302 WHEELER. 



cocoons — exceedingly thin coverings of pinkish silk. As this took 

 place out of the nest, in the jungle, they must be covered with wood and 

 leaves. The vital necessity of this was not apparent, for none of this 

 debris was incorporated into the silk of the cocoons, which were clean 

 and homogeneous. Yet the hundreds of ants gnawed and tore and 

 labored to gather this little dust, as if their very lives depended upon 



it When first brought from the nest, the larvae lay quite straight 



and still, but almost at once they bent far over in the spinning position. 

 Then some officious worker would come along, and the unfortunate 

 larva would be snatched up, carried off, and jammed down in some 

 neighboring empty space, like a bolt of cloth rearranged upon a shelf. 

 Then another ant would approach, antenna the larva, disapprove, and 

 again shift its position. It was a real survival of the lucky, as to who 



should avoid being exhausted by kindness and over-solicitude 



There was no order of packing. The larv?e were fitted together any- 

 way, and meagerly co\'ered with dust of wood and shreds of cloth. 

 One big tissue of wood nearly an inch square was too great a tempta- 

 tion to be left alone, and during the course of my observation it covered 

 in turn almost every group of larvae in sight, ending by being acci- 

 dentally shunted over the edge and killing a worker near the kitchen 

 middens. There was only a single layer of larvae; in no case were they 

 piled up, and when the platform became crowded, a new column was 

 formed and hundreds taken outside. To the casual eye there was no 

 difference between these legionaries and a column bringing in booty 

 of insects, eggs and pupse; yet here all was solicitude, never a bite too 

 severe, or a blunder of undue force." These observations show that 

 in thus covering their larvje with foreign particles just before pupation, 

 Eciton hurchelli behaves exactly like many other ants, a fact which 

 Beebe did not know, for legless larvse, like those of ants, cannot, of 

 course, spin their cocoons without a temporary covering of earth or 

 debris to which they can attach their silk. Unfortunately he fails 

 to tell us anything about the castes to which the cocoon-spinning 

 larvae belonged, and we are left to infer that they were probably, in 

 great part at least, workers and soldiers. 



When I looked up the earlier literature on Eciton on my return to 

 Boston I found that Wilhelm Miiller (1886) had found pupae of 

 hurchelli enclosed in cocoons. March 1st, 1885, he found a little to one 

 side of an Eciton file a dealated male Labidus hurchelli which was 

 being partly dragged and partly pushed along by a couple of workers. 

 As the insect, when placed among a lot of workers and soldiers, re- 

 mained unharmed, he naturally inferred that it belonged to the 



