COMPOSITE NATURE OF MIMICRY 241 



substituted for Natural Selection, viz. the theory of 

 External Causes. 



A mimetic appearance is commonly made up of (i) 

 colour, including (a) structural as well as {d) pigment 

 colours ; (2) pattern ; (3) form ; (4) attitude ; (5) move- 

 ment. 



It may be plausible to hold that direct local influences 

 determine colour, but the case becomes much more diffi- 

 cult when structural tints are included, as they frequently 

 are. Thus it might well be held that the dark pigment 

 of a female Hypolimnas and of the Euploea which it 

 resembles are alike the direct effect of the locality they 

 both inhabit. But the most convinced advocate of direct 

 local causes would probably hesitate to explain, by the 

 operation of the same forces, the structurally caused blue 

 sheen which overspreads the dark pigment in some of 

 these mimetic pairs. Similarly with pattern, it is much 

 more difficult to understand the appropriate arrangement 

 of the colours by direct forces than the production of the 

 tints themselves ; still more difficult to understand how 

 such forces could modify shape, and, again, more difficult 

 to see how they could mould the nervous and muscular 

 systems so as to produce appropriate attitudes and move- 

 ments. Most difficult of all to understand, except on a 

 theory of Selection, how the several elements in the 

 complex set of changes could be kept in their proper 

 relationship and guided to a definite end, viz. the pro- 

 duction of a superficial resemblance to another species. 



The objection to the theory of Internal Causes is not 

 that it is inadequate to produce each of the effects, but 

 that it is in the highest degree improbable that so com- 

 plete and harmonious an effect could be frequently 

 produced accidentally by the combination of such diverse 

 elements. 



It is useless to maintain that these resemblances are 

 the uniform result of uniform forces peculiar to the 

 locality ; for investigation proves that the results are 

 very far from uniform. They appear at almost any point 

 in the structure of the body, superficial or deep-seated, 

 generally at many points in a single individual both 



rOCLTON R 



