in size. The poor ant could not remove it herself, and, being a queen, never left the nest, so that I had no opportunity of doing so. For more than three months none of her companions performed this kind office for her.

With reference to this part of the subject, also, I have made some experiments.

January 3, 1876.-I immersed an ant (Lasius niger) in water for half an hour; and when she was then to all appearance drowned, I put her on a strip of paper leading from one of my nests to some food. The strip was half an inch wide; and one of my marked ants belonging to the same nest was passing continually to and fro over it to some food. The immersed ant lay there an hour before she recovered herself; and during this time the marked ant passed by eighteen times without taking the slightest notice of her.

I then immersed another ant in the water for an hour, after which I placed her on the strip of paper as in the preceding case. She was three-quarters of an hour before she recovered: during this time two marked ants were passing to and fro; one of them went by eighteen times, and the other twenty times; and two other ants also went over the paper; but none of them took the slightest notice of their drowned friend.

I then immersed another ant for an hour, and put her on the strip of paper. She took an hour to recover. The same two marked ants as in the previous observation were at work. One passed thirty times, the

other twenty-eight times, besides which five others passed by; but not one took the slightest notice.

I immersed three ants for eight hours, and then put them on the strip of paper. They began to recover in three-quarters of an hour, but were not quite themselves till half an hour afterwards. During the first three-quarters of an hour two marked ants passed, each four times; and two others also went by. During the following half-hour the two marked ants passed sixteen times, and three others; but none of them took any notice.

I immersed another ant for forty minutes, and put her on the strip of paper. She recovered in twenty minutes, during which time the marked ones, which were the same as in the preceding case, went by fourteen times without taking any notice.

I immersed two ants for ten hours, and then placed them on the strip of paper. The same two marked ants passed respectively eighteen and twenty-six times, and one other passed by also without taking any notice. After this I left off watching.

I immersed two ants for four hours, and then put them on the strip of paper. They began to recover in an hour, during which two marked ants, not the same as in the preceding case, passed respectively twenty-eight and ten times, and two others went by; but none of them took any notice

I immersed an ant for an hour, and then put her on the same strip of paper as in the previous cases. A marked

ant passed her twelve times; three others also went by but took no notice of her; but, on the other hand, a fourth picked her up and carried her off into the nest.

Again, I immersed an ant for an hour, and put her on the strip of paper. The marked ant passed twice, after which she did not return. Soon after, another ant came by and, picking up the immersed one, carried her off to the nest.

I do not bring forward these cases as proof or even as evidence that ants are less tender to friends in distress than previous observers have stated to be the case; but they certainly show that tenderness is not invariably the rule; and, especially when taken in connexion with the following cases, they are interesting illustrations. of the individual differences existing between ants-that there are Priests and Levites, and good Samaritans among them, as among men.

As evidence both of their intelligence and of their affection for their friends, it has been said by various observers that when ants have been accidentally buried they have been very soon dug out and rescued by their companions. Without for one moment doubting the facts as stated, we must remember the habit which ants have of burrowing in loose fresh soil, and especially their practice of digging out fresh galleries when their nests are disturbed.

It seemed to me, however, that it would not be difficult to test whether the excavations made by ants under the circumstances were the result of this general

habit, or really due to a desire to extricate their friends.

With this view I tried the following experiments:—

(1) On August 20 I placed some honey near a nest of Lasius niger on a glass surrounded with water, and so arranged that in reaching it the ants passed over another glass covered with a layer of sifted earth, about one-third of an inch in thickness. I then put some ants to the honey, and by degrees a considerable number collected round it. Then at 1.30 P.M. I buried an ant from the same nest under the earth, and left her there till 5 P.M., when I uncovered her. She was only just covered by the earth, and was none the worse, but during the whole time not one of her friends had taken the least notice of her.

(2) I arranged (September 1) some honey again in the same way. At 5 P.M. about fifty ants were at the honey, and a considerable number passing to and fro. I then buried an ant as before, taking of course one from the same nest. At 7 P.M. the number of ants at the honey had nearly doubled. At 10 P.M. they were still more numerous, and had carried off about twothirds of the honey. At 7 A.M. the next morning the honey was all gone, two or three were still wandering about, but no notice had been taken of the prisoner, whom I then let out. In this case I allowed the honey to be finished, because I thought it might perhaps be alleged that the excitement produced by such a treasure distracted their attention, or even (on the principle of doing the greatest good to the greatest number) that

they were intelligently wise in securing a treasure of food before they rescued their comrade, who, though in confinement, was neither in pain nor danger. So far as the above ants, however, are concerned, this cannot, I think, be urged.

(3) On the 8th September I repeated the experiment, burying some ants at 4 P.M. Up to 6.3 no attempt had been made to release them. I let them out and buried some more. The next morning, at 7 A.M., the honey was all gone, some ants were still wandering about, but no notice had been taken of the captives, whom I then liberated.

(4) I then (August 21) made exactly the same experiment with Myrmica ruginodis, as representing the other great family of ants, and with the same result.

In order to test the affection of ants belonging to the same nest for one another I tried the following experiments. I took six ants from a nest of F. fusca, imprisoned them in a small bottle, one end of which was covered with a layer of muslin. I then put the muslin close to the door of the nest. The muslin was of open texture, the meshes, however, being sufficiently small to prevent the ants from escaping. They could not only, however, see one another, but communicate freely with their antennæ. We now

watched to see whether the prisoners would be tended or fed by their friends. We could not, however, observe that the least notice was taken of them. The experiment, nevertheless, was less conclusive than