26o THE POPULAR SCIENCE MONTHLY. 



head. This frontal eye is totally wanting in the otlier workers, and 

 is not known in any other kind of ant. Their special functions are 

 unknown. None of this sj^ecies are pugnacious. 



The work of reproduction begins with the rainy season. The union 

 probably takes place in the night, for in the morning the neighborhood 

 of the nest will be strewed with the females, and the dead bodies of the 

 males, the former already fertile, from whom the workers make it their 

 duty to tear away the wings. The true females are incapable of at- 

 tending to the wants of their offspring ; and it is on the poor, sterile 

 workers, who are denied all the other pleasures of maternity, that 

 the care devolves. The successful d^hut of the winged males and 

 females depends likewise on the workers. Great activity reigns in an 

 ants'-nest on the exodus of the winged individuals. The workers clear 

 the roads of exit, and show the most lively interest in their departure, 

 althougli it is highly improbable that any of them will return to the 

 same colony. They are of large size, the female measuring two and a 

 quarter inches in expanse of wing ; the male is not much more tlian 

 half the size. They swarm in vast numbers, but are so eagerly preyed 

 upon by insectivorous animals that but few of the impregnated females 

 escape the slaughter to found new colonies. An immense amount of 

 labor would be saved to the ants, if, instead of raising annually myr- 

 iads of winged males and females to perish, they raised only a few 

 wingless males and females, which, free from dangei*, might remain in 

 their native nests ; and, as Fritz Muller says, he who does not admit 

 the paramount importance of intercrossing must of course wonder 

 why the latter manner of reproduction has not long since taken the 

 place, through natural selection, of the production of winged males 

 and females. But the wingless individuals would of course have to 

 pair always with their near relatives, while by swarming a chance is 

 given for the intercrossing of individuals not nearly related. 



Fig. 2. Sauba Ant, Female. 



Resembling the saliba, in being vegetable-feeders, are the harvest- 

 ing-ants {Atla stnictor, A. barhara, Pheidole vnegacephala^ etc.). It 

 has been a fashion among naturalists to set down as pure invention 

 the accounts by classical writers of the accumulation of cereals by ants 

 for winter consumption, and to assume that the Biblical injunction to 



