This gas extinguished flame, gave no diminution with oxygene, and the slightest possible with nitrous gas. When an electric spark was passed through it, mingled with oxygene; no inflammation, or _perceptible_ diminution took place.[84] We may consequently conclude that it was nitrogene, mingled with a minute portion of common air, expelled from the water. [84] The gas was examined by those tests in order to prove that no water had been decomposed. The charcoal was diminished in bulk to one half nearly, but the loss of weight could not be ascertained, as its pores were filled with mercury. Now 5 cubic inches of nitrous oxide were absorbed by the water, consequently 5 were decompounded by the charcoal; and these produced 5,1 cubic inches of nitrogene; and by giving their oxygene to the charcoal, apparently 2,4 of carbonic acid. But 5 cubic inches of nitrous oxide weigh 2,5 grains, and 5,1 cubic inches of nitrogene 1,55; then 2,5-1,55 =,95. So that reasoning from the relative specific gravities of nitrogene and nitrous oxide, 2,5 grains of the last are composed of 1,55 nitrogene, and,95 oxygene. But from many experiments made on the specific gravity of carbonic acid, in August, 1799, I concluded that 100 cubic inches of it weighed 47,5 grains, thermometer being 60,1°, and barometer 29,5. Consequently, making the necessary corrections, 2,4 cubic inches of it weigh 1,14 grains; and on Lavoisier’s and Guyton’s[85] estimation of its composition, these 1,13 grains contain 8,2 of oxygene. [85] See the curious paper of this excellent philosopher, on the combustion of the diamond, in which he proves that charcoal is, in fact, oxide of diamond. Annales de Chimie, xxxi. So that, drawing conclusions from the quantity of carbonic acid formed in this experiment, 2,5 grains of nitrous oxide will be composed of,82 oxygene, and 1,68 nitrogene. The difference between these estimations is considerable, and yet not more than might have been expected, if we consider the probable sources of error in the experiment. 1. It is likely that variable minute quantities of hydrogene remain combined with charcoal, even after it has been long exposed to a red heat. 2. It is probable that the nitrogene and carbonic acid produced were capable of dissolving more water than that held in solution by the nitrous oxide; and if so, they were more condensed than if saturated with moisture, and hence the quantity of carbonic acid under-rated. We may consequently suppose the estimation founded on the quantity of nitrogene evolved, most correct; and making a small allowance for the difference, conclude, that 100 grains of nitrous oxide are composed of about 37 oxygene, and 63 nitrogene; existing in a much more condensed state than when in their simple forms. The tolerable accuracy of this statement will be hereafter demonstrated by a number of experiments on the combustion of different bodies in nitrous oxide, detailed in Research II.