he vast expanse
of ocean, bears but few obvious traces of a high temperature.
Nevertheless, it is highly probable from ordinary volcanic phenomena
that the interior of the earth still possesses a temperature of
incandescence.

A large body when heated takes a longer time to cool than does a small
body raised to the same temperature. A large iron casting will take days
to cool; a small casting will become cold in a few hours. Whatever may
have been the original source of heat in our system--a question which
we are not now discussing--it seems demonstrable that the different
bodies were all originally heated, and have now for ages been gradually
cooling. The sun is so vast that he has not yet had time to cool; the
earth, of intermediate bulk, has become cold on the outside, while still
retaining vast stores of internal heat; while the moon, the smallest
body of all, has lost its heat to such an extent that changes of
importance on its surface can no longer be originated by internal fires.

We are thus led to refer the origin of the lunar craters to some ancient
epoch in the moon's history. We have no moans of knowing the remoteness
of that epoch, but it is reasonable to surmise that the antiquity of the
lunar volcanoes must be extremely great. At the time when the moon was
sufficiently heated to originate those convulsions, of which the mighty
craters are the survivals, the earth must also have been much hotter
than it is at present. When the moon possessed sufficient heat for its
volcanoes to be active, the earth was probably so hot that life was
impossible on its surface. This supposition would point to an antiquity
for the lunar craters far too great to be estimated by the centuries and
the thousands of years which are adequate for such periods as those with
which the history of human events is concerned. It seems not unlikely
that millions of years may have elapsed since the mighty craters of
Plato or of Copernicus consolidated into their present form.

We shall now attempt to account for the formation of the lunar craters.
The most probable views on the subject seem to be those which have been
set forth by Mr. Nasmyth, though it must be admitted that his doctrines
are by no means free from difficulty. According to his theory we can
explain how the rampart around the lunar crater has been formed, and how
the great mountain arose which so often adorns the centre of the plain.
The view in Fig. 28 contains an imaginary