238 SEXUAL DLMORPHISM 



as danger is concerned, would clearly have been quite as well 

 off if they had mimicked their mates instead of imitating 

 species with which they had nothing to do. All the species 

 of Danais above mentioned are mimicked by other species as 

 well as by the varieties oiPapilio cenea, and, further, the latter 

 varieties are connected by intermediate forms. Even sup- 

 posing that we admit the utility of the existing mimicry, the 

 question in regard to its origin is whether the individual 

 variations in the two sexes of P. cenea were originally the 

 same, that is, whether the survival of particular individuals 

 was the determining factor. This essential point is entirely 

 ignored in the theory of selection as the explanation of the 

 mimicry, or rather we may say that the original similarity 

 of variation tendencies and directions is assumed. But in 

 support of this very important assumption we have no evi- 

 dence whatever. On the other hand, if the original variations 

 were not similar, if the male has never varied in the direction 

 of the disguise which the females have assumed, then selection 

 or survival is no explanation. The explanation lies in the 

 difference of variation in definite directions. Even if we 

 had no evidence to show that the females are subjected to 

 different conditions from the males, conditions which might 

 have determined the direction in which they have been 

 modified, at least it is certain that the selection theory in this 

 case is founded on an unsupported assumption, and as it is 

 probable that the habits and conditions of life of the sexes 

 are different, and that those of the females are similar to those 

 of the species they mimic, we have good reason to hold it 

 probable that in this case particular conditions have caused 

 variations in definite directions in one sex, and not in the 

 other. 



But it so happens that we have very distinct evidence of 

 the subjection of the two sexes to different conditions, 

 evidence recorded by Sir Roland Trimen, F.R.S., the original 



