any telescope in existence, can
directly reveal the duplicity of beta Aurigae to the eye--i. e., we can
not see the two stars composing it, because they are so close that their
light remains inextricably mingled after the highest practicable
magnifying power has been applied in the effort to separate them. But
the spectroscope shows that the star is double and that its components
are in rapid revolution around one another, completing their orbital
swing in the astonishingly short period of _four days_! The combined
mass of the two stars is estimated to be two and a half times the mass
of the sun, and the distance between them, from center to center, is
about eight million miles.

The manner in which the spectroscope revealed the existence of two
stars in beta Aurigae is a beautiful illustration of the unexpected and,
so to speak, automatic application of an old principle in the discovery
of new facts not looked for. It was noticed at the Harvard Observatory
that the lines in the photographed spectrum of beta Aurigae (and of a few
other stars to be mentioned later) appeared single in some of the
photographs and double in others. Investigation proved that the lines
were doubled at regular intervals of about two days, and that they
appeared single in the interim. The explanation was not far to seek. It
is known that all stars which are approaching us have their spectral
lines shifted, by virtue of their motion of approach, toward the violet
end of the spectrum, and that, for a similar reason, all stars which are
receding have their lines shifted toward the red end of the spectrum.
Now, suppose two stars to be revolving around one another in a plane
horizontal, or nearly so, to the line of sight. When they are at their
greatest angular distance apart as seen from the earth one of them will
evidently be approaching at the same moment that the other is receding.
The spectral lines of the first will therefore be shifted toward the
violet, and those of the second will be shifted toward the red. Then if
the stars, when at their greatest distance apart, are still so close
that the telescope can not separate them, their light will be combined
in the spectrum; but the spectral lines, being simultaneously shifted in
opposite directions, will necessarily appear to be doubled. As the
revolution of the stars continues, however, it is clear that their
motion will soon cease to be performed in the line of sight, and will
become more and