264 THE SOCIAL LIFE OF ANIMALS 



and is known as a worker. Apparently the funda- 

 mental difference can be brought about only by the 

 treatment which the developing grub receives after 

 hatching, and is not a matter of heredity. Just how 

 the workers are stimulated to give one or more grubs 

 the treatment that will allow them to develop their 

 full reproductive capacities is not fully known. If, 

 however, the queen bee dies or is removed from the 

 colony, workers will start enlarging one or more of 

 the cells which contain developing grubs, change 

 their care and feeding and so allow them to trans- 

 form into fertile reproductives. Perhaps they are 

 kept from doing so when a queen is present by some- 

 thing like a social hormone, which there is good rea- 

 son for thinking is produced by the even more social 

 termites. 



The mechanism which results in caste formation 

 among ants need not be the same as that in wasps 

 and bees, since it is generally conceded that they 

 had a separate social evolution. For years two theo- 

 ries have been promoted as to how ant castes came 

 into being. One group of students thought that ant 

 castes were determined as were those of bees and 

 wasps, by care and food; another group w^as equally 

 sure that the differences were hereditary. After con- 

 fessedly wavering between the two views in his long 

 study of ants. Professor Wheeler in his posthumous 



