets to the westward. I remember, 

 however, speculating that there must have been some im- 

 migration during the glacial period from North America 

 or Japan; but I cannot remember what my grounds were. 

 Some of the plants, I think, show an affinity with Australia. 

 I am very glad that you like LyelPs chapter on Oceanic 

 Islands, for I thought it one of the best in the part which 

 I have read. If you do not receive the big photo of me in 

 due timCj let me hear. — Yours very sincerely, 



Ch. Darwin. 



The following refers to Wallace's article, ** A Theory of 

 Birds' Nests,'' in Andrew Murray's Journal of Travel, 

 i. 73. He here treats in fuller detail the view already pub- 

 lished in the Westminster Review for July, 1867, p. 38. 

 The rule which Wallace believes, with very few exceptions, 

 to hold good is, '' that when both sexes are of strikingly 

 gay and conspicuous colours, the nest is . . . such as to 

 conceal the sitting bird; while, whenever there is a strik- 

 ing contrast of colours, the male being gay and conspicu- 

 ous, the female dull and obscure, the nest is open and the 

 sitting bird exposed to view." At this time Wallace allowed 

 considerably more influence to sexual selection (in combina- 

 tion with the need of protection) than in his later writings. 

 See his letter to Darwin of July 23, 1877 (p. 298), which 

 fixes the period at which the change in his views occurred. 

 He finally rejected Darwin's theory that colours ^^ have 

 been developed by the preference of the females, the more 

 ornamented males becoming the parents of each successive 

 generation." {Bee " Darwinism," 1889, p. 285.) 



Down, Bromley, Kent, S.E, April 15, 1868. 



My dear Wallace, — I have been deeply interested by 

 your admirable article on Birds' Nests. I am delighted 

 to see that we really differ very little— not more than two 

 men almost always will. You do not lay much or any 

 stress on new characters spontaneously appearing in one 



212 



