not notice this, astronomers
know it and have to make allowance for it. The effect is most
noticeable in the case of the sun when he is going down, for the
atmosphere bends his rays up, and though we see him a great glowing red
ball on the horizon, and watch him, as we think, drop gradually out of
sight, we are really looking at him for the last moment or two when he
has already gone, for the rays are bent up by the air and his image
lingers when the real sun has disappeared.

[Illustration: A STICK THRUST INTO THE WATER APPEARS CROOKED.]

Therefore in looking through the luminous stuff that forms a comet's
tail astronomers might well expect to see the stars displaced, but not a
sign of this appears. It is difficult to imagine, therefore, what the
tail can be made of. The idea is that the sun exercises a sort of
repulsive effect on certain elements found in the comet's head--that is
to say, it pushes them away, and that as the head approaches the sun,
these elements are driven out of it away from the sun in vapour. This
action may have something to do with electricity, which is yet little
understood; anyway, the effect is that, instead of attracting the matter
toward itself, in which case we should see the comet's tails stretching
toward the sun, the sun drives it away! In the chapter on the sun we had
to imagine something of the same kind to account for the corona, and the
corona and the comet's tails may be really akin to each other, and
could perhaps be explained in the same way. Now we come to a stranger
fact still. Some comets go right through the sun's corona, and yet do
not seem to be influenced by it in the smallest degree. This may not
seem very wonderful at first perhaps, but if you remember that a dash
through anything so dense as our atmosphere, at a pace much less than
that at which a comet goes, is enough to heat iron to a white heat, and
then make it fly off in vapour, we get a glimpse of the extreme fineness
of the materials which make the corona.

Here is Herschel's account of a comet that went very near the sun:

'The comet's distance from the sun's centre was about the 160th part of
our distance from it. All the heat we enjoy on this earth comes from the
sun. Imagine the heat we should have to endure if the sun were to
approach us, or we the sun, to one 160th part of its present distance.
It would not be merely as if 160 suns were shining on us all at once,
but, 160 times 160, according to a rule w