340 ANIMAL AGGREGATIONS 



discussion, have social significance. In more loosely organized social 

 life we would expect the cohesiveness shown by animals at the 

 aggregation level to make up a greater part of the total unifying 

 forces which operate to produce a social unit. Further, we must 

 recognize that at least a part of the forces which operate to bind 

 members of families together are similar to those which operate in 

 the case of other kinds of social groupings. 



It is worth re-emphasizing, as Child (1924) and Alverdes (1927) 

 have recognized, that both the congregation and the family bases of 

 societies are in fact but two different types of aggregations, so that 

 at all events aggregations of some sort are essential for the develop- 

 ment of the social habit. In other words, this phase of the problem 

 of social origins is not the question whether the social habits as seen 

 among birds, ants, men, and others arose from aggregations such as 

 we have been discussing in earlier sections of this book or from some 

 sort of family; but rather it becomes a question as to the kind of 

 aggregation which gave immediate rise to them, since the family 

 type is only one of a number of kinds of aggregations, as we have 

 seen from Deegener's outline. 



The main value of the detailed classification of social organiza- 

 tion given by Deegener lies in his recognition and elaboration of the 

 essential unity of sex-conditioned and asexually conditioned social 

 tendencies. Throughout his two main categories of loose accidental 

 unions, or associations, and essential groupings, or societies, Dee- 

 gener has recognized important subdivisions, differentiated accord- 

 ing to whether the groups are formed on a sexual basis or prolonga- 

 tion of some sort of family relations, or whether by the gathering 

 of individuals from different parents into more or less well-integrated 

 groups. The former type, while common enough in homotypic 

 groupings, does not compose all such groups; while the latter is 

 most characteristic, though not necessarily exclusively so, of social 

 bands or flocks which contain more than one species. 



In order to appreciate the importance of social organizations 

 which are not fundamentally united on a sexual basis, one should 

 review Deegener's classification to find the extent to which he recog- 



