ctically no gas will escape. Hold them quietly this way,
the tube of gas uppermost, for not less than one full minute
by the clock. A minute and a half is not too much time. Now
have some one light a match for you, or else go to a lighted
Bunsen burner.

Take the test tubes apart gently and hold the lower one, which
was full of air, with its mouth to the flame. What has the gas
in the upper tube done? Now hold the flame to the upper test
tube, which was full of gas. What happens? Has all the gas
gone out of it?

[Illustration: FIG. 149. Filling a test tube with gas.]

[Illustration: FIG. 150. The lower test tube is full of air; the upper,
of gas. What will happen when the cardboard is withdrawn?]

As you well know, gas is much lighter than air; you can make a balloon
rise by filling it with gas. Yet part of the gas went _down_ into the
lower tube. The explanation is that the molecules of gas and those of
air were flying around at such a rate that many of the gas molecules
went shooting down among the air molecules, and many of the molecules
of air went shooting up among those of gas, so that the gas and the
air became mixed.

DIFFUSION IN LIQUIDS. Diffusion takes place in liquids, as you know.
For when you put sugar in coffee or tea and do _not_ stir it, although
the upper part of the tea or coffee is not sweetened, the part nearer
the sugar is very sweet. If you should let the coffee or tea, with the
sugar in the bottom, stand for a few months, it would get sweet all
through. Diffusion is slower in liquids than in gases, because the
molecules are so very much closer together.

OSMOSIS. One of the most striking and important facts about diffusion
is that it can take place right through a membrane. Try this
experiment:

EXPERIMENT 86. With a rubber band fasten a piece of parchment
paper, made into a little bag, to the end of a piece of glass
tubing about 10 inches long. Or make a small hole in one end
of a raw egg and empty the shell; then, to get the hard
part off the shell, soak it overnight in strong vinegar or
hydrochloric acid diluted about 1 to 4. This will leave a
membranous bag that can be used in place of the parchment bag.
Fill a tumbler half full of water colored with red ink, and
add enough cornstarch to make the water milky. Pour into the
tube enough of a strong sugar solution to fill the membranous
bag at its base and