At this stage the two discoverers joined forces, and letters passed almost daily between them, describing the results of experiments which one or other had made. And just prior to the meeting of the British Association at Oxford in August 1895, it was decided that the proof of the existence of a new constituent gas in air was sufficiently clear to render it advisable to make to the Association a short announcement of the discovery. The statement was received with surprise and interest; chemists were naturally somewhat incredulous that air, a substance of which the composition had been so long and so carefully studied, should yield anything new. One of the audience inquired whether the name of this new substance had been discovered; as a matter of fact it was then under consideration. But it was still conceivable, although improbable, that the new gas was being produced by the very processes designed for its separation, and attention was first turned to devising a complete proof of its actual presence in air. Now it is known that the rates of diffusion of gases through a narrow opening, or through a number of minute holes, such as exist in a pipe of porous clay, _e.g._ a tobacco-pipe stem, are in inverse proportion to the square roots of the densities of the gases. Oxygen is, in round numbers, sixteen times as dense as hydrogen; the square roots of 16 and 1 being 4 and 1, it was found by Graham, who first carefully investigated this subject, that four times as much hydrogen would pass through a porous diaphragm, in a given time, as oxygen. The _compound_ of hydrogen and oxygen, however, in the state of gas, _viz._ steam, is not separated by such a process into its constituents; it diffuses as such, and since it is nine times as dense as hydrogen, the relative rates of diffusion of steam and hydrogen are as 1: √9, or as 1 to 3; that is, for every 3 parts of hydrogen passing through such a septum, 1 part of steam would pass in the same time.