which _b c_ is the
section shall be perpendicular, or nearly so, to the two end surfaces
of the rhomb _b x, c y_.

[Illustration: Fig. 34.]

The invention of the Nicol prism was a great step in practical optics,
and quite recently such prisms have been constructed of a size and
purity which enable audiences like the present to witness the
chromatic phenomena of polarized light to a degree altogether
unattainable a short time ago.

(The two prisms employed in these experiments were lent to me by my
lamented friend Mr. William Spottiswoode, and they were manufactured
by Mr. Ahrens, an optician of consummate skill.)

Sec. 2. _Colours of Films of Selenite in Polarized Light_.

Two Nicol prisms play the same part as the two plates of tourmaline.
Placed with their directions of vibration parallel, the light passes
through both; while when these directions are crossed the light is
quenched. Introducing a film of mica between the prisms, the light, as
in the case of the tourmaline, is restored. But notice, when the film
of mica is _thin_ you have sometimes not only light, but _coloured_
light. Our work for some time to come will consist of the examination
of such colours. With this view, I will take a representative crystal,
one easily dealt with, because it cleaves with great facility--the
crystal gypsum, or selenite, which is crystallized sulphate of lime.
Between the crossed Nicols I place a thick plate of this crystal; like
the mica, it restores the light, but it produces no colour. With my
penknife I take a thin splinter from the crystal and place it between
the prisms; the image of the splinter glows with the richest colours.
Turning the prism in front, these colours gradually fade and
disappear, but, by continuing the rotation until the vibrating
sections of the prisms are parallel to each other, vivid colours again
arise, but these colours are complementary to the former ones.

Some patches of the splinter appear of one colour, some of another.
These differences are due to the different thicknesses of the film. As
in the case of Hooke's thin plates, if the thickness be uniform the
colour is uniform. Here, for instance, is a stellar shape, every
lozenge of the star being a film of gypsum of uniform thickness: each
lozenge, you observe, shows a brilliant and uniform colour. It is
easy, by shaping our films so as to represent flowers or other
objects, to exhibit such objects in hues unattainable by art. Here,
fo