LETTERS. 201 



united, and in not a few cases re-separated. I have during 

 the last 5 or 6 years been making a most laborious series of 

 experiments, by which I shall be able, I think, to demonstrate 

 the wonderful good derived from crossing, and I am almost 

 sure but shall not know till the end of the summer that I shall 

 be able to prove that the good is precisely of the same kind 

 which the adult individual derives from slight changes of 

 conditions. 



" With my sincere thanks for your interest in my work, I 

 remain, dear Sir, Yours very faithfully, 



"Ch. Dakwin." 



_J]ieJ ollowing letter is of ^reat interest in relatio n 

 tnrnf^yiy pro);^lfti iis of sexual Selection, protec tive 

 re semblance, mimicry, etc. :— - 



" Jan. 23, 1872. " Down. 



" Dear Sir — The point to which you refer seems to me a 

 very difficult one. 1** the comparison of the amount of 

 variability in itself would be difficult. 2"*^'^' of all characters, 

 colour seems to be the most variable, as we see in domes- 

 ticated productions. (3) I fully agree that selection if 

 long continued gives fixity to characters. We see the reverse 

 of this in the great variability of fancy races, now being 

 selected by rtian. But to give fixity, selection must be con- 

 tinued for a very long period : pray consider on this head 

 what I have said in the ^ngm^jibou^ the_ ya^iab^^ 

 characters developed in an extraordinary manner, in com- 

 parison with'^'the same characters in allied species. The 

 setScHon liitlBt also be for a definite object, and not for 

 anything so vague as beauty, or for the superiority of one male 

 in its weapons over another male, which can in like manner be 

 modified. This at least seems to me partly to account for the 

 general variability of secondary sexual characters, jn the case 

 jifmimetic insects, there is another element of doubt, as the 

 imitated form may be undergoing change which will_J)e 

 fol lowed by the iniitating forni. This la tter co nsideration 

 see ms to me , as remarked in my ' l^esceiit of_Manj^_to_y]T6w 

 mucli light onThow the process of imitation first began. 



