After I had made the discovery of the _marine acid air_, which the vapour of spirit of salt may properly enough be called, and had made those experiments upon it, of which I have given an account in the former part of this work, and others which I propose to recite in this part; it occurred to me, that, by a process similar to that by which this _acid_ air is expelled from the spirit of salt, an _alkaline_ air might be expelled from substances containing volatile alkali. Accordingly I procured some volatile spirit of sal ammoniac, and having put it into a thin phial, and heated it with the flame of a candle, I presently found that a great quantity of vapour was discharged from it; and being received in a vessel of quicksilver, standing in a bason of quicksilver, it continued in the form of a transparent and permanent air, not at all condensed by cold; so that I had the same opportunity of making experiments upon it, as I had before on the acid air, being in the same favourable circumstances. With the same ease I also procured this air from _spirit of hartshorn_, and _sal volatile_ either in a fluid or solid form, i. e. from those volatile alkaline salts which are produced by the distillation of sal ammoniac with fixed alkalis. But in this case I soon found that the alkaline air I procured was not pure; for the fixed air, which entered into the composition of my materials, was expelled along with it. Also, uniting again with the alkaline air, in the glass tube through which they were conveyed, they stopped it up, and were often the means of bursting my vessels. While these experiments were new to me, I imagined that I was able to procure this air with peculiar advantage and in the greatest abundance, either from the salts in a dry state, when they were just covered with water, or in a perfectly fluid state; for, upon applying a candle to the phials in which they were contained, there was a most astonishing production of air; but having examined it, I found it to be chiefly fixed air, especially after the first or second produce from the same materials; and removing my apparatus to a trough of water and using the water instead of quicksilver, I found that it was not presently absorbed by it. This, however, appears to be an easy and elegant method of procuring fixed air, from a small quantity of materials, though there must be a mixture of alkaline air along with it; as it is by means of its combination with this principle only, that it is possible, that so much fixed air should be retained in any liquid. Water, at least, we know, cannot be made to contain much more than its own bulk of fixed air. After this disappointment, I confined myself to the use of that volatile spirit of sal ammoniac which is procured by a distillation with slaked lime, which contains no fixed air; and which seems, in a general state, to contain about as much alkaline air, as an equal quantity of spirit of salt contains of the acid air.