BATESIAN AND MULLERIAN MIMICRY It is obvious that this resemblance between the butterflies of distasteful groups is, as Bates recognized, a different thing from the mimicry he sought to explain by his hypothesis—the likeness of a harmless or palatable species to a formidable or nauseous one which advertises its qualities by warning characters, especially colours, but often by movement, attitude or sound. The mimic gains advantage by the display of false warning characters. This is called "Batesian Mimicry" after its discoverer. The resemblance between the distasteful species themselves was later explained by Fritz Muller as an adaptation which economizes life by facilitating the education of enemies. If two distasteful species are alike they will share between them the lives which must be sacrificed before the enemies have learned to associate the warning characters with their special means of defence. If they are unlike, the lives must be contributed by each of them independently. These resem blances, which are a form of common or combined warning, are called "Miillerian mimicry." A Batesian mimic may be compared to an unscrupulous tradesman who copies the advertisement of a successful firm ; Mullerian mimicry to a combination between firms to adopt a common advertisement and share the expense.
The qualities of Mullerian mimics differ widely in effectiveness and the species differ immensely in their relative numbers and in their capacities for variation. Hence models and mimics exist among them no less than in Batesian mimicry and in both there is the same evidence of the development of a mimetic from a dif ferent non-mimetic appearance still retained by allied species and often by the non-mimetic male. Those in second row are suffi cient to show the likeness to the model which may be attained by a MUllerian mimic. But while Batesian mimicry is never an advantage and may be a disadvantage to its model, Miillerian mimicry is never a disadvantage but generally (theoretically al ways) an advantage. Although this criterion is in itself simple, it is extremely difficult, with our present imperfect knowledge of mimics and their enemies in life, to apply. The decision therefore rests on indirect evidence, which makes a different appeal to different minds, so that opinions differ widely as to the relative importance of the two kinds of mimicry.
It will be generally agreed that the resemblance of a Batesian mimic to its model is closely related to protective resemblance.
In fact Bates originally included both resemblances under mimicry.
A. R. Wallace and nearly all later writers on the subject have differed from him in this because, although the two classes are so near akin, the species in one are conspicuous, in the other well concealed. They may, however, be conveniently grouped together under deceptive (or apatetic) resemblance. Both classes include palatable species, much sought after by the enemies of the group to which they belong. Any evidence supporting the conclusion that a mimic has been developed from a species with concealing colours would be evidence that it is a Batesian and not a MUllerian mimic. Such evidence may generally be obtained by comparison with its closest non-mimetic allies, and in butterflies often by com paring a mimetic female with its non-mimetic male, paying special attention to the under-surface, upon which concealing colours are specially developed. In the tropical American butterfly Proto gonius, conspicuously mimetic on the upper surface and dead leaf-like on the under, Kaye has recently observed that the latter appearance is so transparent that when the insect is sailing with expanded wings above the observer, he can only see the colours of the upper side.
If, on the other hand, the above comparisons lead to the conclu sion that a mimic belongs to a group with warning colours, espe cially one which supplies models for mimicry,—we should, in the opinion of the present writer, be justified in considering it as Mill lerian. Special evidence is offered by some mimics which retain warning characters independent of the mimetic appearance, and by others which, themselves resembling a central model, never theless act as models for still more outlying members of an association. Still stronger evidence may sometimes be found in reciprocal relationship between the species of two groups or between the groups themselves, A usually supplying models and B their mimics, but B also including models mimicked by species in A. In any case it is often difficult to draw the line between Bate sian and Miillerian mimicry.