VI. _Minute examination of the decomposition of Nitrate of Ammoniac._ Into a retort weighing 413,75 grains, and of the capacity of 7,5 cubic inches, 100 grains of pulverised compact nitrate of ammoniac were introduced. To the neck of this retort was adapted a recipient, weighing 711 grains, tubulated for the purpose of communicating with the mercurial airholder, and of the capacity of 8,3 cubic inches. Temperature being 50° and atmospheric pressure 30,6, the recipient was inserted into a vessel of cold water, and made to communicate with the airholder. The heat of a spirit lamp was then slowly applied to the retort: the salt quickly began to decompose, and to liquify. The temperature was so regulated, as to keep up an equable and slow decomposition. During this decomposition, no luminous appearance was perceived in the retort; the gas that came into the airholder was very little clouded, and much water condensed in the receiver. After the process was finished, the communication between the mercurial airholder and the recipient was preserved till the common temperature was restored to the retort. The volume of the gas in the cylinder was 85,5 cubic inches. The absolute quantity of nitrous oxide in those 85,5 cubic inches, it was difficult to ascertain with great nicety, on account of the common air previously contained in the vessels. 45 measures of it, exposed to well boiled water, diminished by agitation to 8 measures. So that reasoning from the quantity of air, which should have been expelled from the water by the nitrous oxide, we may conclude that the 85,5 cubic inches were nearly pure. The retort now weighed 419,25 grains, consequently 5,5 grains of salt remained in it. This salt was chiefly collected about the lower part of the neck, and contained rather more water than the compact nitrate, as in some places it was crystalised. The recipient with the fluid it contained, weighed 759 grains. It had consequently gained in weight 48 grains. Now the 85,5 cubic inches of nitrous oxide produced, weigh about 42,5 grains; and this added to 48 and 5,5, = 96 grains; so that about 4 grains of salt and fluid were lost, probably by being carried over and deposited by the gas.[86] [86] This was actually the case; for on examining the conducting tube the day after the experiment, some minute crystals of prismatic nitrate of ammoniac were perceived in it. As much of the fluid as could be taken out of the recipient, weighed 46 grains, and held in solution much nitrate of ammoniac with superabundance of acid. This acid required for its saturation, 3⅛ of carbonate of ammoniac (containing, as well as I could guess), about 20 per cent alkali.