SIE JOHN LrBBOCK ON ANTS, BEES, AND WASPS. 



277 



the line de was now outside them. This change, however, did 



not at all discompose Kg. 2. 



the ants ; but instead 



of going, as before, 



through the tunnel and 



between the rows of 



bricks to a, they walked 



exactly along the old 



path to e. 



I then arranged mat- 

 ters as before, but with- 

 out the tunnel and with 

 only three pairs of bricks 



(fig. 3). When an ant had got quite used to the path dto e, 

 I altered the position of the 

 bricks and food to/" (fig. 4), 

 making a difference of 8 

 inches in the position of the 

 latter. The ant came as before, 

 walked up to the first brick, 

 touched it with her antennae, 

 but then followed her old 

 line to a. From there she 

 veered towards the food, and 

 verysoonfoundit. When she 

 was gone, I altered it again, 

 as shown in fig. 5 ; she re- 

 turned after the usual inter- 

 val, and went again straight 

 to a; then, after some 

 wanderings, to /, and at 

 length, but only after a lapse 

 of 25 minutes, found the 

 food at ff. These experi- 

 ments were repeated more 

 than once, and always with 

 similar results. I then va- 

 ried matters by removing 

 the bricks, which, however, 

 did not seem to make anj' 

 difierence to the ants. 



