vn THE COLOURS OF THE LEPIDOPTERA _ 147 



of variable coloration, or the nature of the adjust- 

 ments by which it is rendered possible. Those who 

 are interested in the question will find it discussed 

 in numerous other popular books besides those 

 already mentioned. It is, however, of some interest 

 for our purpose to note that, according to recent 

 research, it is not so much the colour of the environ- 

 ment which directly affects the larvae as the intensity 

 of the light (see Garbowski). In other respects also 

 some of the first crude statements made on the 

 subject are being modified. It may, indeed, be 

 accepted as a general truth in biology that when- 

 ever a series of phenomena seem to be susceptible 

 of an extremely simple and beautiful explanation, 

 that explanation is wrong and founded on an im- 

 perfect acquaintance with the phenomena in ques- 

 tion. All recent progress in biological theory has 

 shown that life is not simple but complex, and has 

 involved a substitution of extremely complex theories 

 for simple ones. 



The Colours of Butterflies 



In butterflies the colour-phenomena are much 

 more complex than in caterpillars, for while in the 

 latter the colour is a direct pigmental effect, in the 

 former the two factors of structure and pigment occur 

 in combination. In both kinds of colour in butter- 

 flies the important elements are the scales of the 

 wings. These are outgrowths of the chitinous 

 cuticle, and consist of a double membrane ; the outer 

 membrane is frequently much differentiated, and 

 may display rows of blunt projections, which are 



