do not correspond to the natural bilateral type, would fail to
give aesthetic pleasure. Such, however, is not the case. Non-symmetrical arrangements of space are often extremely pleasing.

<1> The following is adapted from the author's _Studies in
Symmetry, Harvard Psych. Studies_, vol i, 1902.

This contradiction disappears if we are able to show that the
apparently non-symmetrical arrangement contains a hidden
symmetry, and that all the elements of that arrangement
contribute to bring about just that bilateral type of motor
impulses which is characteristic of geometrical symmetry.

A series of experiments was arranged, in which one of two
unequal lines of white on a black background being fixed in
an upright position a certain distance from the centre, the
other was shifted until the arrangement was felt to be pleasing.
It was found that when two lines of different sizes were opposed,
their relative positions corresponded to the relation of the
arms of a balance, that is, a small line far from the centre
was opposed by a large one near the centre. A line pointing
out from the centre fitted this formula if taken as "heavy,"
and pointing in, if taken as "light." Similarly, objects of
intrinsic interest and objects suggesting depth in the third
dimension were "heavy" in the same interpretation. All this,
however, did not go beyond the proof that all pleasing space-
arrangements can be described in terms of mechanical balance.
But what was this mechanical balance? A metaphor explains
nothing, and no one will maintain that the visual representation
of a long line weights more than a short one. Moreover, the
elements in the balance were so far heterogeneous. The