till be the same, provided only that the masses remain unaltered. In
this we observe a profound difference between the attraction of
gravitation and magnetic attraction. In the latter case the attraction
is not perceptible at all in the great majority of substances, and is
only considerable in the case of iron.

In our account of the solar system we have represented the moon as
revolving around the earth in a _nearly_ circular path, and the planets
as revolving around the sun in orbits which are also approximately
circular. It is now our duty to give a more minute description of these
remarkable paths; and, instead of dismissing them as being _nearly_
circles, we must ascertain precisely in what respects they differ
therefrom.

If a planet revolved around the sun in a truly circular path, of which
the sun was always at the centre, it is then obvious that the distance
from the sun to the planet, being always equal to the radius of the
circle, must be of constant magnitude. Now, there can be no doubt that
the distance from the sun to each planet is approximately constant; but
when accurate observations are made, it becomes clear that the distance
is not absolutely so. The variations in distance may amount to many
millions of miles, but, even in extreme cases, the variation in the
distance of the planet is only a small fraction--usually a very small
fraction--of the total amount of that distance. The circumstances vary
in the case of each of the planets. The orbit of the earth itself is
such that the distance from the earth to the sun departs but little from
its mean value. Venus makes even a closer approach to perfectly circular
movement; while, on the other hand, the path of Mars, and much more the
path of Mercury, show considerable relative fluctuations in the distance
from the planet to the sun.

It has often been noticed that many of the great discoveries in science
have their origin in the nice observation and explanation of minute
departures from some law approximately true. We have in this department
of astronomy an excellent illustration of this principle. The orbits of
the planets are nearly circles, but they are not exactly circles. Now,
why is this? There must be some natural reason. That reason has been
ascertained, and it has led to several of the grandest discoveries that
the mind of man has ever achieved in the realms of Nature.

In the first place, let us see the inferences to be drawn from the fact
th