338 ANIMAL AGGREGATIONS 



certain conditions when two similar asexual cells are crowded to- 

 gether within a limited volume of medium. If this suggestion is 

 sound, it would indicate that mass physiology of animals is much 

 more primitive and fundamental than can be considered to be the 

 case under the assumption that the gregarious or social habit in 

 animals is at bottom an outgrowth only of the association of young 

 individuals with one or both parents, and therefore usually a result 

 of sexual reproduction. It is assumed that in special cases or at 

 critical periods in social evolution the period of the association be- 

 comes lengthened and the family comes to react as a unit under 

 many conditions. Students of social life in insects, especially as it 

 exists among wasps, bees, and ants, usually adopt this explanation 

 in some form for the origin of the social habit. Wheeler, in his 

 summaries of studies on ants (1913a, 1918), and more recently in his 

 review of social life in insects (1923, 1928), regards the insect colony 

 as the result of an extension of the affiliation of mother and off- 

 spring. Wheeler's particular contribution is his theory that mutual 

 feeding, which he calls "trophallaxis," is the bond that unites parent 

 and offspring in the social insects; the mother feeds on secretions 

 from the larvae and so is bound to them through self-interests, 

 while they receive food from the mother to their own advantage. 

 Wheeler shows that the social habit, meaning thereby a more or 

 less prolonged association of young and adults, has arisen de novo 

 at least thirty different times among insects alone, in nearly that 

 many natural taxonomic families or subfamilies, belonging to five 

 different orders. The gradual development of the mother-offspring 

 family from the solitary insects is shown almost diagrammatically 

 by the growth of the social habit among the solitary wasps (Wheeler, 



1923)- 



Herbert Spencer's suggestion that colony life arose from the con- 

 sociation of adult individuals for nonsexual, co-operative purposes 

 was an early recognition of that type of social unit and, at the time 

 it was advanced (1893-94), was not well grounded on proved fact. 

 Spencer suggests that in some cases permanent swarms arise from 

 such consociation and that natural selection establishes such of 

 these groupings as are advantageous. In terms of human society, 



