I took some pains to ascertain the quantity of diminution, in fresh made and very highly-inflammable air from iron, at which it ceased to be inflammable, and, upon the whole, I concluded that it was so when it was diminished a little more than one half; for a quantity which was diminished exactly one half had something inflammable in it, but in the slightest degree imaginable. It is not improbable, however, but there may be great differences in the result of this experiment. Finding that water would imbibe inflammable air, I endeavoured to impregnate water with it, by the same process by which I had made water imbibe fixed air; but though I found that distilled water would imbibe about one fourteenth of its bulk of inflammable air, I could not perceive that the taste of it was sensibly altered. FOOTNOTES: [4] To try this, after every explosion, which immediately follows the presenting of the flame, the mouth of the phial should be closed (I generally do it with a finger of the hand in which I hold the phial) for otherwise the inflammable air will continue burning, though invisibly in the day time, till the whole be consumed. SECTION IV. _Of AIR infected with ANIMAL RESPIRATION, or PUTREFACTION._ That candles will burn only a certain time, in a given quantity of air is a fact not better known, than it is that animals can live only a certain time in it; but the cause of the death of the animal is not better known than that of the extinction of flame in the same circumstances; and when once any quantity of air has been rendered noxious by animals breathing in it as long as they could, I do not know that any methods have been discovered of rendering it fit for breathing again. It is evident, however, that there must be some provision in nature for this purpose, as well as for that of rendering the air fit for sustaining flame; for without it the whole mass of the atmosphere would, in time, become unfit for the purpose of animal life; and yet there is no reason to think that it is, at present, at all less fit for respiration than it has ever been. I flatter myself, however, that I have hit upon two of the methods employed by nature for this great purpose. How many others there may be, I cannot tell.