THE PRINCIPLE OF CO-OPERATION 359 



be packed with Mya so that the openings of their siphons fairly 

 crowd the surface of the mud. At times and in favorable locations 

 jelly fishes, or even minute copepods, may discolor the sea for miles; 

 the entrance to the White Sea may be covered by red streaks pro- 

 duced by the presence of multitudes of starfish eggs (Mesiacev, 

 1927). 



Brues (1926) estimated that the Hymenorus beetle population of 

 a single panicle of Florida yucca would be about 15,000, and cites 

 case after case of well-established insect aggregations. Some of the 

 more striking include the hibernating aggregations of ladybird 

 beetles {Hippodamia convergens) in northern California, of which 

 Carnes (191 2) records that two men in a single day can gather from 

 1,200,000 to double that number from the hibernating masses 

 among the pine needles. A thousand chinch bugs have been found 

 In the shelter of a single tuft of grass 3 inches in diameter (Headlee, 

 1910). Howard (1898, 1901) records flights of a chrysomeUd beetle, 

 one of which formed a belt 15 feet thick and a hundred yards wide 

 over the course of the Gila River. The flight continued for 2 days. 

 Cicadas, monarch butterflies, migratory locusts, and many Diptera, 

 including Polenia rudis, Muscina, house flies, midges, and other 

 insects, are known to collect in great numbers. In this survey I 

 have not mentioned the collections of insects about electric lights, 

 or the insects in the shore drift of lakes, or the vast collections of 

 the more strictly social species, or the type of relationship usually 

 called "symbiosis." 



The recapitulation we have just made summarizes the evidence 

 on two points. Aggregations of animals with little or no group or- 

 ganization, which possess survival values for the aggregants, have 

 been demonstrated sufficiently widely throughout the animal king- 

 dom to indicate that if studied in the other taxonomic units with the 

 proper methods they can probably be demonstrated to occur in ah 

 the larger taxonomic divisions. Certainly they have already been 

 demonstrated in groups sufficiently widespread to indicate that the 

 absence of such group protection, other things being equal, is to be 

 regarded as an exception rather than the rule. Such aggregations 

 are ecologically as well as taxonomically widespread, and they are 



