80 Recent Advances in Stellar Astronomy
the “apertures” is that between the centres of the outer
mirrors, and the power of the instrument is more than
doubled. Working with this, Michelson and Pease, in
December 1920, found that the fringes disappeared in the
light of Betelgeuse (Alpha Orionis) when the mirrors
were ten feet apart, while in the case of other stars the
fringes were clearly visible with a much wider separation—
proving that the cause of their disappearance was not any
defect of the instrument, or atmospheric disturbance, but
a measurable diameter of Betelgeuse. Calculation showed
the diameter to be 0".047—a small quantity according to
ordinary standards, but corresponding to enormous dimen-
sions in miles, for the parallax of this star (according to
the average of the results of several good observers) is
only 0".020. The diameter of the star is therefore 2.3
times the distance which separates the Earth and the Sun,
or 215 millions of miles. In the following summer Pease
extended this success by measuring the diameter of Arc-
turus, 0".022, and of Antares, 0".040. The parallaxes of
these stars, ".097 and ".009, are known with greater per-
centage accuracy than that of Betelgeuse, and the computed
diameters are correspondingly more trustworthy. That of
Arcturus comes out 21 million miles and of Antares 400
million—considerably larger than the orbit of Mars.
A few other stars, notably Aldebaran, probably have
angular diameters large enough to measure in the same
fashion. To measure smaller stars, it might appear nec-
cessary to build a still larger interferometer. But this has
been avoided by Michelson’s ingenuity. By a very simple
device, he is able to produce, in the same field of view as
the fringes already described, another set, formed by light
from the same star and adjustable to any desired degree
of visibility. In this way the visibility of the main fringes