rings, with its mouth projecting over
the shelf. When the phial is thus conveyed into the jar, the cork may
easily be removed, and may also be put into it again at pleasure, and
conveyed the same way out again.

When any thing, as a gallipot, &c. is to be supported at a considerable
height within a jar, it is convenient to have such _wire stands_ as are
represented fig. 5. They answer better than any other, because they take
up but little room, and may be easily bended to any shape or height.

If I have occasion to pour air from a vessel with a wide mouth into
another with a very narrow one, I am obliged to make use of a funnel,
fig. 6, but by this means the operation is exceedingly easy; first
filling the vessel into which the air is to be conveyed with water, and
holding the mouth of it, together with the funnel, both under water with
one hand, while the other is employed in pouring the air; which,
ascending through the funnel up into the vessel, makes the water
descend, and takes its place. These funnels are best made of glass,
because the air being visible through them, the quantity of it may be
more easily estimated by the eye. It will be convenient to have several
of these funnels of different sizes.

In order to expel air from solid substances by means of heat, I
sometimes put them into a _gun-barrel_, fig. 7, and filling it up with
dry sand, that has been well burned, so that no air can come from it, I
lute to the open end the stem of a tobacco pipe, or a small glass tube.
Then having put the closed end of the barrel, which contains the
materials, into the fire, the generated air, issuing through the tube,
may be received in a vessel of quicksilver, with its mouth immersed in a
bason of the same, suspended all together in wires, in the manner
described in the figure: or any other fluid substance may be used
instead of quicksilver.

But the most accurate method of procuring air from several substances,
by means of heat, is to put them, if they will bear it, into phials full
of quicksilver, with the mouths immersed in the same, and then throw the
focus of a burning mirror upon them. For this purpose the phials should
be made with their bottoms round, and very thin, that they may not be
liable to break with a pretty sudden application of heat.

If I want to expel air from any liquid, I nearly fill a phial with it,
and having a cork perforated, I put through it, and secure with cement,
a glass tube, bended in t