INSECT SOCIETIES 



been made of the effects of nutrition on the size, structure and 

 fertility of Hymenoptera it may be possible to argue the case 

 in greater detail. 



The beginning of termite societies was probably quite 

 different. There are several kinds of insects, such as the 

 Embiids, in which all the developmental stages are known to 

 live together, sharing a common silken shelter but not help- 

 ing one another in any other way. What seems to have hap- 

 pened in the termites was that the developmental stages began 

 to play a part in the life of the community as important as 

 that of the adults. This was possible in them because of their 

 gradual metamorphosis and the general similarity of the 

 early stages to the adult. The caste system of the termites 

 would originally have been founded exclusively on age, the 

 half-grown larvae or nymphs doing most of the work while 

 the queen laid the eggs. This system would increase the 

 chances of variations in nutrition during the developmental 

 period. As in the Jlymenoptera, an increased response to the 

 effects of nutrition might have been evolved later. Such a 

 scheme would give at least a hypothetical explanation of the 

 origin of the more primitive termite societies. We still 

 know so little of the complex caste system of the higher 

 termites that any discussion of them is premature. 



INSECT SOCIETIES AS FAMILIES 



An insect society is a family, a group of offspring sur- 

 rounding the mother or royal couple. Their mutual toler- 

 ance is founded on growing up in the same nest. If ant 

 colonies are split into two in artificial nests, the two halves 

 may fight if reunited after an interval of some months. Judg- 

 ing by ants and honey bees, mutual tolerance depends on 

 sharing a common nest-smell, in natural conditions acquired 

 during development. Ants of the same species but from 



203 



