ARMY ANTS — SCHNEIRLA 399 



None of the hundreds of colonies in this species which have been 

 studied at various times of year in these investigations has failed to 

 show the described correspondence between brood condition and the 

 colony activity pattern, with the exception of one natural case in 

 which the queen had died and all semblance of the normal pattern 

 characteristic of the species had disappeared. Such cases may be not 

 infrequent, but extinction must be rapid, as indicated by our tests. 

 Without further broods, as our field experiments show, regular no- 

 madic function disappears in a colony, and the only way in which the 

 worker population can be saved is by fusion with another colony of 

 the species.^ 



The timing device of these recurrent and surprisingly regular cycles 

 is no external event such as the lunar phases which influence rhythmic 

 behavior in many other animals. Rather it lies within each colony, 

 in the casual relationship between brood condition and colony be- 

 havior. An active larval brood as it grows provides more and more 

 tactual and chemical stimulation to the adult workers, who are cor- 

 respondingly responsive. (The nature of the intimate stimulative re- 

 lationship normally prevailing between workers and brood has been 

 demonstrated in laboratory observations and tests.) The brood 

 thereby energizes the colony in an accelerating manner dm'ing its 

 larval development. A great raid then ordinarily occurs each day, 

 leading by complex stages into a colony emigration sometime after 

 dusk. However, the essential increment of social stunulation vanishes 

 abruptly when this larval brood matures and becomes enclosed in 

 cocoons, and without it the level of colony excitement falls to a new 

 low at which raids are always too small to provide a basis for emigra- 

 tion. Each statary phase ends when the emergence of a mature pupal 

 brood as highly active callow workers suddenly raises the colony 

 excitation level to a point at which the large daily raids and emigra- 

 tions can occur. The hordes of newly emerged workers lose their 

 initial excitatory effect after a few days of further maturation, where- 

 upon they behave essentially like ordinary adult workers; however, in 

 the meantime a new larval brood has reached a condition of high stim- 

 ulative potency, so that a nomadic colony function continues without 

 a break until this brood in its turn is mature, and so on. 



Thus the key to periodic emigration evidently lies in the fact that 

 when the colony is overstimulated with sufficient intensity and persist- 

 ence, the population responds to environmental excitation at a high 

 level of energy and participation. Each morning the initial stimulus 



* After the queen has been absent for more than a few hours, fusion can occur 

 when the colony crosses paths with another of its species. This event throws 

 interesting light on the normal adjustments of workers to the colony queen and 

 upon the imifying effect of her odor in the colony (Schneirla, 1949). 



