260 THE VARIATION OF ANIMALS IN NATURE 



if, as Perkins contends, the whole Eumenine fauna evolved from 

 two species immigrant from the Orient. From a Miillerian 

 standpoint one would rather have expected that all the species 

 would have been alike, that change in colour would have been 

 more retarded compared with change in structure. Exactly 

 the same argument may be applied to the S. American Vespids. 

 In most districts there is more than one large association of 

 unrelated species with similar colour-patterns. Often quite 

 closely related species belong to very different colour-groups. 



An interesting example is known amongst the butterflies of 

 the genus Erebia. This genus of Satyrines is of sombre brown 

 hue with a cryptic under-surface. There is no evidence to 

 suggest that they are not quite palatable to birds, and they 

 would be considered very unlikely insects to form Miillerian 

 associations amongst themselves. Yet Chapman (191 3) and 

 Higgins (1930) have both recorded marked colour convergence 

 between different species in various localities in the Alps. 

 The amount of convergence, though significant, is small and 

 would not make much difference to their appearance on the 

 wing, but this limited geographical polymorphism, with each 

 species having a parallel local form in each district, is what 

 would have been called Miillerian mimicry if the insects had 

 been brightly coloured. It is possible that in reality some 

 edaphic factor is involved. 



These examples are only supplementary to what has already 

 been brought forward on pp. 255-259. The matter is not 

 one capable as yet of proof either way, and we can only 

 state our opinion that it is very doubtful if the mimicry theory 

 can be made to cover all the facts. We may summarise the 

 argument of the previous paragraphs as follows : The fact 

 of mimicry, of striking resemblance between structurally un- 

 related forms, is well established, and the phenomenon is wide- 

 spread, especially amongst insects. In a number of selected 

 examples there is a considerable degree of probability 1 that 



1 A certain number of examples must probably be accepted on the grounds 

 of close degree of resemblance between model and mimic, coincidence of geo- 

 graphical range (often combined with geographical variation) and general evidence 

 as to distastefulness of the model and relative scarcity of the mimic. 



Probably there is no single example in which (a) a model has been proved 

 to be distasteful by its almost invariable rejection by its potential enemies, and 

 (b) a mimic of it is also regularly rejected although actually palatable. The 

 extremely scattered evidence for the mimicry theory makes it very difficult to 

 collate the facts recorded with regard to any particular pair of species. 



