BUCKINGHAJVI. — DIVISION OF LABOR AMONG ANTS. 499 



ence it has. The view of Spencer ('94) and his followers is that any 

 fertilized egg may develop into a queen or worker, or any modification 

 of either, or any transitional stage between the two, merely by a differ- 

 ence of feeding in the larval stage. The larvae which pupate before 

 reaching their maximum growth become workers, the others, queens ; 

 and the several castes of workers are produced from larvae which pupate 

 at different stages. 



Emery ('94, '96, :04) and his followers go farther than this, claiming 

 that the germ-plasm, being plastic, is capable of different degrees and 

 kinds of development according to the stimuli (especially food) acting 

 on the larvae. This peculiarity of the germ-plasm, which is laid down 

 in every female egg, is somatogenic and not blastogenic, and is trans- 

 mitted as an ability of germ-plasm to develop variously in the individ- 

 ual according to the conditions. 



Finally, there are those who hold with Weismann ('92, pp. 455-497) 

 that, since the queen and the workers are very different, the eggs pro- 

 ducing them must contain in the germ-plasm a different " determinant " 

 for each form. The stimulus which calls forth one or another of these 

 forms seems to be difference of feeding in the larval state, but 

 such feeding alone would not be a sufficient cause to produce such 

 differences. 



It seems to me that while feeding, probably only quantitative, is ef- 

 fective in bringing about the various castes in a species, it is a stimulus 

 rather than a cause, for certain structures of the workers and queens 

 of some species show that morphologically the queen and worker of a 

 given species vary independently of each other, and the same thing is 

 true of distinct soldier and worker castes. These facts make it appear 

 that, as "Wheeler says, " while adaptive characters in stature, sculpture, 

 pilosity, and color must depend for their ontogenetic development on 

 the nourishment of the larva, it is equally certain that they have been 

 acquired and fixed during the phylogeny of the species. In other 

 words, nourishment, temperature, and other environmental factors 

 merely furnish the conditions for their attainment of characters pre- 

 determined by heredity." Wheeler feels that we are therefore " com- 

 pelled to agree with Weismann that the characters that enable us to 

 differentiate the castes must be somehow represented in the egg. We 

 may grant this, however, without accepting his conception of repre- 

 sentative units." 



The worker characters are inherited, because the workers are capable 

 of laying eggs (according to Wheeler, much more often than has been 

 supposed), and these eggs develop into males, through which the 

 worker characters are transmitted. 



