timal motivation? It was hoped to obtain an answer by continuance and variation of the experiment.

On February 7, the day following the fourth trial in the regular experiment, the wound chain situation was arranged as variation A at Congo's home tree (Tree 1), to which she was accustomed to be fastened for exercise and play, and with which presumably she was very familiar. The chain was wound twice counter-clockwise about one section of the tree and extended approximately five feet from the tree toward a plate containing her breakfast, the major part of which still remained despite previous work. The plate was fifteen to sixteen feet from the base of the tree and to the south of it. When brought from her cage to the experiment Congo looked up and about the tree as though appraising the situation, and then running directly to the trunk which carried the chain she climbed around it clockwise and promptly measured the length of the free chain by approaching the plate. It was insufficient to enable her to reach the food, so she hastily returned to the tree and with the slack of the chain held in one hand, as if to render more obvious its relation to the tree, she again climbed about the trunk clockwise. Then, descending to the ground, she pulled the chain to make sure that it was free and clear, and proceeded to the plate, which she easily reached with one foot and drew to her. The total time required for this successful performance was about thirty seconds. There was no hesitation whatever in initial attack on the problem or in carrying out the phases of its solution.

This response makes one still more curious as to why the gorilla should not act similarly when the same problem is set on an adjoining tree less than twenty feet away and with the chain wound similarly about a section of the trunk.

On the following day variation A of the experiment was repeated, except that the chain was wound clockwise instead of counter-clockwise. Success came so quickly that it was difficult for the experimenter to follow Congo's movements and to time the performance. Certainly it did not require more than ten seconds.

On February 9, trial 3 of variation A resulted in success, trial 5 of the regular experiment in failure, and trial 4 of variation A in success. When brought to the regular experiment for trial 5, Congo started about the trunk of the tree in the direction required to unwind the chain. Then she suddenly stopped and descended to the ground. It looked as though she knew how to solve the problem, but for some reason did not care to try. On the chance that motivation was inadequate I promptly added to the sweet potatoes already on the plate two ripe bananas, of which she was inordinately fond, and an apple. She looked intently at the offering, seemed interested and eager for it, but made no effort to get it. Instead, for the greater part of the thirty-minute interval, she either climbed about the tree gathering moss, leaves, and bark, or played on the ground near the trunk.

Again it should be emphasized that effort was made to present the same problem in each situation; namely, in variation A at Tree 1, to which Congo was accustomed to be moored, and in the regular experiment at Tree 2, a similar tree only a few yards from the familiar oak. Somewhat different types of chain were used initially, but it was subsequently proved by experiment that this did not influence the behavior.

Beginning on February 10, settings of the regular experiment and of variation A were presented daily in that order. At the home tree conditions were made increasingly difficult by complex windings of the chain, whereas at the other tree they were rendered as simple as possible. Various sections of the two trees were used, and in several of the trials two or three sections were employed simultaneously, the chain often being wound in opposite directions about adjacent portions of the trunk in order to render the task of unwinding more complex and to give me opportunity to note Congo's observational ability and the manner in which she followed the course of the chain.

On February 10, trial 6 of the regular experiment yielded failure, trial 5 of variation A success. In the latter case the

problem was more complicated than previously and she made. a number of false moves, spending nearly a minute in solving her problem. The outstanding features of her performance were pauses and obvious visual inspection of the relation of the chain to the tree, especially when she made a mistake. Usually she saw what should be done, but often her movements tended to get ahead of her visual observation and she was compelled to retrace her steps and by actually pulling on the chain with her hand test its freedom of motion and discover anew the direction which she should take.

Again, on February 11, trial 7 of the regular experiment exhibited only initial movement toward the unwinding of the chain, whereas trial 6 of variation A yielded prompt solution, despite slight desire for the reward of food.

