IOO FLOWERS OF THE FIELD AND FOREST. 



many flowers follow that track in their individual development. 

 A common English forget-me-not is pale yellow when it first 

 opens, then changes to pink, and ends by being blue. A wall-flower 

 i<; first whitish, then yellow, and finally red or blue. An evening 

 primrose has white flowers at first, but at a later period of develop- 

 ment, red ones. Cobaa scandens, which has been flowering lux- 

 uriantly and blossoming perfectly in my study all winter, has 

 constantly shown this kind of evolution of color. It is first green, 

 then lightens much into a very pale-green, or white, and then 

 begins to develop toward purple, passing in some cases as I 

 noticed, through a pronounced pink. Its final color is a strong 

 purple. The garden convolvulus opens, a blushing white, and 

 passes into a full purple. When changes in the color of 

 flowers take place during the process of growth, they are, so 

 far as has been observed, all in this, and never in the opposite 

 direction. 



There can scarcely be good reason to question, I suppose, that 

 the evolution of flowers and of honey-eating insects has gone on 

 side by side, each helping the other. In given cases, the color and 

 form of the floral envelope, the nature of the honey sack, together 

 with the position of the stamens and pistil, are all correlated with 

 the specialized organs and particular habits of the insect tribe 

 whose help is depended upon in the act of pollenization. Owing 

 something, then, to the agency of insects for the possession of all 

 the exquisite beauty and sweetness of flowers, I can make no 

 more appropriate ending for this paper, than by quoting a few lines 

 from Emerson's "Humble-bee." 



Hot mid-summer's petted crone, 

 Sweet to me thy drowsy tone, 



