light and of
heat. The heat acquired by the earth from the flashing of the shooting
stars through our air is quite insensible. It has been supposed,
however, that the heat accruing to the sun from the same cause may be
quite sensible--nay, it has been even supposed that the sun may be
re-invigorated from this source.

Here, again, we must apply the cold principles of weights and measures
to estimate the plausibility of this suggestion. We first calculate the
actual weight of meteoric indraught to the sun which would be adequate
to sustain the fires of the sun at their present vigour. The mass of
matter that would be required is so enormous that we cannot usefully
express it by imperial weights; we must deal with masses of imposing
magnitude. It fortunately happens that the weight of our moon is a
convenient unit. Conceive that our moon--a huge globe, 2,000 miles in
diameter--were crushed into a myriad of fragments, and that these
fragments were allowed to rain in on the sun; there can be no doubt that
this tremendous meteoric shower would contribute to the sun rather more
heat than would be required to supply his radiation for a whole year. If
we take our earth itself, conceive it comminuted into dust, and allow
that dust to fall on the sun as a mighty shower, each fragment would
instantly give out a quantity of heat, and the whole would add to the
sun a supply of heat adequate to sustain the present rate of radiation
for nearly one hundred years. The mighty mass of Jupiter treated in the
same way would generate a meteoric display greater in the ratio in which
the mass of Jupiter exceeds the mass of earth. Were Jupiter to fall into
the sun, enough heat would be thereby produced to scorch the whole solar
system; while all the planets together would be capable of producing
heat which, if properly economised, would supply the radiation of the
sun for 45,000 years.

It must be remembered that though the moon could supply one year's heat,
and Jupiter 30,000 years' heat, yet the practical question is not
whether the solar system could supply the sun's heat, but whether it
does. Is it likely that meteors equal in mass to the moon fall into the
sun every year? This is the real question, and I think we are bound to
reply to it in the negative. It can be shown that the quantity of
meteors which could be caught by the sun in any one year can be only an
excessively minute fraction of the total amount. If, therefore, a
moon-weight