wings often interlocked, during their long imprisonment in a dark, subterranean chamber. When, after a good rest in this position, they were brought to right-about-face, two immediately moved backward a few steps, and came to rest again (Figs. 31 and 32). Move around them in circle, and not a head is turned; make a pass of the hand suddenly towards them, and sometimes there is a slight backward movement, but sometimes there is none; reverse again with heads turned away, and two turn part way around, one of which repeats the movement when repeatedly reversed. If placed on their backs, they will slowly right themselves, though if in the nest-hole they were sometimes contented with this position for a longer time.
I am now convinced that the earlier conclusion was correct, and that the peculiar actions described are due to habit, learned underground, and in relation to getting their food, although I was probably in error in supposing that their instinct of fear was ordinarily delayed until they were ready to leave the tunnel. When such birds are handled daily there can be little doubt that this instinct is liable to be checked. The arched chamber of sand in which these kingfishers lay was 1112 inches in diameter, and 612 inches high, while the tunnel leading to it had a 3-inch bore. Its temperature stood at 77° F., and was seven degrees cooler than that of the air outside. Since these five birds were each about six inches long it is evident that they were closely packed, and that once in the tunnel, no turning movement would be possible, any more than for the adult which after feeding always backs out of the hole. Now we have earlier noticed a tendency among the older young to crawl down the passageway, and meet the parent at the mouth, but that they are hustled back and presumably fed at the nest, hence the probable association between walking backwards and getting food, and hence the curious habit displayed by these birds when they are taken from the ground.
III. Intelligence in Adult Birds
There seems to be little intelligence displayed by birds in regard to the quantity of food served at the nest. What one of the altricious kinds really does in effect is to "test" the reaction of the throat of its nestling, and to await the response. If this is not forthcoming the food is quickly withdrawn, and another is tried. The most responsive bird gets the food, and there is no distribution on any other basis than this. The same bird may thus be fed twice or even three times in succession, the strongest usually getting the most, and the amount which it can take being reflexly determined by the gullet.
On the other hand, intelligence is certainly shown in the kind of food served, and in the treatment which it often receives. Thus a gull
