To five parts of inflammable air, which had been agitated in water till it was diminished about one half (at which time part of it fired with a weak explosion) I put one part of nitrous air, which diminished it one eighth of the whole. This was done in lime-water, without any precipitation of lime. To compare this with common air, I mixed the same quantity, viz. five parts of this, and one part of nitrous air: when considerable crust of lime was formed upon the surface of the lime water, though the diminution was very little more than in the former process. It is possible, however, that the common air might have taken more nitrous air before it was fully saturated, so as to begin to receive an addition to its bulk.
I agitated in water a quantity of nitrous air phlogisticated with iron filings and brimstone, and found it to be so far restored, that three fourths of an ounce measure of nitrous air being put to two ounce measures of it, made no addition to it.
But the most remarkable of these experiments is that which I made with nitrous air itself which I had no idea of the possibility of reducing to a state fit for respiration by any process whatever, at the time of my former publication on this subject. This air, however, itself, without any previous phlogistication, is purified by agitation in water till it is diminished by fresh nitrous air, and to a very considerable degree.
In a pretty long time I agitated nitrous air in water, supplying it from time to time with more, as the former quantity diminished, till only one eighteenth of the whole quantity remained; in which state it was so wholesome, that a mouse lived in two ounce measures of it more than ten minutes, without shewing any sign of uneasiness; so that I concluded it must have been about as good as air in which candles had burned out. After agitating it again in water, I put one part of fresh nitrous air to five parts of this air, and it was diminished one ninth part. I then agitated it a third time, and putting more nitrous air to it, it was diminished again in the same proportion, and so a fourth time; so that, by continually repeating the process, it would, I doubt not, have been all absorbed. These processes were made in lime-water, without forming any incrustation on the surface of it.
Lastly, I took a quantity of common air, which had been diminished and made noxious by phlogistic processes; and when it had been agitated in water, I found that it was diminished by nitrous air, though not so much as it would have been at the first. After cleansing it a second time, it was diminished again by the same means; and, after that, a third time; and thus there can be no doubt but that, in time, the whole quantity would have disappeared. For I have never found that agitation in water, deprived of its own air, made any addition to a quantity of noxious air; though, a priori, it might have been imagined that, as a saturation with phlogiston diminishes air, the extraction of phlogiston would increase the bulk of it. On the contrary, agitation in water always diminished noxious air a little; indeed, if water be deprived of all its own air, it is impossible to agitate any kind of air in it without some loss. Also, when noxious air has been restored by plants, I never perceived that it gained any addition to its bulk by that means. There was no incrustation of the lime-water in the above-mentioned experiment.
It is not a little remarkable, that those kinds of air which never had been common air, as inflammable air, phlogisticated nitrous air, and nitrous air itself, when rendered wholesome by agitation in water, should be more diminished by fresh nitrous air, than common air which had been made noxious, and restored by the same process; and yet, from the few trials that I have made, I could not help concluding that this is the case.
In this course of experiments I was very near deceiving myself, in consequence of transferring the nitrous air which I made use of in a bladder, in the manner described, p. 15. fig. 9. so as to conclude that there was a precipitation of lime in all the above-mentioned cases, and that even nitrous air itself produced that effect. But after repeated trials, I found that there was no precipitation of lime, except, in the first diminution of common air, when the nitrous air was transferred in a glass vessel.
That the calces of metals contain air, of some kind or other, and that this air contributes to the additional weight of the calces, above that of the metals from which they are made, had been observed by Dr. Hales; and Mr. Hartley had informed me, that when red-lead is boiled in linseed oil, there is a prodigious discharge of air before they incorporate. I had likewise found, that no weight is either gained or lost by the calcination of tin in a close glass vessel; but I purposely deferred making any more experiments on the subject, till we should have some weather in which I could make use of a large burning lens, which I had provided for that and other purposes; but, in the mean time, I was led to the discovery in a different manner.
Having, by the last-recited experiments, been led to consider the electric matter as phlogiston, or something containing phlogiston, I was endeavouring to revivify the calx of lead with it; when I was surprized to perceive a considerable generation of air. It occurred to me, that possibly this effect might arise from the heat communicated to the red-lead by the electric sparks, and therefore I immediately filled a small phial with the red-lead, and heating it with a candle, I presently expelled from it a quantity of air about four or five times the bulk of the lead, the air being received in a vessel of quicksilver. How much more air it would have yielded, I did not try.
Along with the air, a small quantity of water was likewise thrown out; and it immediately occurred to me, that this water and air together must certainly be the cause of the addition of weight in the calx. It still remained to examine what kind of air this was; but admitting water to it, I found that it was imbibed by it, exactly like fixed air, which I therefore immediately concluded it must be[12].