he earth's revolution when it is nearest the
sun, and the slowness when it is farther away.

There is a further complication arising from that curious phenomenon
called the precession of the equinoxes, which has to be taken into
account before we can sufficiently comprehend the effect of the
varying eccentricity of the orbit on the earth's seasons. To
understand this feature of precession we should first note that it
means that each year the change from the winter to the summer--or, as
we phrase it, the passage of the equinoctial line--occurs a little
sooner than the year before. The cause of this is to be found in the
attraction which the heavenly bodies, practically altogether the moon,
exercises on the equatorial protuberance of the earth. We know that
the diameter of our sphere at the equator is, on the average,
something more than twenty-six miles greater than it is through the
poles. We know, furthermore, that the position of the moon in relation
to the earth is such that it causes the attraction on one half of this
protuberance to be greater than it is upon the other. We readily
perceive that this action will cause the polar axis to make a certain
revolution, or, what comes to the same thing, that the plane of the
equator will constantly be altering its position. Now, as the
equinoctial points in the orbit depend for their position upon the
attitude of the equatorial plane, we can conceive that the effect is a
change in position of the place in that orbit where summer and winter
begin. The actual result is to bring the seasonal points backward,
step by step, through the orbit in a regular measure until in
twenty-two thousand five hundred years they return to the place where
they were before. This cycle of change was of old called the Annus
Magnus, or great year.

If the earth's orbit were an ellipse, the major axis of which remained
in the same position, we could readily reckon all the effects which
arise from the variations of the great year. But this ellipse is ever
changing in form, and in the measure of its departure from a circle
the effects on the seasons distributed over a great period of time are
exceedingly irregular. Now and then, at intervals of hundreds of
thousands or millions of years, the orbit becomes very elliptical;
then again for long periods it may in form approach a circle. When in
the state of extreme ellipticity, the precession of the equinoxes will
cause the hemispheres in turn each to h