special study of sense-illusions. By these the term illusion is commonly
employed in a narrow, technical sense, and opposed to hallucination. An
illusion, it is said, must always have its starting-point in some actual
impression, whereas a hallucination has no such basis. Thus it is an
illusion when a man, under the action of terror, takes a stump of a
tree, whitened by the moon's rays, for a ghost. It is a hallucination
when an imaginative person so vividly pictures to himself the form of
some absent friend that, for the moment, he fancies himself actually
beholding him. Illusion is thus a partial displacement of external fact
by a fiction of the imagination, while hallucination is a total
displacement.

This distinction, which has been adopted by the majority of recent
alienists[1], is a valuable one, and must not be lost sight of here. It
would seem, from a psychological point of view, to be an important
circumstance in the genesis of a false perception whether the
intellectual process sets out from within or from without. And it will
be found, moreover, that this distinction may be applied to all the
varieties of error which I propose to consider. Thus, for example, it
will be seen further on that a false recollection may set out either
from the idea of some actual past occurrence or from a present product
of the imagination.

It is to be observed, however, that the line of separation between
illusion and hallucination, as thus defined, is a very narrow one. In by
far the largest number of hallucinations it is impossible to prove that
there is no modicum of external agency co-operating in the production of
the effect. It is presumable, indeed, that many, if not all,
hallucinations have such a basis of fact. Thus, the madman who projects
his internal thoughts outwards in the shape of external voices may, for