44 Recent Advances in Stellar Astronomy
but only about one star in a thousand is redder than the
limit, and it is doubtful if any known star is bluer.1
Terrestrial sources of light, on the same scale, are found
to be very red. All but a few of the whitest of them, such
as the carbon arc or the gas-filled tungsten lamp, are red-
der than the reddest of the ordinary types of stars, and
even the exceptions are much redder than the Sun. Our
instinctive standards for estimating color, in fact, are quite
different for the stars and for artificial light. A common-
place, though modern, illustration of this is the “daylight”
incandescent lamp, which is carefully adjusted to give light
of the color of sunlight, but, if seen among other lamps at
night, looks blue.
So far, however, we have been talking only about the
apparent brightness of the stars—the light which we get
from them. What of their real brightness—the light which
they send out? To determine this we must know the dis-
tances of the stars—and this opens up a long and difficult
chapter of practical astronomy.
The first rational estimate of the distance of a star was
reached from photometric considerations, and by no less
distinguished a philosopher than Newton. He proved,
first, that the stars do not shine by reflected sunlight—for
they show no perceptible disks in the telescope and there-
fore they could not reflect light enough to account for their
observed brightness. They must, therefore, be self-
luminous like the Sun. Now Sirius looks rather fainter
than Jupiter. From this fact, by ingenious but thoroughly
sound reasoning, Newton showed that, if Sirius was really
as bright as the Sun, its distance must be fully 100,000
times as great as that which separates the Earth and Sun.
We know now that the actual distance is five times greater;
1 Except the nuclei of certain nebulæ, which may not be ordinary atari.