96 Mr. G. A. K. Marshall on Diaposcmatism, with refer* nee 



the butterflies' enemies will affect the results clue to their 

 attacks. Those which exhibit a high degree of intelligence 

 will obviously profit more quickly by their experience ; in 

 other words, they will acquire the necessary mental 

 association between colour and inedibility by the destruc- 

 tion of comparatively few butterflies. They will therefore 

 operate much less efficiently as producers of Miillerian 

 mimicry than will those enemies which have a compara- 

 tively low degree of intelligence and which therefore 

 require to make many experiments before arriving at the 

 same result. But if there be enemies still lower in the 

 scale and incapable of forming such a mental association 

 at all, then the destruction of butterflies which they 

 would cause would have no effect whatever from a purely 

 mimetic standpoint ; no more than if the insects had been 

 killed by a torrential thunderstorm. Similarly if we 

 suppose that a certain species of bird has specially adapted 

 itself to feed on a genus of insects usually avoided by 

 other insectivorous animals, the attacks of that bird will 

 have no effect in the direction of Miillerian mimicry on 

 that particular genus. In other words, the mental 

 attitude of the enemy towards its prey has an important 

 bearing upon the results which its attacks will produce. 

 Finally, the accumulation of experience does not render au 

 animal more effective as a Miillerian factor, but precisely 

 the reverse ; for as it becomes more skilled in recognising 

 nauseous species, so will it gradually cease its experimental 

 destruction, upon which this kind of mimicry so essentially 

 depends. 



If we turn for a moment to consider whether these 

 arguments are equally applicable in the case of Batesian 

 mimicry, we find, on the contrary, a totally different state 

 of affairs. So far from experience and intelligence being 

 adverse qualities, it is evident that the greater the 

 accumulation of experience and the higher the degree of 

 intelligence possessed by the insectivorous animal, the 

 greater will be its efficiency as a producer of Batesian 

 mimicry. For it will thus be the better enabled to 

 discriminate between the edible mimic and its inedible 

 model, with the result that there will be a more effective 

 selection and a keener elimination of those variations of 

 the mimic which do not come up to a high standard of 

 resemblance. A consideration of this difference in the 

 operation of these two mimetic forces would appear to 



