structural arrangement, your inevitable answer will be, that those
crystals are built by the play of polar forces with which their
molecules are endowed. In virtue of these forces, molecule lays
itself to molecule in a perfectly definite way, the final visible form
of the crystal depending upon this play of its ultimate particles.

Everywhere in Nature we observe this tendency to run into definite
forms, and nothing is easier than to give scope to this tendency by
artificial arrangements. Dissolve nitre in water, and allow the water
slowly to evaporate; the nitre remains and the solution soon becomes
so concentrated that the liquid condition can no longer be preserved.
The nitre-molecules approach each other, and come at length within the
range of their polar forces. They arrange themselves in obedience to
these forces, a minute crystal of nitre being at first produced. On
this crystal the molecules continue to deposit themselves from the
surrounding liquid. The crystal grows, and finally we have large
prisms of nitre, each of a perfectly definite shape. Alum crystallizes
with the utmost ease in this fashion. The resultant crystal is,
however, different in shape from that of nitre, because the poles of
the molecules are differently disposed. When they are _nursed_ with
proper care, crystals of these substances may be caused to grow to a
great size.

The condition of perfect crystallization is, that the crystallizing
force shall act with deliberation. There should be no hurry in its
operations; but every molecule ought to be permitted, without
disturbance from its neighbours, to exercise its own rights. If the
crystallization be too sudden, the regularity disappears. Water may be
saturated with sulphate of soda, dissolved when the water is hot, and
afterwards permitted to cool. When cold the solution is