[7] This occasionally retrospective character of dreams has long been known, and was referred to by the writer of an article on 'Dreams and Dreaming' in the Lancet for 24th November 1877.
[8] The old French case (quoted by Macnish) of a woman, with a portion of her skull removed, whose brain bulged out during dreams but was motionless in dreamless sleep, as well as the more recent similar case known to Hammond (Treatise on Insanity, p. 233), supports the belief that the psychic activity which is not manifested in rememberable dreams is probably at the most of a very shadowy character. Even during waking life psychic activity often falls to a very low ebb; Beaunis, who has investigated this question ('Comment Fonctionne mon Cerveau,' Revue Philosophique, January 1909), describes a condition which he names 'psychic twilight' and regards as frequently occurring.
[9] Lucretius long ago referred to the significance of this fact (lib. iv. vv. 988-994), and he stated that the hallucination persisted for a time even after the dog had awakened. I have never myself been able to see any trace of such hypnagogic hallucination or delusion in dogs who awake from dreams, though I have frequently looked for it; it always seems to me that the dog who seemingly awakes from a dream of hunting grasps the fireside facts of life around him immediately and easily.
[10] This classification of the sources of dreams has, however, been generally accepted for little more than a century. At an earlier period it was not usually believed to cover the whole ground. Thus Des Laurens (A. Laurentius) in the sixteenth century, in his treatise on the Disease of Melancholy (insanity), says that there are three kinds of dreams: (1) of Nature (i.e. due to external causes); (2) of the mind (i.e. based on memories); and, above both these classes, (3) dreams from God and the devil.
[11] M. W. Calkins, 'Statistics of Dreams,' American Journal of Psychology, April 1893.
[12] The simile of the kaleidoscope for the most elementary process of dreaming has often suggested itself. Thus in an article on dreaming in the Lancet (24th November 1877) we read: 'The combinations are new, but the materials are old, some recent, many remote and forgotten.... The turn of the kaleidoscope is instantaneous and any new idea thrown into the field, perhaps in the act of turning, becomes an integral part of the picture.'
[13] Foucault, Le Rêve, p. 182.
[14] This is in accordance with the view of Wundt, who attributes this multiplication of imagery to the retinal element.
[15] Baron Charles Mourre, 'La Volonté dans le Rêve,' Revue Philosophique, May 1903.
[16] Ribot, Psychologie de l'Attention, 1889, chs. i. and ii.