This has nothing to do with the fact that hypnosis is best brought about by suggesting the idea of sleep, that is, the belief that sleep will set in. This belief is indeed effective in removing all the ideas which are awake in the mind which would interfere with the willingness to submit to the suggestions of the hypnotizer. But the fact that belief in sleep and expectation of sleep bring with them the hypnotic state is not a proof that the hypnotic state itself is sleep. Even the mental experiences which can remain in sleep, the dreams, are characteristically different from the hypnotic experience. Thus the dreams show that unselective awakening of ideas which is to be expected from a general decrease of functioning. The hypnotic variation is characterized just by its selective narrowing of consciousness. For the same reason, hypnotism is strikingly different from such diseases of the mind as dementia. Certainly in dementia too, many associations are cut off, but it is not a selective inhibition, it is a haphazard destruction resulting from the degeneration in the brain. The fundamental principle of the hypnotic state lies in its selective character. Inhibited and cut off are those states which are antagonistic to the beliefs in the suggested ideas, and as their antagonism consists in their connection with opposite actions, the whole is again a question of motor setting. No doubt, such new motor setting can precede the normal sleep too; thus the sleeper may be insensitive to any surrounding noises, but perhaps awake at the slightest call from a patient who is intrusted to his care. In that case, one special feature of hypnotism is superadded to sleep but the sleep itself is not hypnotic. Again sleep may go over into a state which shares many characteristic features with hypnotism, that is, somnambulism, and it may be said with a certain truth that hypnotism is artificial somnambulism. But somnambulism, while arising in sleep, is not at all a feature of sleep.