EXPERIMENTS. 261 



oon as she came to the end of the bridge the turned 

 round. 



I then modified the arrangement, placing between 

 the nest and the food three similar pieces of wood. 

 Then when the ant was on the middle piece, I trans- 

 posed the other two. To my surprise this did not at 

 all disconcert them. 



I then tried the arrangement shown in Fig. 24. 



Fig. 24. 



a is a paper bridge leading to the nest ; 6 is a board 

 about 22 inches long by 13 broad, on which is a disk of 

 white paper fastened at the centre by a pin d ; e is 

 some food. When the ants had come to know their way 

 BO that they passed straight over the paper disk on their 

 way from a to e, I moved the disk round with an ant 

 on it, so that / came to g and g to/. As before, the 

 ants turned round with the paper. 



As it might be possible that the ants turned 

 rim(i on account of the changed relative position 

 of external objects, I next substituted a circular 

 box 12 inches in diameter, open at the top, and 



