The obscuration during sleep of the external sensory channels, with the checks on false conclusions they furnish, is not alone sufficient to explain the symbolism of dreams. The dissociation of thought during sleep, with the diminished attention and apperception involved, is also a factor. The magnification of special isolated sensory impressions in dreaming consciousness is associated with a general bluntness, even an absolute quiescence, of the external sensory mechanism. One part of the organism, and it seems usually a visceral part, is thus apt to magnify its place in consciousness at the expense of the rest. As Vaschide and Piéron say, during sleep 'the internal sensations develop at the expense of the peripheral sensations.' That indeed seems to be the secret of the immense emotional turmoil of our dreams. Yet it is very rare for these internal sensations to reach the sleeping brain as what they are. They become conscious, not as literal messages, but as symbolical transformations. The excited or labouring heart recalls to the brain no memory of itself, but some symbolical image of excitement or labour. There is association, indeed, but it is association not along the matter-of-fact lines of our ordinary waking civilised life, but along much more fundamental and primitive channels, which in waking life we have now abandoned or never knew. There is another consideration which may be put forward to account for one group of dream-symbolisms. It has been found that certain hysterical subjects of old standing when in the hypnotic state are able to receive mental pictures of their own viscera, even though they may be quite ignorant of any knowledge of the shape of these viscera. This _autoscopy_, as it has been called, has been specially studied by Féré, Comar, and Sollier.[149] Hysteria is a condition which is in many respects closely allied to sleep, and if it is to be accepted as a real fact that autoscopy occasionally occurs in the abnormal psychic state of hypnotic sleep in hysterical persons, it is possible to ask whether it may not sometimes occur normally in the allied state of sleep. In the hypnotic state it is known that parts of the organism normally involuntary may become subject to the will; it is not incredible that similarly parts normally insensitive may become sufficiently sensitive to reveal their own shape or condition. We may thus, indeed, the more easily understand those premonitory dreams in which the dreamer becomes conscious of morbid conditions which are not perceptible to waking consciousness until they have attained a greater degree of intensity.[150]