Wanting, however, to procure this air in greater quantities, and this method being rather expensive, it occurred to me, that alkaline air might, probably, be procured, with the most ease and convenience, from the original materials, mixed in the same proportions that chemists had found by experience to answer the best for the production of the volatile spirit of sal ammoniac. Accordingly I mixed one fourth of pounded sal ammoniac, with three fourths of slaked lime; and filling a phial with the mixture, I presently found it completely answered my purpose. The heat of a candle expelled from this mixture a prodigious quantity of alkaline air; and the same materials (as much as filled an ounce phial) would serve me a considerable time, without changing; especially when, instead of a glass phial, I made use of a small iron tube, which I find much more convenient for the purpose. As water soon begins to rise in this process, it is necessary, if the air is intended to be conveyed perfectly _dry_ into the vessel of quicksilver, to have a small vessel in which this water (which is the common volatile spirit of sal ammoniac) may be received. This small vessel must be interposed between the vessel which contains the materials for the generation of the air, and that in which it is to be received, as _d_ fig. 8. This _alkaline_ air being perfectly analogous to the _acid_ air, I was naturally led to investigate the properties of it in the same manner, and nearly in the same order. From this analogy I concluded, as I presently found to be the fact, that this alkaline air would be readily imbibed by water, and, by its union with it, would form a volatile spirit of sal ammoniac. And as the water, when admitted to the air in this manner, confined by quicksilver, has an opportunity of fully saturating itself with the alkaline vapour, it is made prodigiously stronger than any volatile spirit of sal ammoniac that I have ever seen; and I believe stronger than it can be made in the common way. In order to ascertain what addition, with respect to quantity and weight, water would acquire by being saturated with alkaline air, I put 1-1/4 grains of rain-water into a small glass tube, closed at one end with cement, and open at the other, the column of water measuring 7/10 of an inch; and having introduced it through the quicksilver into a vessel containing alkaline air, observed that it absorbed 7/8 of an ounce measure of the _air_, and had then gained about half a grain in weight, and was increased to 8-1/2 tenths of an inch in length. I did not make a second experiment of this kind, and therefore will not answer for the exactness of these proportions in future trials. What I did sufficiently answered my purpose, in a general view of the subject.