axis is parallel to these printed lines. This
ellipse represents the orbit of the earth. Every
ellipse has two foci. Suppose the sun to lie in the
left focus of our ellipse: the major axis meets
the curve of the oval at two points, one to the
right and one to the left. The point to the left
evidently makes the nearest approach to the sun
during the whole course of the earth's orbit,
whence it is called the perihelion; the point to
the right, being the most distant, is called the
aphelion. These two extreme points are also
called apsides, whence the name of apsidal line
given to the major axis of the orbit.

Now, in the year 1248, the first day of winter,
or the winter solstice—we are speaking of our
hemisphere—occurred when the earth was passing
the perihelion; and the first day of summer,
or the summer solstice, when the earth was
passing the aphelion. To complete our figure,
let us draw through the centre of the sun a
straight line perpendicular to the major axis of
the ellipse. This line will cut the oval orbit of
the earth at two points, one above and the other
below the major axis. The first is the point
where, in 1248, the earth was at the autumnal
equinox, that is, it was the first day of autumn;
and the second where she was at the vernal
equinox, or on the first day in spring. With
this simple figure in view, the rest of our
explanation is as plain as can be.

During the whole course of a single year
there is no sensible change in the inclination of
the earth's axis; it remains, to all intents and
purposes, parallel to itself; but, in the lapse of
ages, this parallelism remains no longer unaltered.
The earth is slightly swollen, or bulges out at
the equator; the sun's attraction acting on this
swelling has the effect of changing the
inclination of the axis. It is analogous to the
rolling of a top whilst its toe remains spinning
on exactly the same spot of ground.
The top's axis, more or less inclined towards
the ground, describes a conical surface round
the line perpendicular to the plane on which
the top is spinning. The solar attraction,
combined with the diurnal movement, impresses a
similar movement on the globe. This change
of direction has the effect of altering the date of
the equinoxes.

In this way, since 1248, the vernal equinox
has drawn nearer to the perihelion by more than
ten degrees; consequently, the winter solstice,
which was at the perihelion, retreats from it,
the autumnal equinox advances towards the
summer solstice, and the summer solstice
towards the vernal equinox, for all the points of
the orbit follow the same movement. As this
movement goes on, the vernal equinox will at
last take place at the time of the earth's passing
the perihelion; it will then be beyond it, and
will in time take the place of the autumnal
equinox, which will have taken the place of the
vernal equinox, exactly as the two solstices will
also have mutually changed their positions.
That is to say, at that time the order of the
seasons will be reversed, in respect to the four
principal points of the earth's orbit; our spring
and summer will take place at the perihelion, our
autumn and winter at the aphelion. The contrary
will be the case for the southern hemisphere.
Our autumn and winter will then be seven days
longer than in the southern hemisphere, and
every year the sun will shine seven days longer
on the South Pole than on the North Pole.

Be it remarked that the change of the
equinoctial points takes place in a direction opposite
to that of the earth's motion in her orbit, whence
the name of the "Precession of the Equinoxes"
given to this grand phenomenon, which has long
been known to astronomers, although M. Adhémar
was the first to build upon it his theory
of the periodicity of great deluges. The rate
of the movement of precession is so slow that
its entire revolution round the earth's orbit
requires twenty-five thousand eight hundred and
sixty-eight years; but, in fact, it is practically
shortened by another phenomenon, which
modifies the duration of this long period:

In consequence of the attractions exercised by
the planets upon our globe, the major axis of
the earth's orbit, or the apsidal line, changes its
place; it moves in the place of the orbit, and in
the same direction as the earth itself, and,
consequently, in direction contrary to the equinoctial
revolution. Thus—to have a clear idea of the
case—whilst the vernal equinox goes backwards
towards the perihelion, the perihelion, in
consequence of the gradual motion of the apsidal
line, comes forward to meet the vernal equinox.
The effect of the displacement of the apsides is,
therefore, a shortening of the duration of the
revolution of the equinoxes, abbreviating it, in
round numbers, to twenty-one thousand years.
Consequently, every ten thousand five hundred
years, the order of the seasons in the two
hemispheres is reversed, in respect to the equinoctial
and solstitial points; that is, the dates when
spring, summer, autumn, and winter begin.

Now let us follow the bold deductions that
are drawn from these established astronomical
premises. At the poles, the year consists of
only one day and one night; the day lasts as
long as the united spring and summer of the
corresponding hemisphere; the night, as long
as the united autumn and winter; consequently
the North Pole's day is one hundred
and sixty-eight hours longer than the South
Pole's. But, during day, the earth receives
the heat of the sun, whilst, by night, it loses
it by radiation. If the length of the days is
greater than the length of the nights, the earth
is heated; it is cooled in the contrary case.
Hence the North Pole every year lays in a stock
of heat, and the South Pole loses heat. The
former accumulates every year one hundred and
sixty-eight times the quantity of heat received
from the sun in the course of an hour of day, the
second loses one hundred and sixty-eight times
the heat which is dispersed by radiation in the
course of an hour of night. What is the amount
of this difference in the course of several
thousand years—of ten thousand years, for instance?
Three million three hundred and sixty thousand
hours, which are equivalent to two hundred and