suffering, and shortly afterwards he wrote in a letter that 'in feverish dreams a terrible doubt has risen up before me; I have fancied I heard her voice calling to me from the vault where she was laid.' He comforted himself, however, with the thought that this horrible supposition was unjustified, since French doctors had been present at her death. Maury[184] also mentions that he had often had dreams of this type in which the dead appeared as living, though the sight of them always produced astonishment and doubt which the sleeping brain endeavoured to allay by some kind of explanation. Beaunis also describes how he has dreamed with surprise of meeting a friend whom even in his dream he knew to be dead.[185] It is not difficult, in the light of all that we have been able to learn regarding the psychology of the world of dreams, to account for the process here described, for its frequency, and for its poignant emotional effects. This dream type is only a special variety of the commonest species of dream, in which two or more groups of reminiscences flow together and form a single bizarre congruity, a _confusion_ in the strict sense of the word. The death of a friend sets up a barrier which cuts into two the stream of impressions concerning that friend. Thus, two streams of images flow into sleeping consciousness, one representing the friend as alive, the other as dead. The first stream comes from older and richer sources; the second is more poignant, but also more recent and more easily exhausted. The two streams break against each other in restless conflict, both, from the inevitable conditions of dream life, being accepted as true, and they eventually mix to form an absurd harmony, in which the older and stronger images (in accordance with that recognised tendency for old psychic impressions generally to be most stable) predominate over those that are more recent. Thus, in the first observation the dreamer seems to have begun his dream by imagining that his mother was alive as of old; then his more recent experiences interfered with the assertion of her death. This resulted in a struggle between the old-established images representing her as alive and the later ones representing her as dead. The idea that she had come to life again was evidently a theory that had arisen in his brain to harmonise these two opposing currents. The theory was not accepted easily; all sorts of scientific objections arose to oppose it, but there could be no doubt, for his mother was there. The dreamer is in the same position as a paranoiac who constantly seems to hear threatening voices; henceforth he is absorbed in inventing a theory (electricity, hypnotism, or whatever it may be) to account for his hallucinations, and his whole view of life is modified accordingly. The dreamer, in the cases I am here concerned with, sees an image of the dead person as alive, and is therefore compelled to invent a theory to