man, or the orator, which is different from that which applies to the poor man, the bachelor, or, let us say, the merchant. The essential point, then, in this procedure is that the work of interpretation is not applied to the entirety of the dream, but to each portion of the dream-content severally, as though the dream were a conglomerate in which each fragment calls for special treatment. Incoherent and confused dreams are certainly those that have been responsible for the invention of the cipher method. *(3)

* Aristotle expressed himself in this connection by saying that the best interpreter of dreams is he who can best grasp similarities. For dream-pictures, like pictures in water, are disfigured by the motion (of the water), so that he hits the target best who is able to recognize the true picture in the distorted one (Buchsenschutz, p. 65).

*(2) Artemidoros of Daldis, born probably in the beginning of the second century of our calendar, has furnished us with the most complete and careful elaboration of dream-interpretation as it existed in the Graeco-Roman world. As Gompertz has emphasized, he ascribed great importance to the consideration that dreams ought to be interpreted on the basis of observation and experience, and he drew a definite line between his own art and other methods, which he considered fraudulent. The principle of his art of interpretation is, according to Gompertz, identical with that of magic: i.e., the principle of association. The thing dreamed meant what it recalled to the memory- to the memory, of course, of the dream-interpreter! This fact- that the dream may remind the interpreter of various things, and every interpreter of different things- leads, of course, to uncontrollable arbitrariness and uncertainty. The technique which I am about to describe differs from that of the ancients in one essential point, namely, in that it imposes upon the dreamer himself the work of interpretation. Instead of taking into account whatever may occur to the dream-interpreter, it considers only what occurs to the dreamer in connection with the dream-element concerned. According to the recent records of the missionary, Tfinkdjit (Anthropos, 1913), it would seem that the modern dream- interpreters of the Orient likewise attribute much importance to the co-operation of the dreamer. Of the dream-interpreters among the Mesopotamian Arabs this writer relates as follows: "Pour interpreter exactement un songe les oniromanciens les plus habiles s'informent de ceux qui les consultent de toutes les circonstances qu'ils regardent necessaires pour la bonne explication.... En un mot, nos oniromanciens ne laissent aucune circonstance leur echapper et ne donnent l'interpretation desiree avant d'avoir parfaitement saisi et recu toutes les interrogations desirables." [To interpret a dream exactly, the most practised interpreters of dreams learn from those who consult them all circumstances which they regard as necessary for a good explanation.... In a word, our interpreters allow no circumstance to be overlooked and do not give the desired interpretation before perfectly taking and apprehending all desirable questions.] Among these questions one always finds demands for precise information in respect to near relatives (parents, wife, children) as well as the following formula: habistine in hoc nocte copulam conjugalem ante vel post somnium [Did you this night have conjugal copulation before or after the dream?] "L'idee dominante dans l'interpretation des songes consiste a expliquer le reve par son oppose." [The dominant idea in the interpretation of dreams consists in explaining the dream by its opposite.]

*(3) Dr. Alfred Robitsek calls my attention to the fact that Oriental dream-books, of which ours are pitiful plagiarisms, commonly undertake the interpretation of dream-elements in accordance with the assonance and similarity of words. Since these relationships must be lost by translation into our language, the incomprehensibility of the equivalents in our popular "dream-books" is hereby explained. Information as to the extraordinary significance of puns and the play upon words in the old Oriental cultures may be found in the writings of Hugo Winckler. The finest example of a dream-interpretation which has come down to us from antiquity is based on a play upon words. Artemidoros relates the following (p. 225): "But it seems to me that Aristandros gave a most happy interpretation to Alexander of Macedon. When the latter held Tyros encompassed and in a state of siege, and was angry and depressed over the great waste of time, he dreamed that he saw a Satyr dancing on his shield. It happened that Aristandros was in the neighbourhood of Tyros, and in the escort of the king, who was waging war on the Syrians. By dividing the word Satyros into sa and turos, he induced the king to become more aggressive in the siege. And thus Alexander became master of the city." (Sa Turos = Thine is Tyros.) The dream, indeed, is so intimately connected with verbal expression that Ferenczi justly remarks that every tongue has its own dream- language. A dream is, as a rule, not to be translated into other languages.