, or marshes,
because at one time they were believed to be such. Probably in some of
those places, if not in all, water existed millions of years ago; but
ages since they must have lost it either by evaporation or by absorption
into the soil.

"I will not say any more just now, but as we pass above the lunar
surface I will point out a few of the natural features that may be of
interest to you."

M'Allister here paid me the compliment of saying, "Well, Professor, I
always thought astronomy was a very dry and difficult subject; but your
remarks were really very interesting, and quite easy to understand.
There is only one thing that seemed to me rather strange as coming from
a scientific man, and I would like you to explain that."

"Certainly; if there is anything you do not quite understand, you have
only to ask and I will try to clear the matter up," I answered. "What
is it you wish to know?"

"Well," he answered, "I noticed that when you were speaking about the
distance of the moon you always said it was _about_ so far away. Why
didn't you tell us the exact distance? I'm not a scientific man by any
means, but if any one were to ask me the length of a connecting rod on
one of my machines I should say '25 inches,' not 'about 25 inches,' for
that would not do for a practical man!"

"It's like this, M'Allister," I said. "You measure things with a
two-foot rule, which is something you can actually handle, and you know
it is made according to a standard measure and must contain exactly 24
inches. If, however, your rule was 24-1/4 inches long, yet still divided
into twenty-four equal parts, you could measure work with it just the
same, but would know that every measurement was just a little bit out.
If you had no possible means of obtaining another rule, you would have
to put up with a little inexactitude.

"That is just the position in which astronomers are placed; they have to
put up with a measure which they know is not perfectly accurate, yet it
is the best which can be secured.

"Their two-foot rule, so to speak, may be the distance from the earth to
the sun, or the length of the whole diameter of the earth's orbit, and
these cannot be handled like your rule; and although we know the
measurements of these are nearly correct, they are not quite so. Yet the
distances of the moon, planets, stars, &c., have to be measured by these
rules, so it is clear we can only know those distances with a near
approximation to acc