172 Larva of S. ocellatus and its Food-plants. [Feb. 4, 



of the under sides of the leaves and their size, I am allnding to the 

 trees upon which the larvae mentioned in the table were found or were 

 bred. There may be varieties in many cases to which my descrip- 

 tions would not apply. 



8. Conclusion. 



A glance at the table printed above will at once show the 

 nature and amount of the evidence for the conclusion that the 

 larva of Smerinthus ocellatus maintains a colour-relation with the 

 food-plant upon which it was hatched, adjustable within the limits of 

 a single life, and that the predominant colour of the food-plant 

 itself is the stimulus which calls up a corresponding larval colour. 

 This may seem to be a long paper to prove the existence of such a 

 relation for a single species, but it must be remembered that in 

 accepting the conclusion now arrived at, we are admitting the exist- 

 ence of an entirely new resource in the various schemes of larval 

 protection by resemblance to the environment, and one which stands 

 on a very different level from all the others ; in these the gradual 

 working of natural selection has finally produced a resemblance — 

 either general or special — to something which is common to all the 

 food-plants of the larva, or to some one or more of them, the larva 

 being less protected upon the remainder ; but in this case the same 

 gradual process has finally given the larva a power which is (rela- 

 tively) immediate in its action, a power which enables the organism 

 itself to answer with corresponding colours the differences which 

 obtain between its various food-plants. And the action differs no 

 less from the superficially similar cases of much more rapid changes 

 in the colour of other organisms (amphibia, fish, &c.) corresponding 

 to the changing colours of their environment, for in such organisms 

 the external colours act as a stimulus which, through a nervous circle, 

 modifies the condition of existing pigments ; while in the larva it 

 is the amounts and kinds of vegetal pigments made use of and 

 larval pigments deposited that are affected. The influence, in fact, 

 makes itself felt by affecting the absorption and production of 

 pigments rather than their modification when formed; and such a 

 method of gaining protection is, as far as we yet know, unique in the 

 animal kingdom. And such a power is not confined to the species in 

 which its existence has been to some extent completely proved. 

 There are already proofs that many others can maintain a similar 

 colour-relation (see my former paper, and the references given by 

 Mr. Meldola in his translation to Weismann's " Studies in the Theory 

 of Descent," Part II), and I am sure that careful observation will 

 reveal many slight and protective differences among larvas of the 

 same species when found upon differently-coloured food-plants, and 

 will prove that this power is not at all uncommon among the great 



