BEHAVIOR OF ANTS AND MYRMECOPHILES 407 



ably till they have oviposited, but that discord supervenes with 

 maternity and the attraction of the ova. 



Experiments with dealated queens of Tetramorium yielded 

 somewhat similar results, but out of fifteen queens that reared 

 their broods together, three survived. In other species of ants 

 the queens are even more tolerant of one another, e.g., Plagio- 

 lepis pygmaea and Leptothorax unifasciatus. In this connection 

 Emery calls attention to the Argentine ant (Iridomyrmcx kumilis), 

 the colonies of which, according to Newell, may unite in great 

 numbers in the fall, producing enormous hibernating colonies, 

 each containing hundreds of queens. 



Emery (22) shows that larvae which are insufficiently nour- 

 ished produce weak and undersized workers. These inferior 

 workers are found in the first brood of a queen and in colonies 

 where the worker force is small numerically. As the success of 

 a colony of robber-ants depends on the presence of large, strong 

 workers for fighting, Emery considers that a populous host col- 

 ony is necessary. Social parasitism having arisen, in the opinion 

 of Emery, from nest -robbing habits, this necessity for a large 

 force of workers to attend the young has been an important 

 factor in its origin. 



There is no transition in habit from the forms of fusca, through 

 to the typical social parasites, rufa, triincicola, etc., so Emery 

 considers that the social-parasitic habit originated from a muta- 

 tion rather than by a gradual development. In sanguined the 

 mandibles of the male are toothed, while in fusca they- are 

 smooth, which shows that the latter species is even more special- 

 ized than the former. 



Enslin (23) observed Formica cinerea attending the small 

 European Membracid, Gar gar a genistae, which he found in 

 numbers on Sarothamnus scoparius. The plant was frequented 

 by ants which were foraging and also attending a few aphids 

 in the ordinary manner. When one of the ants finds a Gargara 

 nymph it strokes the abdomen with its antennae. The tree- 

 hopper elevates the abdomen and from the anus exserts a rod- 

 like structure on the end of which is a drop of light-colored 

 liquid which the ant imbibes. Then the ant either tries again 

 on the same individual or hunts for another. If the head or 

 thorax instead of the abdomen of the bug is stroked the result 

 is the same. Adults of Gargara are also attended, but not so 



