252 AN UNDISCOVERED GAS. 



of the separation of these imagined elements from each other by means 

 of dilinsion ; bnt Mr. Travers and I pointed ont that the same alteration 

 of spectrum, which was apparently produced by difiusiou, could also be 

 caused by altering the pressure of the gas in the vacuum tube, and 

 shortly after Professor Kunge acknowledged his mistake. 



These considerations, however, made it desirable to subject helium to 

 systematic diffusion in the same way as argon had been tried. The 

 experiments were carried out in the summer of 1896 by Dr. Collie and 

 myself. The result was encouraging. It was foand possible to separate 

 helium into two portions of different rates of diffusion and consequently 

 of different density by this means. The limits of separation, however, 

 were not very great. On the one hand, we obtained gas of a density 

 close on 2; and on the other, a sample of density 2.4 or thereabouts. 

 The difficulty was increased by the curious behavior, which we have 

 often had occasion to confirm, that helium x^ossesses a rate of diffusion 

 too rapid for its density. Thus, the density of the lightest portion of 

 the diffused gas, calculated from its rate of diffusion, was 1.874; but this 

 corresponds to a real density of about 2. After our paper, giving an 

 account of these experiments, had been published, a German investi- 

 gator, Herr A. Hagenbach, repeated our work and confirmed our results. 



The two samples of gas of different density differ also in other 

 properties. Different transparent substances differ in the rate at which 

 they allow light to pass .through them. Thus light travels through 

 water at a much slower rate than through air, and at a slower rate 

 through air than through hydrogen. Kow, Lord Eayleigh found that 

 helium offers less opposition to the passage of light than any other sub- 

 stance does, and the heavier of the two portions into which helium had 

 been split offered more opposition than the lighter portion. And the 

 retardation of the light, unlike what has usually been observed, was 

 nearly proportional to the densities of the samples. The spectrum of 

 these two samples did not differ in the minutest particular. Therefore 

 it did not appear quite out of the question to hazard the speculation 

 that the process of diffusion was instrumental not necessarily in sep- 

 arating two kinds of gas from each other, but actually in removing 

 light molecules of the same kind from heavy molecules. This idea is 

 not new. It had been advanced by Professor Schiitzenberger (whose 

 recent death all chemists have to deplore) and later by Mr. Crookes that 

 what we term the atomic weight of an element is a mean; that when 

 we say that the atomic weight of oxygen is 10 we merel^^ state that the 

 average atomic weight is 16; and it is not inconceivable that a certain 

 number of molecules have a weight somewhat higher than 32, while a 

 certain number have a lower weight. 



We therefore thought it necessary to test this question by direct 

 experiment with some known gas, and we chose nitrogen as a good 

 material with which to test the point. A much larger and more con- 

 venient apparatus for diffusing gases was built by Mr. Travers and 



