We have no other evidence of it." Well, that may very well be, but it does not seem to matter much, does not seem to bear on the topic in hand, if the dream is in any case only a reminiscence of a previous one. That hardly constitutes what we mean by a dream within a dream. "Putting this theory aside," my correspondent goes on, "I may say I have had these double dreams not infrequently, and usually related them at the breakfast table. Once, and only once, I had a triple dream, when I dreamed something, dreamed of relating it, and dreamed that I was dilating on my double dream. At least, such was my impression on awaking, but by the above theory I only dreamed that I had a double dream. This is very curious."

In discussing these rather intricate dealings of the mind it is very difficult, not so much to escape a tangle in ones own head, as to avoid creating a tangle in the readers. One may be very clear about the matter oneself, and yet express one's view in such a way that it is very foggy to other people. I am only too much afraid that this will often be the case in what I myself have written on the subject, for me to cavil at all at what seems to me (and probably by my own want of comprehension) the occasional fogginess of some of the correspondents who have been good enough to send me their experiences. Maury, who seems to have been the greatest experimentalist in dreams that has given his results to the public, experimented by giving a sleeper pin pricks to make him dream of spear thrusts - that is the type of his experiments - and sometimes the sleeper dreamed of spear thrusts "quite good," but sometimes of something quite different, and sometimes of nothing at all - that is, he could remember nothing when he awoke. So what Maury proved amounts to something like this negative conclusion, that the same sensational cause will not always produce the same dream effect.

And let it be said here, at once, for fear of misunderstanding, through inattention, that this is not the same as proving that the same dream is not always produced by the same cause. It does not directly touch that question.

The definition of dream that I ventured on in my tentative essay was a mental operation in sleep that is remembered on awaking. Objection to this may well be taken on the ground that mental operations of one person while asleep are often made obvious by speech or action to another, though the sleeper will perhaps have no knowledge of them when he awakes. Whether this is to be called a dream depends - on what? On the meaning we ascribe to the word. Now, if we take the view that mental operations go on all the while we sleep, but that in so - called dreamless sleep we forget them, then it is evident that we must limit "dreams" to the mental operations we remember. Otherwise it becomes a name of one constant aspect of sleep, which is not what we want But if we take the view that these mental operations go on intermittently, then "dreams" may have a meaning distinct from "sleep," as indicating the moments in which those operations do go on. Whether they go on all the while or intermittently there seems to be a division of opinion among the clever people.

Opinions seem so equally divided that they seem to cancel each other fairly - it is not the first time that the opinions of the learned have been so divided, nor the first time that it has been noticed (Dean Swift, in his pleasant way, noticed it) - so that we are left with a sheet tolerably blank to put our little opinions on: a very happy condition.

Sir Henry Holland, for instance, is of opinion that there is no such thing as dreamless sleep but that we dream, unconsciously, all the while that we are sleeping. Against his opinion we may put that of Dr. Arthur Durham, who not only thinks that there is such a thing as sleep without dreaming, but actually maintains that no sleep is sound in which a dream occurs. Sir Benjamin Brodie is of Sir Henry Holland's opinion, and, to make the balance level, we may cite Lord Brougham, whose view is that we dream only at the moment of transition from sleep to wakefulness. I cannot find that either Maury or Schemer express an opinion on the point, but instances may be cited on the one side or the other, to the end of the chapter. The case of Lord Holland, so often told, who went to sleep when some one was talking to him, woke up to hear the end of the sentence of which he had heard the beginning before he fell asleep, and in the moment* ary interval dreamed a dream whose incidents traversed much time, is often quoted in favour of Lord Brougham, as showing that it is just at the moment of waking that we dream.

What this case, which is typical of many instances, really does show appears to be merely this

- which is amply shown by many dreams - that in dreams the limits of time and space do not exist, that we can dream, to put it popularly, as fast as we can think, the incidents of the dream drama succeeding each other as quickly as thought can move, from one continent or from one century to another. This fact, which is established by all experience, is exactly what would be expected a priori, a comfortable state of things that does not always exist for the flattery of our poor reason.

It is a singular thing that these learned men should have come so terribly to loggerheads about the question. Those who oppose Lord Brougham and Dr. Durham cite the case of a man or a dog dreaming in sleep from which they do not wake immediately. They obviously "dream," in the popular sense of the word. The man talks in his sleep, the dog growls and trembles, and, as obviously, they do not dream at the moment only of transition from sleep to wakefulness. This seems to knock the bottom clean out of the argument of Lord Brougham and Co. And so it really would, if this twitching and growling of the dog, this talking of the man, were to be taken as signs of dreaming properly so - called We shall find, I think, that this is only popularly, and improperly, to be called dreaming, and also find that in this case, as in so many others since the days that Locke wrote his famous essay on "Words," it is because of the double entente, the double meaning, of the word they are discussing, rather than any real difference of opinion, that the learned have disagreed.