214 JOHN L. PRICER. 



This experiment was repeated a number of times and each trial 

 gave unmistakable evidence that the ant which discovered the 

 larvae in some way conveyed the intelligence to others. I have 

 also had similar evidence of communication when I placed the 

 carcass of a white grub in the feeding-room of a colony that had 

 been deprived of animal food for some time. In the succeeding 

 trials of this experiment, however, it never happened again that 

 all the larvse were returned to the nest through the straight tube. 

 They were as often carried over the vertical loop as through any 

 other tube, and this happened just as frequently after the ants had 

 used the tubes for three or four months as at first. The ants 

 always work as if in great haste to get the larvse back to the nest, 

 and it seems that if they had had any discretion whatever, they 

 would have chosen the shorter and less difficult- route. In one 

 instance, however, I watched an ant while making six trips from 

 a cluster of larvse which had been carried into the straight tube. 

 Three of the six times, when returning to the larvse, she entered 

 the arched tube from the vestibule and proceeded until she came 

 to the incline and then each time turned about and found the 

 straight tube and the larvse. She seemed to remember that the 

 larvae were in the straight tube and so knew that she was in the 

 wrong tube when she came to the incline. She also seemed not 

 to be tracking herself as she returned to the larvae. 



Experiment V. 



Three islands were formed by inverting two-inch Petri dishes 

 in four-inch ones and filling the larger ones with water. One of 

 these islands, which I will designate as A, was connected with the 

 feed-room of the apparatus used in experiment IV., by means of 

 a glass tube which was bent in such a way as to be partly im- 

 mersed in water in a Petri dish. This made it impossible for the 

 ants to crawl back to the feeding-room over the tube and escape 

 to the table. The other islands, which I will designate as B and 

 C respectively, were connected with island A by means of bridges 

 of cardboard eight inches long and one half inch wide. B was 

 placed in a direct line with the tube leading to A, and C was 

 placed at right angles to the tube, opposite A. 



Larvse taken from the colony in the nest were placed on B and 



