f the height, and now that the two tubes
are side by side you see the water in the smaller tube standing twice as
high as it does in the larger tube. In the same way, if I were to take a
tube as fine as a hair the water would go up ever so much higher. It is
for this reason that this is called Capillarity, from the Latin word
_capillus_, a hair, because the action is so marked in a tube the size
of a hair.

[Illustration: Fig. 8.]

Supposing now you had a great number of tubes of all sizes, and placed
them in a row with the smallest on one side and all the others in the
order of their sizes, then it is evident that the water would rise
highest in the smallest tube and less and less high in each tube in the
row (Fig. 8), until when you came to a very large tube you would not be
able to see that the water was raised at all. You can very easily
obtain the same kind of effect by simply taking two square pieces of
window glass and placing them face to face with a common match or small
fragment of anything to keep them a small distance apart along one edge
while they meet together along the opposite edge. An india-rubber ring
stretched over them will hold them in this position. I now take this
pair of plates and stand it in a dish of coloured water, and you at once
see that the water creeps up to the top of the plates on the edge where
they meet, and as the distance between the plates gradually increases,
so the height to which the water rises gradually gets less, and the
result is that the surface of the liquid forms a beautifully regular
curve which is called by mathematicians a rectangular hyperbola (Fig.
9). I shall have presently to say more about this and some other curves,
and so I shall not do more now than state that the hyperbola is formed
because as the width between the plates gets greater the height gets
less, or, what comes to the same thing, because the weight of liquid
pulled up at any small part of the curve is always the same.

[Illustration: Fig. 9.]

If the plates or the tubes had been made of material not wetted by
water, then the effect of the tension of the surface would be to drag
the liquid away from the narrow spaces, and the more so as the spaces
were narrower. As it is not easy to show this well with paraffined glass
plates or tubes and water, I shall use another liquid which does not wet
or touch clean glass, namely, quicksilver. As it is not possible to see
through quicksilver, it will not do