33§ C. H. TURNER. 



they or their immediate ancestors had ever experienced a nest 

 with a crack leading into the brood chamber. 



To make sure that the covering of the cracks mentioned above 

 was not a mere coincidence, I removed gently the trash that was 

 covering crack b. In less than an hour a few ants were busy 

 covering it. At intervals of about a week, this experiment was 

 repeated twelve times ; always yielding the same results. Usu- 

 ally one or two ants did the covering ; at no one time have I 

 seen more than four thus occupied. 



That the ants were bent not on covering just any cracks that 

 entered the nest, but only the cracks that affected the brOod 

 chamber is evidenced by the fact that, although these experi- 

 ments covered a period of eight months, at no time did the ants 

 cover the crack a or the free edges of the glass covers of com- 

 partments A and B (Fig. i). Furthermore ants bearing trash 

 would frequently cross the edges of the covers to A and B and 

 even the crack a and pass on and deposit their burdens on the 

 crack b or the edges c, d, or f. There is yet another evidence 

 of this statement. 



If this behavior is for the purpose of covering cracks the exist- 

 ence of which alters the conditions in the brood chamber, a 

 crack crossing the brood chamber should produce a greater dis- 

 turbance and hence should be covered by the ants first. There- 

 fore I divided the cover to the living chamber C into two equal 

 parts which were so adjusted as to leave between them a trans- 

 verse crack e (Fig. 2) wide enough to make quite an opening yet 

 too narrow to permit the passage of ants to and fro. Whenever 

 this was done, and it was repeated over a dozen times, the first 

 crack to be covered was c, and after e, b (Fig. 2). In each case 

 only a few ants covered the cracks. Usually only one or two 

 were thus employed ; at no one time were over six thus occu- 

 pied. Remember that that nest contained at least two hundred 

 workers. 



Whenever the trash was removed from crack e (Fig. 2), were 

 it done ever so gently, the induced restlessness of the ants within 

 the nest indicated that they were much disturbed. If I gently 

 breathed against the uncovered crack, the ants within rushed 

 about in all directions as though panic stricken. To one who 



