enter both systems in our diagram, expressing, by the names given them, their relation to consciousness. (See illustration.)

The last of the systems at the motor end we call the preconscious (Pcs.) to denote that the exciting processes in this system can reach consciousness without any further detention, provided certain other conditions are fulfilled, e.g., the attainment of a definite degree of intensity, a certain apportionment of that function which we must call attention, etc. This is at the same time the system which holds the keys of voluntary motility. The system behind it we call the unconscious (Ucs), because it has no access to consciousness except through the preconscious, in the passage through which the excitation-process must submit to certain changes. *

* The further elaboration of this linear diagram will have to reckon with the assumption that the system following the Pcs represents the one to which we must attribute consciousness (Cs), so that P = Cs.

In which of these systems, then, do we localize the impetus to dream-formation? For the sake of simplicity, let us say in the system Ucs. We shall find, it is true, in subsequent discussions, that this is not altogether correct; that dream-formation is obliged to make connection with dream-thoughts which belong to the system of the preconscious. But we shall learn elsewhere, when we come to deal with the dream-wish, that the motive-power of the dream is furnished by the Ucs, and on account of this factor we shall assume the unconscious system as the starting- point for dream-formation. This dream-excitation, like all the other thought-structures, will now strive to continue itself in the Pcs, and thence to gain admission to the consciousness.

Experience teaches us that the path leading through the preconscious to consciousness is closed to the dream-thoughts during the day by the resisting censorship. At night they gain admission to consciousness; the question arises: In what way and because of what changes? If this admission were rendered possible to the dream-thoughts by the weakening, during the night, of the resistance watching on the boundary between the unconscious and the preconscious, we should then have dreams in the material of our ideas, which would not display the hallucinatory character that interests us at present.

The weakening of the censorship between the two systems, Ucs and Pcs, can explain to us only such dreams as the Autodidasker dream but not dreams like that of the burning child, which- as will be remembered- we stated as a problem at the outset in our present investigations.

What takes place in the hallucinatory dream we can describe in no other way than by saying that the excitation follows a retrogressive course. It communicates itself not to the motor end of the apparatus, but to the sensory end, and finally reaches the system of perception. If we call the direction which the psychic process follows from the unconscious into the waking state progressive, we may then speak of the dream as having a regressive character. *

* The first indication of the element of regression is already encountered in the writings of Albertus Magnus. According to him the imaginatio constructs the dream out of the tangible objects which it has retained. The process is the converse of that operating in the waking state. Hobbes states (Leviathan, ch. 2): "In sum our dreams are the reverse of our imagination, the motion, when we are awake, beginning at one end, and when we dream at another" (quoted by Havelock Ellis, loc. cit., p. 112). -

This regression is therefore assuredly one of the most important psychological peculiarities of the dream- process; but we must not forget that it is not characteristic of the dream alone. Intentional recollection and other component processes of our normal thinking likewise necessitate a retrogression in the psychic apparatus from some complex act of ideation to the raw material of the memory-traces which underlie it. But during the waking state this turning backwards does not reach beyond the memory-images; it is incapable of producing the hallucinatory revival of the perceptual images. Why is it otherwise in dreams? When we spoke of the condensation-work of the dream we could not avoid the assumption that by the dream-work the intensities adhering to the ideas are completely transferred from one to another. It is