or less
perfect cloud-belts suggestive of those on Jupiter. A view of our earth
would be, as it were, intermediate between a view of Jupiter and of
Mars. In the latter case the appearance of the permanent features of the
planet is only to a trifling extent obscured by clouds floating over the
surface. Our earth would always be partly, and often perhaps very
largely, covered with cloud, while Jupiter seems at all times completely
enveloped.

From another class of observations we are also taught the important
truth that Jupiter is not, superficially at least, a solid body. The
period of rotation of the planet around its axis is derived from the
observation of certain marks, which present sufficient definiteness and
sufficient permanence to be suitable for the purpose. Suppose one of
these objects to lie at the centre of the planet's disc; its position is
carefully measured, and the time is noted. As the hours pass on, the
mark moves to the edge of the disc, then round the other side of the
planet, and back again to the visible disc. When it has returned to the
position originally occupied the time is again taken, and the interval
which has elapsed is called the period of rotation of the spot.

If Jupiter were a solid, and if these features were engraved upon its
surface, then it is perfectly clear that the time of rotation as found
by any one spot must coincide precisely with the time yielded by any
other spot; but this is not observed to be the case. In fact, it would
be nearer the truth to say that each spot gives a special period of its
own. Nor are the differences very minute. It has been found that the
time in which the red spot (the latitude of which is about 25 deg. south) is
carried round is five minutes longer than that required by some peculiar
white marks near the equator. The red spot has now been watched for
about twenty years, and during most of that time has had a tendency to
rotate more and more slowly, as may be seen from the following values of
its rotation period:--

In 1879, 9h. 55m. 33.9s.
In 1886, 9h. 55m. 40.6s.
In 1891, 9h. 55m. 41.7s.

Since 1891 this tendency seems to have ceased, while the spot has been
gradually fading away. Generally speaking, we may say that the
equatorial regions rotate in about 9h. 50m. 20s., and the temperate
zones in about 9h. 55m. 40s. Remarkable exceptions are occasionally met
with. Some small black spots in north latitude 22 deg., which broke out in
1880 an