The first of the foregoing analyses supplies an instance where a loss of carbon was decisive. In the second and third, as in the whole series of successful analyses, the carbonic acid gas may be taken as exactly equal to half the condensation due to the formation of water, and as containing half the oxygen which was required for complete combustion. The quantity of marsh gas present was equal to half the oxygen required for its complete com- bustion, to half the condensation due to generated water, and to the volume of carbonic acid gas which was produced. As this was a uniform result in all the samples, it is manifest that the constitution of the inflammable principle of fire-damp is identical with that of marsh gas or light carburetted hydrogen. The proportion of carbon and hydrogen indicated by analysis, sufficiently demonstrate the absence of such gases as hydrogen, carbonic oxide, and olefiant gas. Their absence, however, was proved by other methods. A portion of fire-damp was mixed in a tube with chlorine of known purity, and the mixture kept for a quarter of an hour in a dark place, when the chlorine was absorbed by milk of lime : the original quantity of firedamp was always recovered, except a slight loss due to the mere washing to absorb the chlorine. The absence of olefiant and carbonic oxide gases was also proved by means of spongy platinum. In 1824, soon after the curious action of spongy platinum in causing the combination of oxygen and hydrogen gases was made known by Doebereiner, both Dr. Henry and myself pointed out the obstacles to that action, occasioned by carbonic oxide, olefiant gas, and some other gases. — (Philosophical Transactions, and Edinbro' Philosophical Journal for 1824.)— And Dr. Henry at the same time shewed that marsh gas differs remarkably in this respect from carbonic oxide and olefiant gases, as it offers scarcely any impediment to the action of platinum. Agreeably to those researches, it follows that if fire-damp contained merely marsh gas, oxygen, and nitrogen, spongy platinum introduced at common temperatures, or even heated to 300° F., would not produce any sensible effect; and that if a small quantity of an explosive mixture,* made with 1 measure of oxygen, and 2 measures of hydrogen gases, were added to the fire-damp, spongy platinum should cause
- By the expression " explosive mixture," I hereafter mean a mixture made with 1
measure of oxygen, and 2 measures of hydrogen gases. VOL. II. 4 A
