again and thrown away like the rest. Under such circumstances, then, it seems that ants make no difference between friends and strangers.
It may, however, be said in this experiment that, as ants do not recover from chloroform, and these ants were therefore to all intents and purposes dead, we should not expect that much difference would be made between friends and strangers. I therefore tried the same experiment, only, instead of chloroforming the ants, I made them intoxicated. This was a rather more difficult experiment. No ant would voluntarily degrade herself by getting drunk, and it was not easy in all cases to hit off the requisite degree of this compulsory intoxication. In all cases they were made quite drunk, so that they lay helplessly on their backs. The sober ants seemed much puzzled at finding their friends in this helpless and discreditable condition. They took them up and carried them about for a while in a sort of aimless way, as if they did not know what to do with their drunkards, any more than we do. Ultimately, however, the results were as follows: The ants removed twenty-five friends and thirty strangers. Of the friends, twenty were carried into the nest, where no doubt they slept off the effect of the spirit—at least, we saw no more of them—and five were thrown into the water. Of the strangers, on the contrary, twenty-four were thrown into the water; only six were taken into the nest, and four of these were shortly afterward brought out again and thrown away.
The difference in the treatment of friends and strangers was, therefore, most marked.
Dead ants, I may add, are always brought out of the nest, and I have more than once found a little heap on one spot, giving it almost the appearance of a burial-ground.
I have also made some experiments on the power possessed by ants of remembering their friends. It will be recollected that Huber gives a most interesting account of the behavior of some ants, which, after being separated for four months, when brought together again, immediately recognized one another, and "fell to mutual caresses with their antennæ." Forel, however, regards these movements as having indicated fear and surprise rather than affection, though he also is quite inclined to believe, from his own observation, that ants would recognize one another after a separation of some months. The observation recorded by Huber was made casually; and neither he nor any one else seems to have taken any steps to test it by subsequent experiments. The fact is one, however, of so much interest, that it seemed to me desirable to make further experiments on the subject. On the 4th of August, 1875, therefore, I separated one of my nests of F. fusca into two halves, which I kept entirely apart.
I then from time to time put an ant from one of these nests into the other, introducing also a stranger at the same time. The stranger
