stigate this in a simple way, the following experiment
was planned and carried out by Mr. B.

He wished to find out whether, if two detached surfaces of different
sizes be gazed at together, the linear distances of the field of
vision (the whole scene visible at once) would be at all misjudged. To
test this, he put in the window (W)[5] of the dark room a filling of
white cardboard in which two square holes had been cut (S S'). The
sides of the squares were of certain very unequal lengths. Then a slit
was made between the middle points of the sides of the squares next to
each other, so that there was a narrow path or trough joining the
squares between their adjacent sides. Inside the dark room he arranged
a bright light so that it would illuminate this trough, but not be
seen by a person seated some distance in front of the window in the
next room. A needle (D) was hung on a pivot behind the cardboard, so
that its point could move along the bright trough in either direction;
and on the needle was put the armature (A) of an electro-magnet which,
when a current passed, would be drawn instantly to the magnet (E), and
so stop the needle exactly at the point which it had then reached. A
clock motor (Cm) was arranged in such a way as to carry the needle
back and forth regularly over the slit; and the electro-magnet was
connected by wires with a punch key (K) on a table beside the subject
in the next room. All being now ready, the subject, Mr. S., is told to
watch the needle which appears as a bead of light travelling along the
slit, and stop it when it comes to the middle point of the line, by
pressing the electric key. The experimenter, who stands behind the
window in the dark room, reads on a scale (mm.) marked in millimetres
the exact point at which the needle stops, releases the needle by
breaking the current, thus allowing it to return slowly over the line
again. This gives the subject another opportunity to stop it at what
he judges to be the exact middle of the line, and so on. The
accompanying figure (Fig. 7) shows the entire arrangement.

[Footnote 5: This and the following letters in parentheses refer to
Fig. 7]

[Illustration: FIG. 7]

A great many experiments performed in this way, with the squares set
both vertically and horizontally, and with several persons, brought a
striking and very uniform result. The point selected by the subject as
the middle is regularly too far toward the smaller square. Not a
littl