(l) In ordinary circumstances we are made aware of the existence of such things as chairs and tables through the sense of sight; but chairs and tables may, in certain conditions, be accepted by the mind as existing where they have no existence. They do not, in that case, reach the mind through the eye. They may be said to be seen directly by the mind itself without being imaged on the retina. The eye plays no part in this kind of vision. The mind, in other words, may be said, in certain conditions, to see without the aid of the eye. But the things so seen have no reality.

They are hallucinations. I give them that name because I can find no better. My difficulty, however, is scarcely with the name itself. It is rather with the fact that it habitually occurs in an environment which differs from that in which I am now to use it. As generally employed it refers to a manifestation of disease - often toxic in its nature. But the phenomena, which I here call hallucinations, are physiological. Disease plays no part in them. They have been called - 'sane hallucinations.'

It is beyond question, I think, that all the five senses - sight, hearing, touch, taste, and smell - may appear to be concerned in the hallucinations to which I now refer; but it is desirable at once to point out that those things, which present themselves as socalled sane hallucinations, need not have reached the mind, at any previous time, through some one of the organs of sense.

Sounds, tastes, and odours, which have no existence, are all thus imagined, but perhaps the most vivid fancies are those which seem to be furnished by the eye. All seeing need not actually he through the eye. There is what is called a non-optical or mind's eye vision, which does not reach the brain from an excited retina, but may be regarded as generated in the brain. Memory-images may so arise; that is, objects, which have at some time been seen through the eye, may be recalled by the mind, and may be then erroneously accepted as actually existing and present, when they do not exist. In like manner, creation-images may arise; that is, objects which have never been seen through the eye may appear to be seen as actually present, when no such objects exist. Both memory - and creation-images appear as these so-called sane hallucinations.

It is not my business here to deal with their appearance in waking life when there is no suspicion of disease, and when they may seem to be called up by the will and sometimes made to serve a useful purpose, as when an architect sees the building he has designed. Still less is it my present business to discuss their appearance as the outcome of disease, when, in certain conditions, the eye may even seem to follow changes in their position, as if they were being at the moment pictured on the retina and sent from it to the brain and mind. At present I am concerned only with their appearance in ordinary dreams.

(2) It is safe to say that no person ever gets through twenty-four hours of life without having hallucinations of this kind. They go, indeed, to make up our dreams. We see, and hear, and feel, and taste, and smell things in our dreams which have no reality, but which appear to the mind of the dreamer to be real.

Perhaps the hallucinations of delirium, whether toxic or non-toxic, which are commonly regarded as occurring in persons who are awake but in some state of disease, may bo more correctly regarded as occurring in persons who are only awake in a limited sense-who may, at least, be asleep to the extent and in the sense of being unable to control and direct currents of thought.

The hallucinations which are pathological, and also those occurring in dreams which are not pathological, may lead to various kinds of bodily action, and in this way they may be said to influence conduct.

(3) It seems to me that there is no such thing as dreamless sleep. During the whole continuance of sleep, the mind, I believe, is occupied with a certain kind of thinking which works round what I have called hallucinations. I do not expect to be able to prove the correctness of this opinion as to the persistence of dreams all through sleep, but I think that it can easily he shown to be possibly correct. I go further, and say that many things show that it is probably correct. 1 may not be able to prove absolutely its correctness, hut it is proper to bear in mind that it is quite as difficult to prove absolutely that it is not correct. My difficulty is frankly avowed. Many things, however, are taught in Biology as being certainly true, in regard to which a like avowal could be made but is not made. There is what has been called a 'conjectural biology.'

We do not and we cannot remember much of what we have been thinking about, while we are awake. This is unquestionably true in a very large sense. But nevertheless we do not doubt that we have been thinking continuously. We do not suppose that at any time all thinking had ceased, though we may be completely unable to recall what it was about.

(4) It is not a new thing to hold that there is no sleep without dreaming-in other words, that dreaming goes on unceasingly all through sleep. I have stated my own opinion strongly, but the same opinion has been nearly as strongly expressed by others. Sir Benjamin Brodie, for instance, may be said to express it when he writes: 'I believe that I seldom, if ever, sleep without dreaming.' Sir Henry Holland expresses it still more plainly when he says: 'No moment of sleep is without some condition of dreaming.' Godwin says much the same thing when he asserts that 'sleep is not a suspension of thought' -in other words, that dreaming is sleep-thinking. Dr. John Reid still more clearly holds the opinion, though he does not furnish me with a short apt quotation. Hazlitt, too, may be taken as holding that there is no such thing as dreamless sleep.

Descartes and his followers may, perhaps, be regarded as holding that the mind is unceasingly at work in sleep - even in the 'profoundest sleep' though 'the memory retains it not,' and Isaac Watts says that 'the soul never intermits its activity,' and that we may 'know of sleeping thoughts at the moment they arise, and not retain them the next moment.'