164 



C. H. TURNER. 



The larger particles were carried in the mouth and deposited 

 therein. Every few moments the ant would enter the burrow 

 and compact the sand with its head. Frequently it would leave 

 the hole and wander about for a few moments. It always soon 

 returned to the covering. After the hole had been filled to the 

 level of the ground, the ant continued to pile on sand. When I 

 left, at 10:03 A. M., the ant was still piling on sand. At no 

 time was it assisted by any other ant, although numerous ants 

 passed by it. 



With few exceptions, on each mound there is one or more trash 

 piles. In over two hundred mounds examined I noticed only one 



a 



FIG. 3. FIG. 4. 



FIG. 3. Mound of P. badins. a, entrances. The mound is covered with peb- 

 bles, and dead stalks of grass and weeds are scattered over it. Greatest length, 32 

 in. ; greatest width, 25 in. ; greatest height, 5 in. 



FIG. 4. Mound of P. badius, surface view, a, entrances; b, debris, mostly plant 

 matter, from the nest. The mound is covered with pebbles. On the outer edge 

 there are a few stalks of grass. At one time this mound had four openings. Greatest 

 length, 32 in. ; greatest width, 27 in. 



exception (Fig. 3). On this trash pile is dumped debris from the 

 nest, supplemented by plant material obtained in the surrounding 

 territory. The major portion of the pile is composed of plant 

 matter, the remainder of remnants of insects. In some cases the 

 trash-pile surrounds the nest-opening (Figs. I, 5, 7, 8); but in 

 most cases it does not seem to bear any definite relation to that 

 opening (Figs. 2, 4, 6). 



It seems self-evident that a mound as conspicuous as this must 

 play some important role in either the present or the past behavior 

 of the species. It is possible that it serves more than one pur- 



