584 PKOr. E. B. POTJLTON : NATTJKAL SELECTION 



This series, whicli we intend to keep separate, when more com- 

 plete, will aflFord very valuable evidence on this poiut. Already 

 it proves that the members of the groups which converge round 

 Limnas chrysippus, Amauris echeria, aud the black and white 

 species o^ Amauris in Natal are upon the wing together. This 

 evidence has been very kindly supplied me by Mr. Gruy A. K. 

 Marshall. I have similar, but less complete, evidence as regards 

 some of the Central and S. American groups. 



It will probably be conceded that the phenomena generally 

 are likely to exhibit the same relationship in time which has 

 been already proved to exist for many of them. This conclusion, 

 however, is a considerable difficulty in the way of the theory of 

 external causes as well as a further difficulty to the theory 

 of internal causes. As regards the latter, the time relationship 

 is an entirely unexplained coincidence ; as regards the former, it 

 is a coincidence which leaves much to be explained. It is diffi- 

 cult enough to believe that local forces could produce local 

 resemblance ; it is a further difficulty that the resemblances are 

 contemporaneous. If, as is probable, the forces are supposed to 

 act during larval life, they must include in their effects an 

 influence on the rate of growth and development, an adjustment of 

 the duration of stages delicate enough to bring the various species 

 into the phases in which the resemblance is shown at similar 

 times of the year. But such effects are entirely diflferent from 

 those which are manifest in the resemblance itself, and add 

 a further complexity to a result already shown to be so complex 

 that the theory of external causes fails to supply an interpreta- 

 tion (see Section 7, p. 576). 



But it has been shown in many cases, and is probably true in 

 al], that the time relationships between the species which exhibit 

 these resemblances are not confined to their appearance at the 

 same season of the year. They are such that they fly together 

 under those conditions of light which render the resemblance 

 visible to enemies. "When moths resemble butterflies, they are 

 mostly species which are as truly day-flying as the butterflies 

 themselves ; in other cases they are species which fly readily by 

 day when disturbed. Similarly with the species of various orders 

 which resemble Hymenoptera. The case of Coleoptera recently 

 suggested to me by Mr. Grahan is peculiarly interesting. It is 

 known in so many cases that beetles which are about by day 

 possess finely facetted eyes as compared with the larger fewer 



