seen only by a practised eye under the most favorable conditions.
But it is always there. The latest suggestion is that it is a tail
of the earth, of the same kind as the tail of a comet!

We know that the motions of the heavenly bodies are predicted with
extraordinary exactness by the theory of gravitation. When one
finds that the exact path of the moon's shadow on the earth during
a total eclipse of the sun can be mapped out many years in
advance, and that the planets follow the predictions of the
astronomer so closely that, if you could see the predicted planet
as a separate object, it would look, even in a good telescope, as
if it exactly fitted over the real planet, one thinks that here at
least is a branch of astronomy which is simply perfect. And yet
the worlds themselves show slight deviations in their movements
which the astronomer cannot always explain, and which may be due
to some hidden cause that, when brought to light, shall lead to
conclusions of the greatest importance to our race.

One of these deviations is in the rotation of the earth.
Sometimes, for several years at a time, it seems to revolve a
little faster, and then again a little slower. The changes are
very slight; they can be detected only by the most laborious and
refined methods; yet they must have a cause, and we should like to
know what that cause is.

The moon shows a similar irregularity of motion. For half a
century, perhaps through a whole century, she will go around the
earth a little ahead of her regular rate, and then for another
half-century or more she will fall behind. The changes are very
small; they would never have been seen with the unaided eye, yet