ed,
however, that it is not heavier than common air, which must have been
manifest, if so great a diminution of the quantity had been owing, as
Dr. Hales and others supposed, to the elasticity of the whole mass being
impaired. After making several trials for this purpose, I concluded that
air, thus diminished in bulk, is rather lighter than common air, which
favours the supposition of the fixed, or heavier part of the common air,
having been precipitated.

An animal will live nearly, if not quite as long, in air in which
candles have burned out, as in common air. This fact surprized me very
greatly, having imagined that what is called the _consumption_ of air by
flame, or respiration, to have been of the same nature, and in the same
degree; but I have since found, that this fact has been observed by many
persons, and even so early as by Mr. Boyle. I have also observed, that
air, in which brimstone has burned, is not in the least injurious to
animals, after the fumes, which at first make it very cloudy, have
intirely subsided.

I must, in this place, admonish my reader not to confound the simple
_burning of brimstone_, or of matches (_i. e._ bits of wood dipped in
it) and the burning of brimstone with a burning mirror, or any _foreign
heat_. The effect of the former is nothing more than that of any other
_flame_, or _ignited vapour_, which will not burn, unless the air with
which it is surrounded be in a very pure state, and which is therefore
extinguished when the air begins to be much vitiated. Lighted brimstone,
therefore reduces the air to the same state as lighted wood. But the
focus of a burning mirror thrown for a sufficient time either upon
brimstone, or wood, after it has ceased to burn of its own accord, and
has become _charcoal_, will have a much greater effect: of the same
kind, diminishing the air to its utmost extent, and making it thoroughly
noxious. In fact, as will be seen hereafter, more phlogiston is expelled
from these substances in the latter case than in the former. I never,
indeed, actually carried this experiment so far with brimstone; but from
the diminution of air that I did produce by this means, I concluded
that, by continuing the process some time longer, it would have been
effected.

Having read, in the Memoirs of the Philosophical Society at Turin, vol.
I. p. 41. that air in which candles had burned out was perfectly
restored, so that other candles would burn in it again as well as ever,