periods of time in order that everything possible to learn may be
discovered.

Our own moon is one of the enigmas of the mathematical astronomer.
Observations show that she is deviating from her predicted place,
and that this deviation continues to increase. True, it is not
very great when measured by an ordinary standard. The time at
which the moon's shadow passed a given point near Norfolk during
the total eclipse of May 29, 1900, was only about seven seconds
different from the time given in the Astronomical Ephemeris. The
path of the shadow along the earth was not out of place by more
than one or two miles But, small though these deviations are, they
show that something is wrong, and no one has as yet found out what
it is. Worse yet, the deviation is increasing rapidly. The
observers of the total eclipse in August, 1905, were surprised to
find that it began twenty seconds before the predicted time. The
mathematical problems involved in correcting this error are of
such complexity that it is only now and then that a mathematician
turns up anywhere in the world who is both able and bold enough to
attack them.

There now seems little doubt that Jupiter is a miniature sun, only
not hot enough at its surface to shine by its own light The point
in which it most resembles the sun is that its equatorial regions
rotate in less time than do the regions near the poles. This shows
that what we see is not a solid body. But none of the careful
observers have yet succeeded in determining the law of this
difference of rotation.

Twelve years ago a suspicion which had long been entertained that