composition of all members of this system by assuming that they
were all once a part of the same body, and you may say brothers
and sisters of the sun, instead of its offspring. It also makes
size the only factor determining temperature and density, but of
course modified by age, since otherwise Jupiter would have a far
less developed crust than that with which we find it. I have
always considered the period from the molten condition to that
with a crust as comparatively short, which stands to reason, for
radiation has then no check; and the period from the formation of
the crust, which acts as a blanket, to the death of a planet, as
very long. I have not found this view clearly set forth in any
of the books I have read, but it seems to me the simplest and
most natural explanation. Now, granted that the solar system was
once a nebula, on which I think every one will agree--the same
forces that changed it into a system of sun and planets must be
at work on fifty-one M. Canum venaticorum, Andromeda, and ninety-
nine M. Virginis, and must inevitably change them to suns, each
with doubtless a system of planets.