field for
inquiry. The difficulty was, of course, to get the light of one star
separated from all the rest, because the light of one star is very faint
and feeble to cast a spectrum at all. Yet by infinite patience
difficulties were overcome. One star alone was allowed to throw its
light into the telescope; the light passed through a prism, and showed a
faint band of many colours, with the expected little black lines cutting
across it more or less thickly. Examinations have thus been made of
hundreds of stars. In the course of them some substances as yet unknown
to us on earth have been encountered, and in some stars one
element--hydrogen--is much stronger than in others; but, on the whole,
speaking broadly, it has been satisfactorily shown that the stars are
made on the same principles as our own sun, so that the reasoning of
astronomers which had argued them to be suns was proved.

[Illustration: THE SPECTRUM OF THE SUN AND SIRIUS.]

We have here in the picture the spectrum of the sun and the spectrum of
Arcturus. You can see that the lines which appear in the band of light
belonging to Sirius are also in the band of light belonging to the sun,
together with many others. This means that the substances flaming out
and sending us light from the far away star are also giving out light
from our own sun, and that the sun and Sirius both contain the same
elements in their compositions.

This, indeed, seems enough for the spectroscope to have accomplished; it
has interpreted for us the message light brings from the stars, so that
we know beyond all possibility of mistake that these glowing, twinkling
points of light are brilliant suns in a state of intense heat, and that
in them are burning elements with which we ourselves are quite familiar.
But when the spectroscope had done that, its work was not finished, for
it has not only told us what the stars are made of, but another thing
which we could never have known without it--namely, if they are moving
toward us or going away from us.

CHAPTER XIII

RESTLESS STARS

You remember we have already remarked upon the difficulty of telling how
far one star lies behind another, as we do not know their sizes. It is,
to take another similar case, easy enough to tell if a star moves to one
side or the other, but very difficult by ordinary observation to tell if
it is advancing toward us or running away from us, for the only means we
have of judging is if it gets larger o