596 Professor Sir J. J. Thomson [Jan. 17. 



when heated, came to the conclusion that, for practical purposes, a 

 piece of iron must be regaided as an inexhaustible reservoir of gas. 

 There are some interesting features about the emission of gas from 

 a heated solid. If the body is kept for a long time in a vacuum at a 

 high temperature, the emission of gas becomes too small to be detected: 

 if after this treatment the temperature is raised considerably, 

 there will be a further copious emission of gas, which again diminishes 

 as the heating continues. After it has fallen to zero, all that is neces- 

 sary is to raise the temperature again and you will get a fresh supply 

 of gas ; and as far as my experience goes, after you have got all the 

 gas you can out of the solid by heating it, you have only to expose 

 it to cathode rays to get a fresh outburst. This effect of increased 

 temperature in renewing the stream of gas from the solid seems to 

 me to be too large to be accounted for merely by an increase in the 

 rate of diffusion of the absorbed gas from the interior to the surface ; 

 it seems to be more analogous to the case of the emission of the 

 water of crystallization from some salts. There are some salts, 

 for example, copper sulphate, which when heated lose theii" water 

 of crystallization in stages ; thus, if the temperature is raised to a 

 certain value, some of the water of crystallization comes off, but the 

 rest remains fixed, and you may keep the salt at this temperature 

 for ever without getting rid of all the water of crystallization ; on 

 raising the temperature, however, fresh water of crystallization is 

 given off. 



Something of this kind seems to take place in the case of gases 

 absorbed in metals, and there ceem to be indications that there 

 is some kind of chemical combination between the gas and the 

 metal. This absorbed gas may influence the behaviour of the sub- 

 stance. For example, an ordinary carbon filament gives off, when 

 raised to a white heat, large quantities of negatively electrified cor- 

 puscles ; but Pring and Parker * have shown that when great precau- 

 tions are taken to get rid of the absorbed gas, the emission of these 

 corpuscles falls to less than one-millionth of their previous vakie. 

 It is in the gases given off by certain metals when they are bombarded 

 by cathode rays that I have found an unf aiUng source of the substance, 

 which I shall denote by Xg, giving the Hue corresponding to the 

 atomic weight 3. The arrangement I have used for investigating 

 the presence of this gas is shown in Fig. 4. A is a vessel communi- 

 cating with the bulb B in which the positive rays are produced by 

 two tul:)es, one of which is a very fine capillary tube, while the other 

 one is five or six millimetres in diameter ; taps are inserted so that 

 one or both of these vessels can be closed, and the vessels A and B 

 isolated from each other, A is provided with a curved cathode such 



Phil. Mag., xxiii. p. 192. 



