INTRODUCTION. xix 



under one of these two heads ; at any rate it appears the male is small 

 and pale in this group by special selection in relation to some special 

 habit. At the same time such species as B. neustria falling in Class 3, 

 actually belong to Class 1, the pigmentation of the male, though pale etc., 

 being richer than that of the ? . As to the richer colours in the males of 

 Class 1 it may occur in this way. Each individual succeeds in elabora- 

 ting so much colour. The male generally being smaller is able to con- 

 centrate it more, the larger female has to spread it, e.g., in one or two 

 species of the Hybernidce that have measureable wings as $ 's, these are 

 usually much darker than those of the males, and it was very remarkable 

 that when Dr. Chapman and I searched for Diurncea fagella this year 

 we found the females darker than the males. This looks as if the male 

 could dispose of the excess of pigment in his larger wings the female 

 being unable to do so. It must be understood however, that it is not 

 intended here to suggest that each ovum during its life produces so 

 much colour, and that when it is decided that it is to be say a ? (of large 

 size) it has then to spread that colour over a larger area, but that such 

 a tendency is inherent ; probably when the female first grew a larger 

 wing, the membrane was slightly expanded, without increasing the 

 number of scales and therefore became paler than the male, and vice 

 versa in those females with smaller and darker wings. In Class 2 where 

 the scales in the $ become more concentrated, one can understand that 

 if " natural selection " found some advantage in this the female would 

 become larger and paler constitutionally. In Class 2 also, as Euthemonia 

 russula, Agrotis cinerea, Agrotis puta, Rusina tenebrosa, Lithosia pyg- 

 mceola etc., although we find the female smaller and darker we have 

 to bear in mind that the paler males, like many butterflies (Lyccena) 

 &c., are in reality more richly coloured and have a greater amount of 

 pigmentation, and if we accept Wallace's theory that the darkness of 

 female coloration in these butterflies is due to " natural selection," it 

 becomes much more easy to accept the same theory (and I think it is 

 really the true one), with regard to these moths ; and if the males here 

 are really more richly coloured although in reality paler, it quite 

 agrees with the physiological explanation of the origin of the matter 

 which I propose offering. Not that any general rule is likely to meet 

 each individual case but certainly a general explanation appears now 

 to be becoming quite possible. It is quite probable that once the 

 colour has been actually set up as a distinct secondary sexual cha- 

 racter as in Bombyx quercus, B. rubi etc., and the habits of the sexes differ 

 very much, " natural selection " steps in, and the sexes tend to vary in- 

 dependently, almost as if they were different species, but less readily 

 no doubt, each attempting as it were to carry the other with it. 

 Summarising then, I think that Classes 1, 4 and 3 (in part) have 

 almost the same origin. Class 3 (in part) consists of special cases. 

 Class 2, females darker because smaller, i.e., larger males require 

 more energy for wing membrane. Now it appears to me that the 

 explanation of these " secondary sexual characters " relating to colour, 

 are to a certain extent and at any rate in particular instances, ex- 

 plicable on physiological grounds. Every field naturalist is aware 

 that the larvae of the sexes are not very different in size and that 

 large larvae frequently produce males, and small larvae, females. It 

 is also now fully recognised that colour is probably the result of sur- 



