finally travel to the ear, though
we are perfectly sure that these undulations have taken a circuitous
course. It is hardly necessary to remind the reader that the deeply
organized tendency to mistake the direction of the visible or audible
object in these cases has from remote ages been made use of as a means
of popular delusion. Thus, we are told by Sir D. Brewster, in his
entertaining _Letters on Natural Magic_ (letter iv.), that the concave
mirror was probably used as the instrument for bringing the gods before
the people. The throwing of the images formed by such mirrors upon smoke
or against fire, so as to make them more distinct, seems to have been a
favourite device in the ancient art of necromancy.

Closely connected with these illusions of direction with respect to
resting objects, are those into which we are apt to fall respecting the
movements of objects. What looks like the movement of something across
the field of vision is made known to us either by the feeling of the
ocular muscles, if the eye follows the object, or through the sequence
of locally distinct retinal impressions, if the eye is stationary. Now,
either of these effects may result, not only from the actual movement of
the object in a particular direction, but from our own movement in an
opposite direction; or, again, from our both moving in the first
direction, the object more rapidly than ourselves; or, finally, from our
both moving in an opposite direction to this, ourselves more rapidly
than the object. There is thus always a variety of conceivable
explanations, and the action of past experience and association shows
itself very plainly in the determination of the direction of
interpretation. Thus, it is our instinctive tendency to take apparent
movement for real movement, except when the fact of our own movement is
clearly present to consciousness, as when we are walking, or when we are
sitting behind a horse whose movement we see. And so when the sense of
our own movement becomes indistinct, as in a railway carriage, we
naturally drift into the illusion that objects, such as trees, telegraph
posts, and so on, are moving, when they are perfectly still. Under the
same circumstances, we are apt to suppose that a train which is just
shooting ahead of us is moving slowly.

Similar uncertainties arise with respect to the relative movement of two
objects, the eye being supposed to be fixed in space. When two objects
seem to pass one another, i