side of the
glass and gets spilled about, whereas if you pour quickly there is no
room for the great quantity of water to pass into the bottle all at
once, and so it gets spilled again. But if you take a piece of stick or
a glass rod, and hold it against the edge of the tumbler, then the water
runs down the rod and into the bottle, and none is lost (Fig. 7); you
may even hold the rod inclined to one side, as I am now doing, but the
water runs down the wet rod because this elastic skin forms a kind of
tube which prevents the water from escaping. This action is often made
use of in the country to carry the water from the gutters under the roof
into a water-butt below. A piece of stick does nearly as well as an iron
pipe, and it does not cost anything like so much.

[Illustration: Fig. 7.]

I think then I have now done enough to show that on the surface of
water there is a kind of elastic skin. I do not mean that there is
anything that is not water on the surface, but that the water while
there acts in a different way to what it does inside, and that it acts
as if it were an elastic skin made of something like very thin
india-rubber, only that it is perfectly and absolutely elastic, which
india-rubber is not.

You will now be in a position to understand how it is that in narrow
tubes water does not find its own level, but behaves in an unexpected
manner. I have placed in front of the lantern a dish of water coloured
blue so that you may the more easily see it. I shall now dip into the
water a very narrow glass pipe, and immediately the water rushes up and
stands about half an inch above the general level. The tube inside is
wet. The elastic skin of the water is therefore attached to the tube,
and goes on pulling up the water until the weight of the water raised
above the general level is equal to the force exerted by the skin. If I
take a tube about twice as big, then this pulling action which is going
on all round the tube will cause it to lift twice the weight of water,
but this will not make the water rise twice as high, because the larger
tube holds so much more water for a given length than the smaller tube.
It will not even pull it up as high as it did in the case of the smaller
tube, because if it were pulled up as high the weight of the water
raised would in that case be four times as great, and not only twice as
great, as you might at first think. It will therefore only raise the
water in the larger tube to hal