An experiment was therefore devised, in which a large quantity of air was made to stream slowly through a long train of stems of churchwarden tobacco-pipes, placed inside a glass tube, the latter being closed at each end, except for the entrance and exit tubes of the tobacco-pipes; in the encasing glass tube a vacuum was maintained, and the gases, passing through the walls of the pipe-stems, were pumped off and discharged. According to what has just been said, these should be the lighter gases, nitrogen and oxygen, which ought to pass through the porous stems more quickly than the supposed heavier constituent of air; while the air issuing from the end of the train of pipes should contain relatively more of the heavier constituent, and should in consequence have a greater weight than an equal volume of air. But it was obviously convenient to remove the oxygen before weighing this sample of altered air, and this was done in the usual way by passing the mixed gases over red-hot copper. It was found that such nitrogen was even heavier than ordinary atmospheric nitrogen; not much, it is true, but still consistently heavier. The denser constituent could, in fact, be concentrated by this means. The proof was therefore indubitable that the new gas existed in air as such. There is another method of proof, however, which was not left untried. Experiment showed that the solubility of the new gas in water is considerably greater than that of nitrogen, although less than that of oxygen. In 100 volumes of water at the ordinary temperature, about 1·5 volumes of nitrogen will dissolve, about 4·5 volumes of oxygen, and about 4 volumes of the new gas, to which the name finally chosen for it, “argon,” may now be applied. Now the proportion in which the constituents of a mixture of gases will dissolve in a solvent is conditioned first by their relative solubilities, and second, by their relative proportion. Thus, if air be considered to be simply a mixture of 1 volume of oxygen and 4 volumes of nitrogen, the gas extracted from water which has been shaken with air will have the composition-- Oxygen 1 x 4·5 = 4·5 volumes Nitrogen 4 x 1·5 = 6·0 " So that the proportion of oxygen to nitrogen in such a mixture of gases is considerably greater than in air: instead of being approximately 1 to 4, it is nearly 4·5 to 6. The discovery of this law concerning the composition of the gases dissolved in liquids was due to Dr. Henry, one of the biographers of Dalton.