ar. First of all, the mind during sleep wants what M.
Taine calls the corrective of a present sensation. When awake under
ordinary circumstances, any momentary illusion is at once set right by a
new act of orientation. The superior vividness of the external
impression cannot leave us in any doubt, when calm and self-possessed,
whether our mental images answer to present realities or not. On the
other hand, when asleep, this reference to a fixed objective standard is
clearly impossible. Secondly, we may fairly argue that the mental images
of sleep approximate in character to external impressions. This they do
to some extent in point of intensity, for, in spite of the diminished
excitability of the centres, the mode of stimulation which occurs in
sleep may, as I have hinted, involve an energetic cerebral action. And,
however this be, it is plain that the image will gain a preternatural
force through the greatly narrowed range of attention. When the mind of
the sleeper is wholly possessed by an image or group of images, and the
attention kept tied down to these, there is a maximum reinforcement of
the images. But this is not all. When the attention is thus held captive
by the image, it approximates in character to an external impression in
another way. In our waking state, when our powers of volition are
intact, the external impression is characterized by its fixity or its
obdurate resistance to our wishes. On the other hand, the mental image
is fluent, accommodating, and disappears and reappears according to the
direction of our volitions. In sleep, through the suspension of the
higher voluntary power of attention, the mental image seems to lord it
over our minds just as the actual impression of waking life.

This much may suffice, perhaps, by way of a general description of the
sleeping and dreaming state. Other points will make themselves known
after we have studied the contents and structure of dreams in detail.

Dreams are commonly classified (_e.g._ by Wundt) with hallucinations,
and this rightly, since, as their common appellation of "vision"
suggests, they are for the most part the semblance of percepts in the
absence of external impressions. At the same time, recent research goes
to show that in many dreams something answering to the "external
impression" in waking perception is the starting-point. Consequently, in
order to be as accurate as possible, I shall divide dreams into
illusions (in the narrow sense) and