136 THE BEATJTrES OF NATURE 



PAST HISTORY OF FLOWERS 



If the views here advocated are correct, it 

 follows that the original flowers were small 

 and green, as wind-fertilised flowers are even 

 now. But such flowers are inconspicuous. 

 Those which are coloured, say yellow or white, 

 are of course much more visible and more 

 likely to be visited by insects. I have else- 

 where given my reasons for thinking that 

 under these circumstances some flowers be- 

 came yellow, that some of them became white, 

 others subsequently red, and some finally blue. 

 It will be observed that red and blue flowers 

 are as a rule highly specialised, such as 

 Aconites and Larkspurs as compared with 

 Buttercups ; blue Gentians as compared with 

 yellow, etc. I have found by experiment 

 that Bees are especially partial to blue and 

 pink. 



Tubular flowers almost always, if not 

 always, contain honey, and are specially suited 

 to Butterflies and Moths, Bees and Flies. 

 Those which are fertilised by Moths generally 



