thus have the somewhat astonishing result that a gaseous globe in
space radiating heat, and thereby growing smaller, is all the time
actually increasing in temperature. But, it may be said, surely this
cannot go on for ever. Are we to suppose that the gaseous mass will go
on contracting and contracting with a temperature ever fiercer and
fiercer, and actually radiating out more and more heat the more it
loses? Where lies the limit to such a prospect? As the body contracts,
its density must increase, until it either becomes a liquid, or a solid,
or, at any rate, until it ceases to obey the laws of a purely gaseous
body which we have supposed. Once these laws cease to be observed the
argument disappears; the loss of heat may then really be attended with a
loss of temperature, until in the course of time the body has sunk to
the temperature of space itself.

It is not assumed that this reasoning can be applied in all its
completeness to the present state of the sun. The sun's density is now
so great that the laws of gases cannot be there strictly followed. There
is, however, good reason to believe that the sun was once more gaseous
than at present; possibly at one time he may have been quite gaseous
enough to admit of this reasoning in all its fulness. At present the sun
appears to be in some intermediate stage of its progress from the
gaseous condition to the solid condition. We cannot, therefore, say
that the temperature of the sun is now increasing in correspondence with
the process of contraction. This may be true or it may not be true; we
have no means of deciding the point. We may, however, feel certain that
the sun is still sufficiently gaseous to experience in some degree the
rise of temperature associated with the contraction. That rise in
temperature may be partly or wholly obscured by the fall in temperature
which would be the more obvious consequence of the radiation of heat
from the partially solid body. It will, however, be manifest that the
cooling of the sun may be enormously protracted if the fall of
temperature from the one cause be nearly compensated by the rise of
temperature from the other. It can hardly be doubted that in this we
find the real explanation of the fact that we have no historical
evidence of any appreciable alteration in the radiation of heat from the
sun.

This question is one of such interest that it may be worth while to look
at it from a slightly different point of view. The sun con