384 THE AMERICAN NATURALIST. [Vol. XXXVII. 



with rotate open flowers are monochromatic, while genera with 

 highly specialized corollas adapted to bees are polychromatic. 

 How have these color differences arisen, and how far are they 

 due to the selective influence of bees ? That bees in collecting 

 nectar adhere closely in their visits to a single species is well 

 known. Even when the flowers are nearly alike in form and 

 color the honeybee often shows a remarkable power of discrim- 

 ination. If all of the flowers blooming at the same time in a 



distinguish between them as readily as when there is a variety 

 of colors contrasting with each other. This is the reason of 

 the development, according to Mailer, of the numerous color 

 differences in bee flowers. There can be no doubt that bees 

 can distinguish between different hues, and can make their visits 

 more quickly and easily because of the contrasts of flowers in 

 coloration. If, says Muller, we assume that one of two forms, 

 closely allied in structure and of the same color, should vary 

 in color, bees would distinguish it at once, and follow it more 

 easily in their visits. He found, moreover, that a majority of 

 bee flowers are red or blue. Of one hundred species of bee 



