neighborhood to a distance exceeding that of the earth. But in neither
form is the hypothesis satisfactory; there is nothing in the
appearance of the corona to indicate that it extends even as far as
the planet Mercury, while as to meteors, the orbits of the known
swarms do not accord with the hypothesis, and we have no reason to
believe that clouds of others exist traveling in the part of space
where they would have to be in order to answer the requirements of the
theory. The extension of the corona in 1878 did not resemble in its
texture the Zodiacal Light.

Now, it has so often happened in the history of science that an
important discovery in one branch has thrown unexpected but most
welcome light upon some pending problem in some other branch, that a
strong argument might be based upon that fact alone against the too
exclusive devotion of many investigators to the narrow lines of their
own particular specialty; and the Zodiacal Light affords a case in
point, when it is considered in connection with recent discoveries in
chemistry and physics. From the fact that atoms are compound bodies
made up of corpuscles at least a thousand times smaller than the
smallest known atom -- a fact which astounded most men of science when
it was announced a few years ago -- a new hypothesis has been
developed concerning the nature of the Zodiacal Light (as well as
other astronomical riddles), and this hypothesis comes not from an
astronomer, but from a chemist and physicist, the Swede, Svante
Arrhenius. In considering an outline of this new hypothesis we need
neither accept nor reject it; it is a case rather for suspension of
judgment.

To begin with, it carries us back to the ``pressure of light''
mentioned in the preceding chapter. The manner in which this pressure