cannot yet be given, for the evidence is contradictory, and the
interpretations put upon it depend largely on the predilections of the
judges.

But, in a broad sense, the sun-spots and the phenomena connected with
them must have a relation to terrestial meteorology, for they prove
the sun to be a variable star. Reference was made, a few lines above,
to the resemblance of the spectra of sun-spots to those of certain
stars which seem to be failing through age. This in itself is
extremely suggestive; but if this resemblance had never been
discovered, we should have been justified in regarding the sun as
variable in its output of energy; and not only variable, but probably
increasingly so. The very inequalities in the sun-spot cycle are
suspicious. When the sun is most spotted its total light may be
reduced by one-thousandth part, although it is by no means certain
that its outgiving of thermal radiations is then reduced. A loss of
one-thousandth of its luminosity would correspond to a decrease of
.0025 of a stellar magnitude, considering the sun as a star viewed
from distant space. So slight a change would not be perceptible; but
it is not alone sun-spots which obscure the solar surface, its entire
globe is enveloped with an obscuring veil. When studied with a
powerful telescope the sun's surface is seen to be thickly mottled
with relatively obscure specks, so numerous that it has been estimated
that they cut off from one-tenth to one-twentieth of the light that we
should receive from it if the whole surface were as brilliant as its
brightest parts. The condition of other stars warrants the conclusion
that this obscuring envelope is the product of a process of
refrigeration which will gradually make the sun more and more variable
until its history ends in extinction. Looking backward, we see a time
when the sun must have been more brilliant than it is now. At that