POWEKS OF COMMUNICATION. 169 



friction of my finger ; on the other hand, the bridge (d) 

 had retained the scent, but was so placed as to lead 

 away from the larvse ; and it will be seen that, undei 

 these circumstances, out of 41 ants which found theii 

 way towards the larvse as far as e, 14 only passed over 

 the bridge / to the larvse, while 27 went over the 

 bridge d to the empty glass m. 



Taking these observations as a whole, 150 ants 

 came to the point e, of which 21 only went on to the 

 larvse, while 95 went away to the empty glass. These 

 experiments, therefore, seem to show that when an 

 ant has discovered a store of food and others flock to 

 it, they are guided in some cases by sight, while in 

 others they track one another by scent. 



I then varied the experiment as follows : — I put an 

 ant (Z. niger) to some larvse as usual, and when she knew 

 her way, I allowed her to go home on her own legs ; but 

 as soon as she emerged from the nest, if she had any 

 friends with her, I took her up on a bit of paper and 

 carried her to the larvse. Under these circumstances very 

 few ants indeed found their way to them. Thus, on June 

 23, 1876, at 5.30, an ant which had been previously 

 under observation was put to some larvse. She took 

 one and returned as usual to the nest. At 5.34 she 

 came out with no less than 10 friends, and was then 

 transferred to the larvse. The others wandered about a 

 little, but by degrees returned to the nest, not one of 

 them finding their way to the larvse. The first ant 

 picked up a larva, returned, and again came out 

 13 



