air was carefully pressed. I then put the middle part of the
stem into a chafing-dish of hot coals, strongly urged with a pair of
bellows; and, pressing the bladders alternately, I made the air pass
several times through the heated part of the pipe. I have also made
this kind of air very hot, standing in water before the fire. But
neither of these methods were of any use.

_Rarefaction_ and _condensation_ by instruments were also tried, but in
vain.

Thinking it possible that the _earth_ might imbibe the noxious quality
of the air, and thence supply the roots of plants with such putrescent
matter as is known to be nutritive to them, I kept a quantity of air, in
which mice had died, in a phial, one half of which was filled with fine
garden-mould; but, though it stood two months in these circumstances, it
was not the better for it.

I once imagined that, since several kinds of air cannot be long
separated from common air, by being confined in bladders, in bottles
well corked; or even closed with ground stopples, the affinity between
this noxious air and the common air might be so great, that they would
mix through a body of water interposed between them; the water
continually receiving from the one, and giving to the other, especially
as water receives some kind of impregnation from, I believe, every kind
of air to which it is contiguous; but I have seen no reason to
conclude, that a mixture of any kind of air with the common air can be
produced in this manner.

I have kept air in which mice have died, air in which candles have
burned out, and inflammable air, separated from the common air, by the
slightest partition of water that I could well make, so that it might
not evaporate in a day or two, if I should happen not to attend to them;
but I found no change in them after a month or six weeks. The
inflammable air was still inflammable, mice died instantly in the air in
which other mice had died before, and candles would not burn where they
had burned out before.

Since air tainted with animal or vegetable putrefaction is the same
thing with air rendered noxious by animal respiration, I shall now
recite the observations which I have made upon this kind of air, before
I treat of the method of restoring them.

That these two kinds of air are, in fact, the same thing, I conclude
from their having several remarkable common properties, and from their
differing in nothing that I have been able to observe. They equally