ater vapor. Indeed, he supposes that even the earth is slowly losing
its water by evaporation into space, and on Mars, owing to the slight
force of gravity there, this process would go on much more rapidly, so
that, in this way, we have a means of accounting for the apparent drying
up of that planet, while we may be led to anticipate that at some time
in the remote future the earth also will begin to suffer from lack of
water, and that eventually the chasms of the sea will yawn empty and
desolate under a cloudless sky.

But it is not certain that the original supply of atmospheric elements
was in every case proportional to the respective force of gravity of a
planet. The fact that Venus appears to have an atmosphere more extensive
and denser than the earth's, although its force of gravity is a little
less than that of our globe, indicates at once a variation as between
these two planets in the amount of atmospheric material at their
disposal. This may be a detail depending upon differences in the mode,
or in the stage, of their evolution. Thus, after all, Dr. Stoney's
theory may be substantially correct and yet Mars may retain sufficient
water to form clouds, to be precipitated in snow, and to fill its canals
after each annual melting of the polar caps, because the original supply
was abundant, and its escape is a gradual process, only to be completed
by age-long steps.

Even though the evidence of the spectroscope, as far as it goes, seems
to lend support to the theory that there is no water vapor in the
atmosphere of Mars, we can not disregard the visual evidence that,
nevertheless, water vapor exists there.

What are the polar caps if they are not snow? Frozen carbon dioxide, it
has been suggested; but this is hardly satisfactory, for it offers no
explanation of the fact that when the polar caps diminish, and in
proportion as they diminish, the "seas" and the canals darken and
expand, whereas a reasonable explanation of the correlation of these
phenomena is offered if we accept the view that the polar caps consist
of snow.

Then there are many observations on record indicating the existence of
clouds in Mars's atmosphere. Sometimes a considerable area of its
surface has been observed to be temporarily obscured, not by dense
masses of cloud such as accompany the progress of great cyclonic storms
across the continents and oceans of the earth, but by comparatively thin
veils of vapor such as would be expected to for