158 WILLIAM MORTON WHEELER. 



In order to study the ants I placed them with their brood and 

 fungus gardens in large Petri dishes. These made excellent 

 artificial nests in which the insects could be kept in perfect health 

 for two to three weeks and conveniently observed under a strong 

 pocket-lens. Within 24 to 36 hours the frail fungus gardens, 

 which inevitably fell to pieces when extracted from the earthen 

 chambers, were completely reconstructed by the workers as an 

 elaborate and rather regular sponge-work with polygonal crypts a 

 quarter to half an inch in diameter. Of course, the flat space, less 

 than half an inch in thickness, to which the ants were confined, 

 compelled them to rebuild their garden in the form of a disc 

 instead of a sphere or ovoid, but this was very advantageous, 

 since it permitted the observer to scrutinize all parts of the 

 structure through the glass cover. (See figures 2 and 3 from 

 photographs by my friend Dr. David Fairchild.) The ants 

 placed their eggs, larvae and pupae on the fungus-covered surfaces 

 and in the crypts. The queen is a very sluggish insect and 

 remained for long hours in a somnolent attitude near the center 

 of the garden, or moved about very slowly and scattered her eggs 

 in the immediate vicinity. These were rather large and broadly 

 elliptical and were permitted to lie where they were laid till 

 sometime after the larvae had hatched. The workers then 

 carried them to other parts of the garden and placed them in 

 contact with fresh hyphae. Since I never saw the workers ad- 

 ministering hyphae or "kohlrabi" to the larvae as described for 

 Atta cephalotes by Tanner (1892), I infer that the latter, when 

 hungry, merely reach out and crop the fungus. 



The larvae are short, thickset and beset with sparse, long, 

 flagellum-like hairs. The head is large and subrectangular, 

 bearing small, acute mandibles covered with acute points. This 

 type of mandible, which I find to be peculiar to the Attini, seems 

 to be adapted to puncturing the delicate fungus hyphae and 

 expressing their juices. Worker pupae were common in the nests, 

 but pupal males and females were much less numerous. A few 

 of the winged adult sexual forms emerged during the last week of 

 July and the first ten days of August. 



The workers were frequently observed in the act of building and 

 rearranging the particles of the substratum of the garden and 



