87 2 EXPERIMENTS ON 



withont counting these, from 3.20 to 8 o'clock, out of 

 fon^-five ants, twenty-nine went up the right pin, while 

 sixteen went up the nine wrong ones. 

 Thus on 



July 13, out of 27 ants, 19 went right and 8 wrong. 

 „ 15 „ 29 „ 17 „ 12 „ 



„ 16 „ 30 „ 16 „ 14 „ 



„ 18 „ 26 „ 23 „ 3 „ 



„ 19 „ 45 „ 29 „ 16 „ 



Or adding them all together, while there were never 

 less than ten pins, out of 157 ants, 104 came up the 

 right pin, and only 53 up the others. 



I was at first disposed to infer from these facta that 

 the first ant must have described the route to its friends, 

 but subsequent observations satisfied me that they 

 might have found their way by scent. 



