334 Mr. P. R. Lowe on Coloration as a Factor in 



"The relationship of colour-pattera to genera-splitting and 

 genera-lumping/') 



Finally, so far as general considerations go, there is the 

 question of 



(5) The relationship of colour-pattern to sex. 



It may be argued that the fact that the male is so often 

 more brightly coloured than the fenoale, is a difficulty which 

 lessens the value of colour-pattern as a factor in generic 

 differentiation. The female, too, as we all know, may in 

 certain instances usurp the brighter plumage of the male. 

 Mr. Rothschild has given me an instance, in the case of 

 the genus Oreomystes of the family Drepanididse, which he 

 considers an example of the difficulties of applying the factor 

 of colour as a <ieneric character. 



In Oreomystes, while the rule is that both male and female 

 are green, we find in O. flammea that the male is all red 

 and female not all red. Well, it seems to me that as far 

 as I understand the problem, the fact that the male 

 generally has a brighter and different colour-pattern from 

 the female, or that the female may take on masculine 

 characters, or that in any given genus the male of certain 

 species may be found with an exceptional type of coloration, 

 proves very little more than this, viz., that, for some reason 

 or other, some restraining factor which, as a rule, is operative 

 in the female is not so operative in the male. 



The fact that, in one species of the genus Oreomystes, the 

 males are red instead of green simply points to the fact that 

 in the Oreomystes germ-cell there was always a latent 

 combination of factors, which at any moment might produce 

 the factor of red, if given the chance; and I have already 

 spoken of the way in which in certain families or groups 

 of families we seem to see two conspicuous and contrasted 

 colour-factors, or perhaps more than two, which are con- 

 stantly turning up in the species of the group. 



The moral seems to me to be that we must ignore the 

 colour-factors of the male, or of the female if she has 

 assumed masculine attributes, and take as our pattern the 



