THE FLOWER AND THE BEE 



for ten minutes. The first received 15 visits from honey-bees, 

 the second 13 visits. The blue corollas, together with the cone 

 of black anthers, were now removed from the flowers of the 

 first group. The two groups were now observed for a second 

 ten minutes; the first received no visits, the second 7 visits 

 from honey-bees. Once a bee hovered around the denuded 

 flowers of the first group, but failed to alight, although they 

 contained an abundance of nectar. There were scattered 

 upon the ground many partially withered corollas and it was 

 interesting to notice that a bee was twice seen to fly down 

 toward them. The value of conspicuousness was here again 

 very clearly established. 



A staminate flower of the garden-squash {Cucurbita maxima) 

 was placed under observation for ten minutes and received 12 

 visits, 8 from honey-bees and 4 from bumblebees {Bombus 

 terricola). The yellow corolla was then removed, and it was 

 watched for a second ten minutes, during which it received only 

 a single visit from a bumblebee. Two squash-flowers, both 

 staminate, growing side by side, their corollas touching, were 

 then selected. Both were observed for ten minutes. Number 

 one received 6 visits — 4 from bumblebees, 2 from honey-bees; 

 while number two received 13 visits, all from bumblebees. 

 The fresher condition of the second flower probably accounted 

 for the larger number of visits. The yellow corolla was now 

 cut away from number two, and both flowers were watched for 

 another ten minutes. No visits were made to the denuded 

 flower, but number one received 12 visits, 6 from honey-bees, 

 and 6 from bumblebees. In the previous experiments the 

 number of visits to the complete flowers were numerous and 

 decisive. On the contrary, they ceased almost entirely to the 

 decorollated flowers, although they contained an ample supply 

 of nectar. That the white, blue, and yellow corollas were bene- 



214 



