A porcelain tube was provided, open at both ends, and well glazed inside and outside, its diameter being about,5 inches. To one end of this, a glass tube was affixed, curved for the purpose of communicating with the water apparatus. With the other end a glass retort was accurately connected, containing a mixture of perfectly caustic slacked lime, and muriate of ammoniac. The water in the apparatus for receiving the gases had been previously boiled, to expel the air it might contain, and during the experiment was yet warm. When the tube had been reddened in a furnace adapted to the purpose, the flame of a spirit lamp was applied to the bottom of the retort. A great quantity of gas was collected in the water apparatus; of this the first portions were rejected, and the last transferred to the mercurial trough. A small quantity examined, did not at all diminish with nitrous gas, and burnt with a lambent white flame, in contact with common air. 2¾ of this gas, equal to 110 grain measures, were fired with 2, equal to 80, of oxygene, in a detonating tube, by the electric spark. They were reduced to 2¼, or 90. On introducing to the remainder a solution of strontian, it became slightly clouded on the top, and an absorption of some grain measures took place. It was evident, then, that in this experiment, charcoal[61] had been somehow present in the tube; which being dissolved by the nascent hydrogene, had rendered it slightly carbonated, and in consequence made the results inconclusive. [61] Though the tube had never been used, and was apparently clean and dry on the inside, it must have contained something in the form of dust, capable of furnishing either hydrocarbonate, or charcoal. A tube of thick green glass carefully made clean, was now employed, inclosed in the porcelain tube. Every other precaution was taken to prevent the existence of sources of error, and the experiment conducted as before. 140 grain measures of the gas produced, fired with 120 of oxygene, left, in two experiments, nearly 110. Solution of strontian placed in contact with the residuum, did not become clouded, and no absorption was perceived. Now 150 measures of gas were destroyed, and if we take Lavoisier’s and Meusnier’s estimation of the composition of water, and suppose the weight of oxygene to be 35 grains, and that of hydrogene 2,6 the hundred cubic inches; the oxygene employed will be to the hydrogene as 243 to 576. Put _x_ for the oxygene, and _y_ for the hydrogene. Then _x_ + _y_ = 150 _x_ : _y_ :: 243 : 576 243_y_ _x_ = —————— 576 819_y_ = 86400 _y_ = 105 _x_ = 45 And 140 - 105 = 35 Consequently, the nitrogene in ammoniac is to the hydrogene as 35: 105 in volume: and 13,3 grains of ammoniac are composed of 10,6 nitrogene, (supposing that 100 cubic inches weigh 30,45 grains) and 2,7 hydrogene.