24 



scatter them about outside the nest. The ants were not repelled by the 

 moist sand, but seemed to dislike its physical properties. By 11 : 30 

 almost every particle of the sand had been removed, and the ants had 

 placed a few larvae on the square of tin on which the sand rested. By 

 1 : 30 p. m., all the sand removed, some of it being carried out and 

 thrown into the water. One of the three bunches of larvae under the 

 orange cover was now on the tin. Time, 3 hours and 5 minutes. 



Experiment 5762, Kerosene. — January 13, 1 : 50 p. m. A small 

 amount of sand moistened with kerosene was placed under the center 





/c';-V;f'V' 



■ " ' Queen 



Fig. 2 



of the orange cover. The ants nearest at once faced towards it, waved 

 their antennas about, and then ran around in confusion. A very few of 

 the ants approached the sand, and even touched it, but quickly jerked 

 back, throwing their antennae about, vigorously, seemingly in great dis- 

 tress. Others circled completely around the sand at a distance of 

 one or two inches. After 15 minutes a few of the ants with some of 

 their larvae were bunched under one corner of the orange cover. In 

 55 minutes, about a fourth of the ants were under the edge of the 

 cover, another fourth were an inch and a half outside the cover, with 

 the queen, while the other half were assembled at a remote corner of 

 the base, as shown in the accompanying sketch (Fig. 2). January 

 14, 9:45 a. m. (19 hours, 5 minutes ), situation about as on the previous 

 day, except that approximately half the ants were under the cover, 

 none of them, however, nearer than five-eighths of an inch to the re- 

 pellent. Most of the ants under the cover were in the neighborhood 

 of the queen, who was just an inch from the edge of the sand. The 



