234 EVIDENCE AGAINST THE 



ditch, while they carried their own friends into the 

 nest, where no doubt they slept off the effects of the 

 spirits. This experiment seemed to show that the 

 recognition was nut effected by means of any sign ; 

 but I thought the suggestion might be tested in 

 another way. 



I made, therefore, the following experiment. I took a 

 few specimens of Formica fusca from two different nests, 

 which I will call A and B, and placed them together. 

 At first they were rather shy ; but after a while they 

 fraternized. After they had lived amicably together 

 for three months, I put two of these ants from nest A 

 into nest B; but they were soon attacked vigorously 

 and driven out of the nest. I thought it desirable to 

 repeat and extend this test. Accordingly, on June 

 16 I put three specimens of F. fusca from my nest 

 No. 81 with the same number from nest No. 71. Then 

 on September 19, one of the six having died in the 

 interval, I put the two from nest 81 into nest 71, and 

 the three from nest 71 into 81. They were all attacked, 

 though not very quickly or vigorously, but eventually all 

 five were expelled. 



Again, on September 25 I took three ants from each 

 of these nests and put the six together. Then on 

 March 19 following (one having died), I put the 

 two from 71 into 81, and the three from 81 into 71. 

 They were all attacked, so that they were evidently 

 recognized as strangers ; hut it seemed to me that the 

 attack was less vigorous, and I could not be sure that 

 they were either kdled or driven out. In the course of 

 the we.k three or four dead ants were brought out of 

 each of the nests; but I could not feel certain that 

 they weie tbose experimeiited with. 



