The specific weight or weights of an equal measure of air and other gases is determined on the same principle as liquids, although a different apparatus is required. A light capped glass globe, with stop-cock, from 50 to 100 cubic inches capacity, is weighed full of air, then exhausted by an air-pump, and weighed empty, the loss being taken as the weight of its volume of air; these figures are carefully noted, because _air_ instead of _water_ is the standard of comparison for all gases. When the specific gravity of any other gas is to be taken, the glass globe is again exhausted, and screwed on to a gas jar provided with a proper stop-cock, in which the gas is contained; and when perfect accuracy is required, the gas must be dried by passing it over some asbestos moistened with oil of vitriol, and contained in a glass tube, and the gas jar should stand in a mercurial trough. (Fig. 70.) The stop-cocks are gradually turned, and the gas admitted to the exhausted globe from the gas jar; when full, the cocks are turned off, the globe unscrewed, and again weighed, and by the common rule of proportion, as the weight of the air first found is to the weight of the gas, so is unity (1.000, the density of air) to a number which expresses the density of the gas required. If oxygen had been the gas tried, the number would be 1.111, being the specific gravity of that gaseous element. If chlorine, 2.470. Carbonic acid, 1.500. Hydrogen being much less than air, the number would only be 69, or decimally 0.069. [Illustration: Fig. 70. A. Glass globe to contain the gas. B. Gas jar standing in the mercurial trough, D. C. Tube containing asbestos moistened with oil of vitriol.] A very good approximation to the correct specific gravity (particularly where a number of trials have to be made with the same gas, such as [Page 56] ordinary coal gas) is obtained by suspending a light paper box, with holes at one end, on one arm of a balance, and a counterpoise on the other. The box can be made carefully, and should have a capacity equal to a half or quarter cubic foot; it is suspended with the holes downward, and is filled by blowing in the coal gas until it issues from the apertures, and can be recognised by the smell. The rule in this case would be equally simple: as the known weight of the half or quarter cubic foot of common air is to the weight of the coal gas, so is 1.000 to the number required. (Fig. 71.) [Illustration: Fig. 71. A. The balance. B. The paper box, of a known capacity. C. Gas-pipe blowing in coal-gas, the arrows showing entrance of gas and exit of the air.]