acting power of our system--they will, in
fact, be nearly all careering round the sun.

Each planet may, in truth, have a certain following of its own. Within
the limited sphere of the earth's predominant attraction, for instance,
extending some way beyond the moon, we may have a number of satellites
that we never see, all revolving regularly in elliptic orbits round the
earth. But, comparatively speaking, these satellite meteorites are few.
The great bulk of them will be of a planetary character--they will be
attendant upon the sun.

It may seem strange that such minute bodies should have regular orbits
and obey Kepler's laws, but they must. All three laws must be as
rigorously obeyed by them as by the planets themselves. There is nothing
in the smallness of a particle to excuse it from implicit obedience to
law. The only consequence of their smallness is their inability to
perturb others. They cannot appreciably perturb either the planets they
approach or each other. The attracting power of a lump one million tons
in weight is very minute. A pound, on the surface of such a body of the
same density as the earth, would be only pulled to it with a force equal
to that with which the earth pulls a grain. So the perturbing power of
such a mass on distant bodies is imperceptible. It is a good thing it is
so: accurate astronomy would be impossible if we had to take into
account the perturbations caused by a crowd of invisible bodies.
Astronomy would then approach in complexity some of the problems of
physics.

But though we may be convinced from the facts of gravitation that these
meteoric stones, and all other bodies flying through space near our
solar system, must be constrained by the sun to obey Kepler's laws, and
fly round it in some regular elliptic or hyperbolic orbit, what chance
have we of determining that orbit? At first sight, a very poor chance,
for we never see them except for the instant when they splash into our
atmosphere; and for them that instant is instant death. It is unlikely
that any escape that ordeal, and even if they do, their career and orbit
are effectually changed. Henceforward they must become attendants on the
earth. They may drop on to its surface, or they may duck out of our
atmosphere again, and revolve round us unseen in the clear space between
earth and moon.

Nevertheless, although the problem of determining the original orbit of
any given set of shooting-stars before it struck us would