188 THE VALUE OF COLOUR 



cases, although there is no reason to suspect 

 any greater variability, the female is commonly 

 a somewhat better mimic than the male and 

 often a very much better mimic. Wallace's 

 principle seems here to supply the obvious in- 

 terpretation ; but it is to be noted that the evo- 

 lution of Mimicry is taking place in colours that 

 are associated with sex. Otherwise, it is impos- 

 sible to explain the fact that the more perfect 

 Mimicry attained by one sex is not immediately 

 transferred to the other. 



(6) When the difference between the patterns 

 of model and presumed ancestor of mimic is very 

 great, the female is often alone mimetic ; when 

 the difference is comparatively small, both sexes 

 are commonly mimetic. The Nymphaline genus 

 Hypolimnas is a good example. In Hypolimnas 

 itself the females mimic Danainae with patterns 

 very different from those preserved by the non- 

 mimetic males: in the sub-genus Euralia, both 

 sexes resemble the black and white Ethiopian 

 Danaines with patterns not very dissimilar from 

 that which we infer to have existed in the non- 

 mimetic ancestor. 



(7) Although a melanic form or other large 

 variation may be of the utmost importance in 

 facilitating the start of a mimetic likeness, it is 

 impossible to explain the evolution of any detailed 

 resemblance in this manner. And even the 

 large colour variation itself may well be the 

 expression of a minute and * continuous ' change 



