Wasps, Bees, and Ants 555 



find in the combination the springs of most if not all ant behavior; and what 

 will explain the complex activities of ants will certainly explain those of all 

 the other so-called "intelligent insects," namely, bees and wasps, both soli- 

 tary and social. 



A final problem in the life of the social insects is that touching the origin 

 and establishment of the various castes or kinds of individuals inside the 

 single species. The presence of two, often widely differing kinds of indi- 

 viduals, namely, male and female, is so familiar as to lose, for some of us, 

 part of its significance and importance. But why the young produced by 

 the union of male and female can differ so widely as they may, that is, to the 

 extent of the difference between male and female, seems to us explicable by 

 the fact that just such two differing parent individuals take part in the pro- 

 duction of the new individuals, and by the fact that such a phenomenon 

 is the usual and ordinary one of heredity. (However little we may under- 

 stand the natural phenomenon or law of heredity just as little do we under- 

 stand gravitation, which we habitually are content to assign as an ultimate 

 cause for certain effects). But with the social insects we have always one, 

 and often more than one, still different individual among the offspring, and 

 one which takes no part whatever in the (embryonic) production of new 

 individuals; it can hand on nothing to the offspring by heredity. The ques- 

 tion is, then, how are two kinds of individuals (male and female) able to 

 produce not only their own kinds, but a third kind which has no part in pro- 

 ducing or fertilizing the egg-cell from which it develops? 



And on the heels of this question conies a second. How is it that if the 

 present-day forms and kinds of animals are due to the results of the com- 

 bined influences of variation, natural selection, and heredity that is, that 

 the inevitably appearing slight congenital differences as they are of advantage 

 or disadvantage in the life of the animal are preserved or destroyed in the 

 species by natural selection how, it may be asked, have the characters of 

 the worker castes been thus determined by selection, for in this case the 

 modified individuals have no part in the transmission of their characteristics 

 by heredity? 



The first question is answered as far as it at present can be in terms not 

 wholly agnostic, by the statement that it is probably true among ants, as 

 has been shown actually to be true with certain other social insects, namely, 

 the termites (p. in) and the honey-bee (p. 525), that the difference between 

 queen (fertile female) and worker (infertile female) is brought about during 

 postembryonal development by differences regulated by the nurses in the 

 quality and quantity of food supplied the developing individuals. Sharp 

 says: "There is a considerable body of evidence suggesting that the quality 

 or quantity of the food or both combined are important factors in the treat- 