On February 12 the results were similar in that the regular experiment, even with an extreme simplification of the problem, failed to command Congo's interested attention, whereas trial 7 of variation A, with a new and somewhat complicated setting involving three sections of the trunk of the tree, yielded solution, after a number of errors, in about one minute. The latter trial clearly exhibited Congo's method. At each critical point in the process of unwinding she stopped to observe the direction which the chain took, gathered it up in one hand and either pushed or threw it before her so that it should be clear of the tree trunk and she should not have to retrace her steps and run the risk of getting it tangled. This technique, doubtless the product of long experience, is excellent. Yet again on February 13 the two forms of experiment were presented successively, with complete failure as usual at the unfamiliar tree and success at the home tree.

On subsequent days between February 17 and March 5, four additional trials, 9 to 12, were given in variation A, the conditions being varied for the purpose of exhibiting observational ability, method, and the effect of exchanging the chains which had been used in the two forms of experiment. Thus, in trials 10 and 11 the other chain was substituted for the one regular

ly used at Congo's home tree. She nevertheless worked diligently at her problem and solved it. This proves that the relatively unfamiliar chain did not inhibit effort.

The attempt to discover reasons for Congo's failure in the regular form of the experiment was continued through a total of sixteen trials, of which numbers 10 to 16 were given between February 15 and March 5. Failure is recorded for trials 10 and 11. In trial 12, on February 18, the familiar chain from the home tree was used at Tree 2. It was wrapped about a section of the trunk three times clockwise, with six feet free. Although she made no attempt to unwind the chain, she two or three times fingered and looked at it as though interested, and twice I thought she was about to start to climb about the trunk. At this point it may be said that no variations, such for example as the exchange of chains in the two forms of experiment, essentially altered Congo's attitudes toward the problem in the diverse settings or her performance.

Finally, on February 19, I decided to concentrate the entire resources of the day on this particular experiment and to offer what would be relatively a magnificent reward for effort. To this end the experimental situation was set as usual; the familiar chain from the home tree was employed at the unfamiliar tree, and the food plate, sixteen feet northward of the tree, contained two large sweet potatoes, two baked bananas, and an orange. Prior to the experiment Congo had taken nothing except a cup of milk for approximately fifteen hours. Several times prior to the trial, which was made at 10:02, she had vocally begged for food while waiting in her cage. When brought to the experiment and attached to the chain she seemed quite ready to work. Within two minutes she began to fool with the chain. Then she moved toward the plate as far as the chain would permit. Returning she walked about the tree in various directions, now gazing toward the house or elsewhere, and again walking about restlessly. It may here be remarked that restlessness is ordinarily a good sign in a problematic situation. It seems to indicate that hope of success has not been abandoned and that things are happening inside.

At precisely 10:08 Congo started about the trunk of the tree counter-clockwise, but she retraced her steps without completing the turn. Fingering the chain as if to satisfy herself that she had started in the right direction, she again moved about the tree, holding the chain in front of her, and this time passing all the way around the trunk in the correct direction. She next descended to the ground and taking the chain in hand proceeded to go around the tree trunk again in the same fashion. Thus she completed the unwinding of the chain and with its full length available, walked deliberately to the plate and emptied it of its contents.

In addition to the unusually large reward offered and the fact that she had not previously had any portion of her breakfast except her cup of milk, an exceptional condition in this trial was an approaching shower. It is entirely possible that dislike of being away from the shelter of her cage during rainfall may have stirred her to attempt the early completion of her task. At any rate the trial demonstrated that under appropriate conditions she could as readily solve the wound chain problem at the unfamiliar tree as at the home tree. Whether the difference in attitude in the two locations is due to motivation or to specific inhibiting factors does not yet appear.

Naturally it appealed to the experimenter as of considerable interest to discover whether the barriers having once been passed, Congo would continue to work smoothly and successfully in the regular wound chain experiment. Therefore on the following day, in trial numbered 14, she again faced the problem which had been so long unsolved, but which in trial 1 and again in trial 13 she had shown herself capable of solving. In this case the setting of the experiment was as usual. Other experiments had preceded this particular one, Congo had had a portion of her breakfast, and on the plate appeared one baked banana and half of a sweet potato. With little hesitation she attacked the problem and in less than twenty seconds she had succeeded in unwinding the chain and in obtaining her reward. It would seem then that resistance to effort had been over