238

EXPERIMENTS WITH DROWNED ANTS.

It

The next morning I could only see two, but none were being attacked, and there were no dead ones. is probable that the paint had been cleaned off the others, but it was not easy to find them all among so many. At any rate, none were being attacked, nor had any been killed.

These observations, therefore, quite confirm those previously made, and seem to show that if pupæ are taken from a nest, kept till they become perfect insects, and then replaced in the nest, they are recognized as friends.

As regards the mode of recognition, Mr. Mc Cook considers that it is by scent, and states that if ants are more or less soaked in water, they are no longer recognized by their friends, but are attacked. He mentions a case in which an ant fell accidentally into some water: "She remained in the liquid several moments, and crept out of it. Immediately she was seized in a hostile manner, first by one, and then another, then by a third, the two antennæ and one leg were thus held. A fourth one assaulted the middle thorax and petiole. The poor little bather was thus dragged helplessly to and fro for a long time, and was evidently ordained to death. Presently I took up the struggling heap. Two of the assailants kept their hold, one finally dropped; the other I could not tear loose, and so put the pair back upon the tree, leaving the doomed immersionist to her hard fate."

His attention having been called to this, he noticed several other cases, always with the same result. I have not myself been able to repeat the observation with the same species, but with two at least of our native ants the results were exactly reversed. In one

RECOGNITION AFTER A YEAR AND NINE MONTHS. 239

case five specimens of Lasius niger fell into water and remained immersed for three hours. I then took them out and put them into a bottle to recover themselves. The following morning I allowed them to return. They were received as friends, and, though we watched them from 7.30 till 1.30 every hour, there was not the slightest sign of hostility. The nest was, moreover, placed in a closed box, so that if any ant were killed we could inevitably find the body, and no ant died. In this case, therefore, it is clear that the immersion did not prevent them from being recognized. Again, three specimens of Formica fusca dropped into water. After three hours I took them out, and, after keeping them by themselves for the night to recover, I put them back into the nest. They were unquestionably received as friends, without the slightest sign of hostility or even of doubt. I do not, however, by any means intend to express the opinion that smell is not the mode by which recognition is effected.

It will be remembered, perhaps, that my ants (Formica fusca) recognized one another after a separation of a year and nine months, though "after some months' separation they were occasionally attacked, as some of the ants, perhaps the young ones, did not recognize them. Still, they were never killed or driven out of the nest, so that evidently when a mistake was made it was soon discovered." Hence it would appear that there are differences in the memory of different species.

In one case Forel had taken some ants from a large nest of Componotus, for the experiments on their sensibility to the ultra-violet rays, to which I have already referred. After his observations were

240

SUPPOSED RECOGNITION BY SCENT.

concluded, he returned them to the nest, some after eight, some after forty-one days. Those which were returned after eight days were at once recognized, while as regards those which had been forty-one days away from home. "On reculait de part et d'autre, se menaçait des mandibules, s'examinait à fond avec les antennes, se mordait même. Plusieurs même allèrent dans leur irritation jusqu'à essayer de décapiter et même à décapiter quelques-unes de leurs anciennes compagnes et soeurs avec leurs mandibules (c'est le mode de combat des Camponotus)! Les fourmis vernies prirent part à ces rixes aussi bien que les non vernies; je les vis même attaquer, et elles étaient à peine moins adroites. Les combats ne cessèrent entièrement qu'au bout d'un ou deux jours, et, à part les quelques victimes du premier jour, l'incident se termina par une alliance."

Forel seems to entertain no doubt that the recognition is effected by a form of smell, which he terms "odorat au contact." He says, "Beaucoup d'insectes ont en outre une sorte d'odorat au contact que nous ne possédons pas et qui permet entre autres aux fourmis de distinguer leurs compagnes de leurs ennemies."

His observations, however, do not favour the hypothesis that the recognition may be by smell. If the ants recognized their companions by any odour characteristic of the community, the lapse of thirty days could not have made any difference. Here the question of memory would not enter, because the perception of the odour would in both cases be continually before them. M. Forel is so excellent an observer, and has so great a knowledge of the ways of ants, that his opinion is entitled to great weight. It

RECOGNITION BY MEANS OF THE ANTENNÆ. 241

would be very interesting to repeat similar observations, for if it turn out to be the case that separations of comparatively few days lead, in some species, to a want of recognition, it would be a strong argument against the hypothesis that this recognition is due to smell.

It certainly seems as if the recognition was effected to a great extent by the antennæ. Not only do the ants cross and recross them, almost, so to say, as two deaf mutes conversing by their fingers; but, as M. Forel has shown, if ants of different species are brought together after the removal of their antennæ they show no signs of hostility. That this latter statement is correct I am quite content to take on M. Forel's authority; but it is not so conclusive as might seem at first sight, because in ants, as in men, "a fellowfeeling makes us wondrous kind," and ants when isolated, and especially when suffering, are much less pugnacious than they are under normal conditions.

R

( 242 )

CHAPTER XII.

ON THE INSTINCTS OF SOLITARY WASPS AND BEES.

THE hive bee and the common wasps are so familiar and so interesting that they have to a great extent diverted attention from the so-called solitary species of the same groups. Few, for instance, are aware that about 4500 species of wild bees are known, and of wasps 1100, of which some 170 and 16 respectively live in Britain.

These insects often live in association, but do not form true communities. Speaking generally, we may say that each female constructs a cell, every species having its own favourite site, sometimes underground, sometimes in a hollow stick, in an empty snail-shell, or built against a wall, a stone, or the branch of a tree. Having completed her cell, the female stores up in it a sufficient supply of food, which in the case of bees consists of pollen and honey; while the wasps select small animals, such as beetles, caterpillars, spiders, etc., each species generally having one kind of prey. The mother then lays an egg, after which she closes up the cell, and commences another. Having thus provided sufficiently for her offspring, she generally takes no further heed of it. This is not, however, an invariable rule in the genus Bembex, for instance, the