CHAP. V Social Life of Animals 87 



the winged males and females which are produced in vast 

 numbers, and which, leaving the termitary in large swarms, 

 may intercross with those produced in other communities, 

 there are (in some if not all of the species) wingless 

 males and females which never leave the termitary where 

 they are born, and which replace the winged males or 

 females whenever a community does not find, in due time, a 

 true king or queen." There is no doubt as to the existence 

 of both winged and wingless royal pairs. According to 

 Grassi, the former fly away in spring, the others ascend the 

 throne in summer. The complementary kings or viceroys 

 die before' winter ; their mates live on, widowed but still 

 maternal, till at least the next summer. 



This replacement of royalty reminds us that hive-bees, 

 bereft of their queen, will rear one from the indifferent grub, 

 but the termites with which we are best acquainted seem 

 almost always to have a reserve of reproductive members. 

 This other difference between termites and ants or bees 

 should be noticed, that in the latter the " workers " are 

 highly-developed, though sterile females, while in the former 

 the workers seem to be arrested forms of both sexes. They 

 are children which do not grow up. 



8. Evolution of Social Life. — To Professor Alfred 

 Espinas both naturalists and sociologists are greatly in- 

 debted for his careful discussion of the social life of animals. 

 It may be useful, therefore, to give an outline of the mode 

 of treatment followed in his work — Des SocidUs Animales : 

 Etude de Psychologie Comparh (Paris, 1877) :— 



Co-operation, which is an essential characteristic of all society, 

 implies some degree of organic affinity. There are, indeed, 

 occasional associations between unrelated forms — " mutualism," in 

 which both associates are benefited; "commensalism," in which 

 the benefit is mainly one-sided ; parasitism, which is distinctly 

 anti-social, deteriorating the host and also the rank of the tempor- 

 arily benefited parasite. Of normal societies whose members are 

 mutually dependent, two kinds may be distinguished — (a) the 

 organically connected colonies of animals, in which there is a 

 common nutritive life ; (b) those associations which owe their origin 

 and meaning to reproduction. Of the latter, some do not become 

 more than domestic, and these are distinguished as conjugal (in 



