*(2) Dreams which have occurred during the first years of childhood, and which have sometimes been retained in the memory for decades with perfect sensorial freshness, are almost always of great importance for the understanding of the development and the neurosis of the dreamer. The analysis of them protects the physician from errors and uncertainties which might confuse him even theoretically.

I will now make a few rather unsystematic remarks relating to the interpretations of dreams, which will perhaps serve as a guide to the reader who wishes to test my assertions by the analysis of his own dreams.

He must not expect that it will be a simple and easy matter to interpret his own dreams. Even the observation of endoptic phenomena, and other sensations which are commonly immune from attention, calls for practice, although this group of observations is not opposed by any psychic motive. It is very much more difficult to get hold of the unwished ideas. He who seeks to do so must fulfil the requirements laid down in this treatise, and while following the rules here given, he must endeavour to restrain all criticism, all preconceptions, and all affective or intellectual bias in himself during the work of analysis. He must be ever mindful of the precept which Claude Bernard held up to the experimenter in the physiological laboratory: "Travailler comme une bete"- that is, he must be as enduring as an animal, and also as disinterested in the results of his work. He who will follow this advice will no longer find the task a difficult one. The interpretation of a dream cannot always be accomplished in one session; after following up a chain of associations you will often feel that your working capacity is exhausted; the dream will not tell you anything more that day; it is then best to break off, and to resume the work the following day. Another portion of the dream-content then solicits your attention, and you thus obtain access to a fresh stratum of the dream-thoughts. One might call this the fractional interpretation of dreams.

It is most difficult to induce the beginner in dream- interpretation to recognize the fact that his task is not finished when he is in possession of a complete interpretation of the dream which is both ingenious and coherent, and which gives particulars of all the elements of the dream-content. Besides this, another interpretation, an over-interpretation of the same dream, one which has escaped him, may be possible. It is really not easy to form an idea of the wealth of trains of unconscious thought striving for expression in our minds, or to credit the adroitness displayed by the dream-work in killing- so to speak- seven flies at one stroke, like the journeyman tailor in the fairy-tale, by means of its ambiguous modes of expression. The reader will constantly be inclined to reproach the author for a superfluous display of ingenuity, but anyone who has had personal experience of dream-interpretation will know better than to do so.

On the other hand, I cannot accept the opinion, first expressed by H. Silberer, that every dream- or even that many dreams, and certain groups of dreams- calls for two different interpretations, between which there is even supposed to be a fixed relation. One of these, which Silberer calls the psycho- analytic interpretation, attributes to the dream any meaning you please, but in the main an infantile sexual one. The other, the more important interpretation, which he calls the anagogic interpretation, reveals the more serious and often profound thoughts which the dream-work has used as its material. Silberer does not prove this assertion by citing a number of dreams which he has analysed in these two directions. I am obliged to object to this opinion on the ground that it is contrary to facts. The majority of dreams require no over-interpretation, and are especially insusceptible of an anagogic interpretation. The influence of a tendency which seeks to veil the fundamental conditions of dream-formation and divert our interest from its instinctual roots is as evident in Silberer's theory as in other theoretical efforts of the last few years. In a number of cases I can confirm Silberer's assertions; but in these the analysis shows me that the dream-work was confronted with the task of transforming a series of highly abstract thoughts, incapable of direct representation, from waking life into a dream. The dream- work attempted to accomplish this task by seizing upon another thought-material which stood in loose and often allegorical relation to the abstract thoughts, and thereby diminished the difficulty of representing them. The abstract interpretation of a dream originating in this manner will be given by the dreamer immediately, but the correct interpretation of the substituted material can be obtained only by means of the familiar technique.