FIVK KINDS OF INDIVIDUALS IN SAUBA. 21 



and watching for approaching danger. They would 

 occasionally drop their bodies to the ground again, run 

 off to one side, and fiercely work their jaws and antennae, 

 as if having detected some strange sounds in the dis- 

 tance. Discerning nothing, they would quickly return 

 to their posts and resume their positions, thus acting 

 as scouts.' ' 



The same thing has been noticed by other natu- 

 ralists. Bates, for instance, states that in the marching 

 columns of Eciton drepanophora the large-headed 

 workers 'all trotted along empty-handed and outside 

 the column, at pretty regular intervals from each other, 

 like subaltern officers in a marching regiment. . . I 

 did not see them change their position, or take any 

 notice of their small-headed comrades;' and he says 

 that if the column was disturbed they appeared less 

 pugnacious than the others. 



In other species, however, of the same genus, Eciton 

 vastator and E. erratica, which also have two distinct 

 kinds of workers, the ones with large heads do appear 

 to act mainly as soldiers. When a breach is made in 

 one of their covered ways, the small workers set to 

 work to repair the damage, while the large-headed ones 

 issue forth in a menacing manner, rearing themselves 

 up and threatening with their jaws. 



In the Sauba Ant of South America (CEcodoma 

 cephalotes\ the complexity is carried still further; 



1 liev. T. S. Savage on the ' Habits of the Driver AnU,' Trant. 

 Knt. Soc., vol. v. p. 12. 



