The name is absent

The Sizes and Masses of the Stars 71
tungsten lamps, which work at a higher temperature,
give us several times as much light for our money as the
old carbon filaments. The same principle, applied to the
stars, confirms the low temperatures of Classes K and
M. But the white stars present an apparent difficulty.
As a black body is raised to very high temperatures—
well above that of the Sun—most of the energy which it
radiates passes into the ultra-violet, where it does not
affect our eyes. Hence, theoretically, the luminous effici-
ency should be a maximum for a temperature not far
from the Sun’s, and fall off again for the hotter stars.
The solution is found in the fact that the observations
were made with light which had come through the
earth’s atmosphere, and been reflected from a silvered
mirror. Our atmosphere is entirely opaque to all ultra-
violet light of wave length less than 3000 Angstrom
units (H that of the visible violet) and silver, though
an admirable reflector for all visible rays, hardly reflects
at all in the ultra-violet below 3000 and 3300. Hence
practically no radiation of shorter wage length than
this and very little beyond 3500, reaches the thermopile,
and the powerful ultra-violet radiation of the white stars
fails to record itself. When allowance is made for this,
and also for the absorption of the long infra-red wave
in the atmosphere, we get the ratio between the amount
of the light of a distant hot body and that portion of
its heat which could have reached the measuring instru-
ment. Calculating this, with the temperatures already
assigned to the various spectral classes, we obtain results
which are in excellent agreement with Coblentz’s obser-
vations throughout their whole range. For the coolest
stars, this affords a better method of measuring tempera-
ture than the color index, which may be considerably

<<   <   3   4   5   6   7   8   9   >   >>