_ be of the Sun. Under no
circumstances can there be in any one year more than 3 eclipses of the
Moon, and in some years there will be none. The reasons for these
diversities are of a technical character, and a full elucidation of them
would not be of interest to the general reader. It may here be added,
parenthetically, that the occasions will be very rare of there being 5
solar eclipses in one year. This last happened in 1823,[2] and will
only happen once again in the next two centuries, namely in 1935. If a
total eclipse of the Sun happens early in January there may be another
in December of the same year, as in 1889 (Jan. 1 and Dec. 22). This will
not happen again till 2057, when there will be total eclipses on Jan. 5
and Dec. 26. There is one very curious fact which may be here
conveniently stated as a bare fact, reserving the explanation of it for
a future page, namely, that eclipses of the Sun and Moon are linked
together in a certain chain or sequence which takes rather more than 18
years to run out when the sequence recurs and recurs _ad infinitum_. In
this 18-year period, which bears the name of the "Saros," there usually
happen 70 eclipses, of which 41 are of the Sun and 29 of the Moon.
Accordingly, eclipses of the Sun are more numerous than those of the
Moon in the proportion of about 3 to 2, yet at any given place on the
Earth more lunar eclipses are visible than solar eclipses, because the
former when they occur are visible over the whole hemisphere of the
Earth which is turned towards the Moon whilst the area over which a
total eclipse of the Sun is visible is but a belt of the Earth no more
than about 150 to 170 miles wide. Partial eclipses of the Sun, however,
are visible over a very much wider area on either side of the path
traversed by the Moon's shadow.

[Illustration: FIG. 2.--THEORY OF A TOTAL ECLIPSE OF THE SUN.]

Confining our attention in the first instance to eclipses of the Sun,
the diagrams fig. 2 and fig. 3 will make clear, with very little verbal
description, the essential features of the two principal kinds of
eclipses of the Sun. In these figures S represents the Sun, M the Moon
and E the Earth. They are not, of course, even approximately drawn to
scale either as to the size of the bodies or their relative distances,
but this is a matter of no moment as regards the principles involved. M
being in sunshine receives light on, as it were, the left hand side,
which faces S the Sun. The shad