MIMICRY 213 



specimens of butterflies which resemble each other 

 more or less closely, and call the asssemblage " a 

 Mullerian association." The process is analogous to 

 that of matching colours in a Berlin-wool shop, and 

 is perhaps less useful. Without direct knowledge of 

 the living insects it must always be hazardous to 

 assert dogmatically that this or that species is a 

 Mullerian mimic. But it appears to me that a useful 

 criterion is afforded by the colouring of the under side 

 of the butterfly. For example, Elymnias lais has a 

 double defence : when the upper wings are exposed 

 to view, as in flight, the insect resembles a distasteful 

 Danaine, when the insect is at rest the under surface 

 only of the wings is seen, and it harmonizes perfectly 

 with the inanimate surroundings. Now the safety of 

 a distasteful butterfly lies in its bold advertisement of 

 nauseous properties, whether the butterfly be at rest 

 or in flight, and it is a significant fact that all those 

 butterflies which have been proved by direct experiment 

 to be nauseous to birds and reptiles are conspicuously 

 coloured on both wing-surfaces, and freely expose 

 themselves at all times ; such as, e.g., the Heliconince and 

 IthomiincK of the New World, and the Danaitice and 

 Acrceince of the Old World. If we, for the sake of 

 argument, assume that Elymnias lais is a distasteful 

 Mullerian mimic, we may imagine that the following 

 combination of circumstances would often occur ; a 

 bird, searching for prey, would discover and seize some 

 protectively coloured butterfly, and finding it good to 

 eat would, on encountering an Elymnias lais at rest 

 naturally associate palatability with the under-side patter 

 of this species ; seizing it the bird would then discover 

 that it had made a mistake, but by this time it would 



