period of the
recurring maximum with greater accuracy.

The course of one of the sun-spot cycles may be described as follows:
For two or three years the spots are both larger and more numerous than
on the average; then they begin to diminish, until in about six or seven
years from the maximum they decline to a minimum; the number of the
spots then begins to increase, and in about four and a half years the
maximum is once more attained. The length of the cycle is, on an
average, about eleven years and five weeks, but both its length and the
intensity of the maxima vary somewhat. For instance, a great maximum
occurred in the summer of 1870, after which a very low minimum occurred
in 1879, followed by a feeble maximum at the end of 1883; next came an
average minimum about August, 1889, followed by the last observed
maximum in January, 1894. It is not unlikely that a second period of
about sixty or eighty years affects the regularity of the eleven-year
period. Systematic observations carried on through a great many years to
come will be required to settle this question, as the observations of
sun-spots previous to 1826 are far too incomplete to decide the issues
which arise.

A curious connection seems to exist between the periodicity of the spots
and their distribution over the surface of the sun. When a minimum is
about to pass away the spots generally begin to show themselves in
latitudes about 30 deg. north and south of the sun's equator; they then
gradually break out somewhat nearer to the equator, so that at the time
of maximum frequency most of them appear at latitudes not greater than
16 deg.. This distance from the sun's equator goes on decreasing till the
time of minimum. Indeed, the spots linger on very close to the equator
for a couple of years more, until the outbreak signalising the
commencement of another period has commenced in higher latitudes.

We have still to note an extraordinary feature which points to an
intimate connection between the phenomena of sun-spots and the purely
terrestrial phenomena of magnetism. It is of course well known that the
needle of a compass does not point exactly to the north, but diverges
from the meridian by an angle which is different in different places and
is not even constant at the same place. For instance, at Greenwich the
needle at present points in a direction 17 deg. West of North, but this
amount is subject to very slow and gradual changes, as well as to very
sma