use too small a charge, the first portion of the gas will be mixed with
the air already in the retort, and I should be obliged to sacrifice the
first portion of the gas, because it would be so much diluted with air;
the first portion must therefore be thrown away. You will find in this
case, that a common spirit-lamp is quite sufficient for me to get the
oxygen, and so we shall have two processes going on for its preparation.
See how freely the gas is coming over from that small portion of the
mixture. We will examine it, and see what are its properties. Now, in this
way we are producing, as you will observe, a gas just like the one we had
in the experiment with the battery, transparent, undissolved by water, and
presenting the ordinary visible properties of the atmosphere. (As this
first jar contains the air, together with the first portions of the oxygen
set free during the preparation, we will carry it out of the way, and be
prepared to make our experiments in a regular, dignified manner.) And,
inasmuch as that power of making wood, wax, or other things burn, was so
marked in the oxygen we obtained by means of the voltaic battery from
water, we may expect to find the same property here. We will try it You
see there is the combustion of a lighted taper in air, and here is its
combustion in this gas [lowering the taper into the jar]. See how brightly
and how beautifully it burns! You can also see more than this,--you will
perceive it is a heavy gas, whilst the hydrogen would go up like a
balloon, or even faster than a balloon, when not encumbered with the
weight of the envelope.

[Illustration: Fig. 22.]

You may easily see that although we obtained from water twice as much in
volume of the hydrogen as of oxygen, it does not follow that we have twice
as much in weight--because one is heavy, and the other a very light gas.
We have means of weighing gases or air; but without stopping to explain,