say to the people whom we hypnotize: “Sleep,” and the suggestions made to them are comparable to the dreams of natural sleep. The mental situation is really analogous in the two cases. In natural sleep we withdraw our interest from the whole outer world; so also in hypnotic sleep, with the exception of the one person who has hypnotized us and with whom we remain in rapport. Again, the so-called “nurse’s sleep” in which the nurse remains in rapport with the child and can be wakened only by him is a normal counterpart of hypnotic sleep. So it does not seem so very audacious to carry over to natural sleep something which is a condition in hypnosis. The assumption that some knowledge about his dream exists in the dreamer and that this knowledge is merely inaccessible to him, so that he himself does not believe he has it, is not a wild invention. Incidentally, we observe here that a third way of approaching the study of dreams is thus opened out for us; we may approach it by the avenue of sleep-disturbing stimuli, by that of day-dreams, and now by that of the dreams suggested during hypnosis. Now perhaps we shall return to our task with greater confidence. We see it is very probable that the dreamer knows something about his dream; the problem is how to make it possible for him to get at his knowledge and impart it to us. We do not expect him immediately to tell us what his dream means, but we do think he will be able to discover its source, from what circle of thoughts and interests it is derived. With errors, you will remember the man was asked how the slip of the tongue “refilled” had come about, and his first association gave us the explanation. The technique we employ in the case of dreams is very simple and is modelled on this example. Here again we shall ask the dreamer how he came to have the dream, and his next words must be regarded as giving the explanation in this case also. It makes no difference to us therefore, whether he thinks that he does or does not know anything about it, and we treat both cases alike. This technique is certainly very simple, nevertheless I am afraid it will provoke most strenuous opposition in you. You will say: “Another assumption, the third! And the most improbable of all! When I ask the dreamer what ideas come to him about the dream, do you mean to say that his very first association will give the desired explanation? But surely he might have no association at all, or heaven only knows what the association might be. We cannot imagine upon what grounds such an expectation is based. It really implies too much trust in Providence, and this at a point where rather more exercise of the critical faculty would better meet the case. Besides, a dream is not like a single slip of the tongue but is made up of many elements. That being so, upon which association is one to rely?”