More often probably it comes from something in our own body, some worry of the digestive business, some muscle growing cramped after lying long in one position, too much warmth, or too much cold, and so on. But the other cases, in which a sleeper's dream has been occasioned, or altered, by a light brought into the room, a noise in the room, and so on, are too many and well established for us to doubt that they occur, even if our own experience did not confirm them.

The Marquis d'Hervey, who made a profound study of the whole question of sleep and dreams, took the view that sleep meant merely the loss of all sense (I am purposely stating his case very crudely), the brain remaining in its normal activity. It is a view, however, which Maury combats with perfect success, as the general verdict allows, pointing out on the one hand that the avenues to the brain by way of the senses are by no means altogether blocked in sleep

! and on the other hand that the activity of the brain in sleep is by no means of its normal kind.

It may appear very singular, and indeed it is most curious, that a similar state of sleep (and of repair of waste) is induced in the members of the body by an influx of blood to the vessels below the normal amount, as in the brain by an excessive influx of the blood (for it is correct enough to say that all parts of the body sleep, in the chemical sense of the word), but Maury points out that there is a congestion hypertmique and also a congestion anemique, as he calls them. Again, we may make this seem less strange by remembering the analogy of the two gouts, of the rich man and of the poor man, as we call them.

It is the view of the Marquis d'Hervey, in some degree following as a natural sequence on his theory that sleep means little else than the abeyance of sensation, that dreams become more distinct in proportion as the sleep becomes deeper. The common phrase of "deep dreamless sleep" he would therefore take to be a contradiction in terms. It is not very easy in the first place to fix an exact meaning to the term deep -in French "profond" - as applied to sleep. In common talk we use it, I think, in two senses. We use it in the first place of a sleep that might be better called "quiet," when the sleeper shows no sign of any mental agitation, no tossing, no mumbling of words; and in another sense we use it of a sleep from which there is difficulty in awaking the sleeper. The two senses are different, for sometimes it is easy to wake a sleeper out of a very quiet sleep and difficult, comparatively, to wake him out of a sleep that gives indications of being troubled. So we cannot dogmatise about this.

Maury, in opposition to d'Hervey, contends that a quiet sleep is dreamless, but it is hard to see that he proves it, any more than d'Hervey strictly speaking proves his contention on the opposite side; but the fact that whenever we wake a man suddenly (unless he be one of the very exceptional people who never retain, when they awake, any consciousness of any mental operation during sleep) he always awakes to find that he was dreaming, is strong evidence, surely, against there being such a thing as dreamless sleep at all. It is not a little remarkable too, and it is evidence that tells much in favour of the Marquis d'Hervey's view, that the dreams of which we give the clearest indications to those who are watching our sleep are not as a rule those which remain most clearly in our waking memory. The somnambulist, I believe, almost invariably forgets not only all that he did in his sleep, but almost all the train of thought that led up to his somnambulistic actions. If you wake a tossing, or a talking sleeper he often will be found to retain but a very vague memory of the cause of his troubled movements and mumblings. Far more likely are you, as the Marquis d'Hervey says, and as all our evidence shows, to get a lucid account of a dream from a man whose sleep has apparently been quite unvexed.

It is a commonplace observation that all memory of a dream vanishes with wonderful quickness, unless its incidents be jotted down, run over in the mind, or told to another, immediately on awaking. But in this there is no mystery. The fact that the faculty of attention is practically in abeyance fully accounts for it, and perhaps the wonder is rather that any memory of mental operations in sleep should remain at all in the waking mind than that it should quickly vanish.

Besides these originating causes of dreams that we have mentioned, Maury has a whole class of creations, phantoms, or whatever we are to call them, to which he ascribes a very preponderating influence in the production of dreams. As described by him, there is not the very slightest doubt that they must have such a preponderating influence. Yet he surely makes a mistake in thinking that as he describes them they are factors in the dream experience of ordinary men and women. He devotes a whole chapter to these phantoms, under the heading of "Hallucinations Hypnagogiques ". The mind of the dozing man, he says (he speaks of his own mind in particular, and clearly he deems his experience to be part of the common lot of man), is beset with a whole host of hallucinations of appearances of people and places. In the sense that the mind, between waking and sleep, goes on with its association of ideas, modified and perhaps altered entirely by some sound dimly heard as sleep draws on, this is of course true, and part of common experience, but until the attention so far falls into abeyance that we may be said to be actually asleep (for the boundary line between sleeping and waking is not fixed hard and fast) we do not, I think - that is to say the majority of us do not - ever find ourselves taking part in a conversation, say, with people we know to be dead, nor commit any of the outrages on common-sense (consequent on mistaking the images that our ideas suggest for objective realities) such as we commit in sleep and such as Maury would have us believe that we commit in the society of those "Hallucinations Hypnagogiques". A considerable number of persons, notwithstanding, seem to have been conscious of these hallucinations, these creations of what they do not wish us to call a "dream world," but may permit us to call a "doze world," and curiously enough some of the people who seem to have been most conscious of them have been some of the learned people who have written on the subject.