Permanence and Evolution. 143 



for (letting alone that sexual selection is quite 

 unproved) is not simply the presence of bright, 

 but the presence of various different bright 

 colours, arranged often in beautifully symmetri- 

 cal patterns, separately characteristic of different 

 living forms. 



It is true that the most conspicuously beautiful 

 birds, as, for instance, humming-birds and sun- 

 birds, are fruit-eating ; but Grant Allen does not 

 attempt to point out any particular relation 

 between their colours and that of the fruits 

 they feed on, though in the Colour Sense (p. 185) 

 he insinuates that this might be done. Yet it is 

 evident that till such a test is applied nothing 

 has been really done for the support of his 

 hypothesis. If the colouration, say of birds, is 

 determined by the predilection of the females, 

 then female birds have an eye for minute shades 

 of difference amid exquisite arrangements of 

 colours, and of different strains of female birds 

 each has retained and improved through count- 



