to get at it. If it is found to be edible (often even when not) it is beaten on the ground and "killed". If good to eat it usually is eaten. In the case of these dead mice, relatively unattractive food, they received the semblance of animated creatures disappearing under a rock (known to have harbored similar refugees in the past) thus presenting exciting possibilities which overcame their indifference of minutes before. Completion of the "pattern", I suppose, called for their being killed and eaten. It is interesting to watch the expanding field of associations growing out of Archie's recognition of the worm box as a container of food. (I do not know the "language"; what I mean is that the to think number of acts of mine which he appears, should lead up to the production of the box for his benefit is growing greater). Thus in the north west corner of cage C is a small bench upon which I often sit and sometimes when there give him worms. If I merely sit there now he is apt to jump up to my knee and wait indefinitely for developments, sometimes examining my clothes. (If there are none he may drop to the ground, mew about my feet, pick up a small stone and hammer it on my shoes, or run his head, neck and shoulders up a trouser leg and snap the elastic of my garters). A movement of a hand to the small pocket of the coat where the box is kept often causes him to jump up again. This pocket is inside of a larger one once and today when I was sitting on the same bench, not thinking of offering him worms, I reached into the larger pocket and produced a package of cigarettes. (K.D. saw this). Archie was up instantly and drew out a cigarette neatly. He must have started when he saw the movement of the hand into the pocket. Again the box can be made to give out a snapping sound (something in the manner of those "frogs" which lecturers use to signal the operator of a projector). I have not yet deliberately tried to attract him by making this sound, but once or twice have made it