vi A THEORY OF BIRDS' NESTS 137 



of colour that this protection can be most readily obtained, 

 since no other character is subject to such numerous and 

 rapid variations. The case I have now endeavoured to illus- 

 trate is exactly analogous to what occurs among butterflies. 

 As a general rule, the female butterfly is of dull and incon- 

 spicuous colours, even when the male is most gorgeously 

 arrayed ; but when the species is protected from attack by a 

 disagreeable odour and taste, as in the Heliconidse, Danaidae 

 and Acrseidae, both sexes display the same or equally brilliant 

 hues. Among the species which gain a protection by imitat- 

 ing these, the very weak and slow-flying Leptalides resemble 

 them in both sexes, because both sexes alike require pro- 

 tection, while in the more active and strong-winged genera 

 Papilio, Pieris, and Diadema it is generally the females only 

 that mimic the protected groups, and in doing so often become 

 actually more gay and more conspicuous than the males, thus 

 reversing the usual and in fact almost universal characters of 

 the sexes. So, in the wonderful Eastern leaf-insects of the 

 genus Phyllium, it is the female only that so marvellously 

 imitates a green leaf ; and in all these cases the difference can 

 be traced to the greater need of protection for the female, on 

 whose continued existence, while depositing her eggs, the 

 safety of the race depends. In Mammalia and in reptiles, 

 however brilliant the colour may be, there is rarely any differ- 

 ence between that of the sexes, because the female is not 

 necessarily more exposed to attack than the male. It may, I 

 think, be looked upon as a confirmation of this view, that no 

 single case is known either in the above-named genera 

 Papilio, Pieris, and Diadema or in any other butterfly, of a 

 male alone mimicking one of the Danaidse or Heliconidaa 

 Yet the necessary colour is far more abundant in the males, 

 and variations always seem ready for any useful purpose. 

 This seems to depend on the general law that each species 

 and each sex can only be modified just as far as is absolutely 

 necessary for it to maintain itself in the struggle for existence, 

 not a step further. A male insect by its structure and habits 

 is less exposed to danger, and also requires less protection, 

 than the female. It cannot, therefore, alone acquire any 

 further protection through the agency of natural selection. 

 But the female requires some extra protection, to balance the 



