ed, the
other is green; when the one is yellow, the other is blue. Here we
have it in our power to demonstrate afresh a statement made in our
first lecture, that although the mixture of blue and yellow pigments
produces green, the mixture of blue and yellow lights produces white.
By enlarging our aperture, the two images produced by the spar are
caused to approach each other, and finally to overlap. The one image
is now a vivid yellow, the other a vivid blue, and you notice that
where these colours are superposed we have a pure white. (See fig. 43,
where N is the end of the polarizer, Q the quartz plate, L a lens, and
B the bi-refracting spar. The two images overlap at O, and produce
white by their mixture.)

[Illustration: Fig. 43.]

Sec. 8. _The Magnetization of Light._

This brings us to a point of our inquiries which, though rarely
illustrated in lectures, is nevertheless so likely to affect
profoundly the future course of scientific thought that I am unwilling
to pass it over without reference. I refer to the experiment which
Faraday, its discoverer, called the 'magnetization of light.' The
arrangement for this celebrated experiment is now before you. We have,
first, our electric lamp, then a Nicol prism, to polarize the beam
emergent from the lamp; then an electro-magnet, then a second Nicol,
and finally our screen. At the present moment the prisms are crossed,
and the screen is dark. I place from pole to pole of the
electro-magnet a cylinder of a peculiar kind of glass, first made by
Faraday, and called Faraday's heavy glass. Through this glass the beam
from the polarizer now passes, being intercepted by the Nicol in
front. On exciting the magnet light instantly appears upon the screen.
By the action of the magnet upon the heavy glass the plane of
vibration is caused to rotate, the light being thus enabled to get
through the analyzer.

The two classes into which quartz-crystals are divided have been
already mentioned. In my hand I hold a compound plate, one half of it
taken from a right-handed, and the other from a left-handed crystal.
Placing the plate in front of the polarizer, I turn one of the Nicols
until the two halves of the plate show a common puce colour. This
yields an exceedingly sensitive means of rendering visible the action
of a magnet upon light. By turning either the polarizer or the
analyzer through the smallest angle, the uniformity of the colour
disappears, and the two halves of the qua