STERILE INSECTS. 237 



tion, if it be granted that the shape, size, and color- 

 ing of the butterfly's wings are variable. It is, of 

 course, questioned whether, even with this rigid 

 selection, slight, indefinite variations are enough to 

 account for the final completion of some of the 

 wonderful pieces of mimicry, but this objection is so 

 similar to one already discussed that we need do no 

 more than mention it, and indicate that it is fre- 

 quently regarded as a very great difficulty for 

 natural selection. 



Sterile Insects. 



There is one special difficulty to the principle, of 

 so much weight as to deserve notice here, although 

 it is not our purpose to consider special cases. This 

 is the difficulty in seeing how the working castes of 

 various insects could have arisen, and it is so much 

 of an obstacle that Darwin at first thought it fatal 

 to his theory. Many insects, such as ants and bees, 

 live in large communities, thousands of them com- 

 bining to live in mutual dependence on each other. 

 The labor of the colony is divided among different 

 members, each being fitted to perform certain work. 

 In the beehive is found the queen bee, who pro- 

 duces all of the eggs ; the drones, whose sole duty 

 is to fertilize the eggs ; and, finally, the workers, 

 who take no share in reproduction, but perform the 

 manual labor for the colony, searching for food, 

 building the hive, caring for the young, etc. It is 

 this last class that concerns us. The worker bees are 

 very different from the other classes. Not only 

 have they different instincts, which impel them to 



