24 Flowers and tJicir Pedigrees. 



grasses, where the stamens hang out in long pendu- 

 lous clusters, and the pollen is easily wafted by the 

 breeze from their waving filaments to the pistils of 

 surrounding flowers. In such cases as these, the 

 stamens are generally very long and mobile, so that 

 the slightest breath shakes them readily ; while the 

 sensitive surface of the pistil is branched and feathery, 

 so as readily to catch any stray passing grain of wind- 

 borne pollen. 



But there are other flowers which have adopted a 

 different method of getting the pollen conveyed from 

 one blossom to another, and this is upon the heads 

 and legs of honey-eating insects. From the very 

 first, insects must have been fond of visiting flowers 

 for the sake of the pollen, which they used to eat up 

 without performing any service to the plant in return, 

 as they still feloniously do in the case of several wind- 

 fertilised species ; and to counteract this bad habit on 

 the part of their unbidden guests, the flowers seem to 

 have developed a little store of honey (which the 

 insects prefer to pollen), and thus to have turned their 

 visitors from plundering enemies into useful allies and 

 friends. For even the early pollen-eaters must often 

 unintentionally have benefited the plant, by carrying 

 pollen on their heads and legs from one flower to 

 another ; but when once the plant took to producing 



jT^"^ 



