tion, which at first sight seems very plausible, might be raised against the assertion that the disproportion between dream content and dream thought justifies the conclusion that an abundant condensation of psychic material has taken place in the formation of dreams. For we so often have the impression that we have dreamed a great deal throughout the night and then have forgotten the greater part. The dream which we recollect upon awakening would thus be only a remnant of the total dream-work, which would probably equal the dream thoughts in range if we were able to remember the former completely. In part this is certainly true; there can be no mistake about the observation that the dream is most accurately reproduced if one tries to remember it immediately after awakening, and that the recollection of it becomes more and more defective towards evening. On the other hand, it must be admitted that the impression that we have dreamed a good deal more than we are able to reproduce is often based upon an illusion, the cause of which will be explained later. Moreover, the assumption of condensation in the dream activity is not affected by the possibility of forgetting in dreams, for it is proved by groups of ideas belonging to those particular parts of the dream which have remained in the memory. If a large part of the dream has actually been lost to memory, we are probably deprived of access to a new series of dream thoughts. It is altogether unjustifiable to expect that those portions of the dream which have been lost also relate to the thoughts with which we are already acquainted from the analysis of the portions which have been preserved.
In view of the great number of ideas which analysis furnishes for each individual element of the dream content, the chief doubt with many readers will be whether it is permissible to count everything that subsequently comes to mind during analysis as a part of the dream thoughts—to assume, in other words, that all these thoughts have been active in the sleeping state and have taken part in the formation of the dream. Is it not more probable that thought connections are developed in the course of analysis which did not participate in the formation of the dream? I can meet this doubt only conditionally. It is true, of course, that particular thought
