530 PROCEEDINGS OF THE ACADEMY OF [Oct., 



Strips of paper or of wood-shavings laid over the trail. When the 

 ant had well established her trail across the paper or shaving I could 

 sometimes in her absence change it, replacing it exactly by a new 

 one of the same color and material, and I have gradually increased 

 its width from five millimeters to fifteen, or from one to three times 

 her length, without causing the slightest distraction of the ant from 

 her steady journeys to and fro over it." This proves that the ant 

 does not smell her way at every point, and that familiarity with 

 certain objects under her feet is gradually acquired. A dissimilar 

 object, or an old object in a different place, never failed to distract 

 the ant. The frequent placing of new objects upon her path, or 

 repeated interruptions of her vvork, always caused her to change 

 her route or to abandon her work. 



]\rany ants, where records were kept, gained speed in the carry- 

 ing in of pupje, the number of journeys accomplished during an 

 hour increasing always with experience of the runs, unless special 

 hindrances occurred. If there be no greater stimulus in the greater 

 amount of scent laid down, the gain in speed must arise from 

 added familiarity with the road. 



I occasioned one of my colonies to move from a Lubbock nest to 

 its annex over a bridge eight inches long, once or twice a day 

 during ten days, and the colony gradually reduced the time 

 required for a complete change of loca'ion from over an hour to 

 twenty minutes. 



When an ant discovers a barrier across one of the runs in the 

 maze, she does not more than two or three times follow her trail 

 to the barrier, but altogether changes her route. In the change 

 she does not merely cut off the loop in her own path, but she 

 frequently takes a different direction. 



erefl from shock-effect and by their activities in()icated their readiness for use 

 in experiment. Before their recovery the ants were listless and al)nornially 

 irritable ; and they attacked with self-destructive violence any moving thing 

 that touched them. One antenna performs all the functions of a pair. In 

 examining hundreds of ants, I found many with a single antenna, or with 

 one antenna and the long proximal joint of the other, and these ants, in- 

 cluding queens so maimed, were living normal lives. But I never found in 

 its native nest any antennaless ant. The sense of taste is not lost Avith the 

 antennai. Ants kept without food for three days lapped honeyed cake with 

 evident relish immediately after they were deprived of their antennae. 

 Their sensitivity to light, heat and humidity also remains unimpaired. No 

 part of an antenna that had been clipped was regenerated during three sub- 

 sequent months that I kept the clipped ants under my observation. 



* My best results were in using moistened brown blotting paper, care being 

 taken that its edges across the path should be exactly even with the surface 

 of the earth covering the floor. 



