328 THE COLOURS OF ANIMALS 



for believing that a width which just prevents the 

 appearance of colour is an indication of want of 

 vitality. 



We must also remember that these iridescent 

 tints occur in combination with colours produced in 

 other ways. If we take a hypothetical case, the 

 inadequacy of ' surplus vitality ' as an explanation 

 becomes at once apparent. 



Let us suppose that a male bird becomes more 

 beautiful in appearance, and that the change consists 

 in the addition of white feathers, of new tints or 

 shades in the colours due to pigments, and of those 

 due to interference. 1 We must therefore suppose that 

 a ' surplus of vitality ' favours the disappearance of 

 pigment and the substitution of bubbles of gas, in 

 one part, although albinism affords rather strong 

 evidence that such a result is certainly not an indica- 

 tion of strength : we must suppose that the same 

 cause favours slight changes in the chemical con- 

 stitution of pigments, in other parts, involving the 

 excessively unlikely hypothesis that the aesthetic 

 value of the results is a measure of the difficulty 

 involved in their production : and we must finally 

 suppose that, elsewhere, the same cause is efficient 

 in adjusting with mathematical precision the width 

 of spaces in the tissue, although it is wildly im- 

 probable that the minute differences which correspond 



1 Admitting, for the sake of argument, that this cause is effective 

 in birds, as it certainly is in insects. 



