[ 113 ] 



XIV. The rreparat'ion and Properties of an Intensely Radio- 

 active Gas frotn Metallic Mercuri/. Bij Hon. R. J. Strutt, 

 Fellow of Trinity College, Cambridge ^. 



SOME time back f I described a series of experiments on 

 the leakage of positive electricity from hot metals. 

 These experiments were conducted at a pressure of 1 cm. of 

 mercury, in order to avoid the leakage of electricity which 

 always takes place in a vessel containing air even at ordinary 

 temperatures, and which is now known to be due to a feeble 

 radioactivity of the walls. On some occasions it was found 

 that after heating the apparatus for some hours, allowing it 

 to cool, and admitting air to atmospheric pressure, the leakage 

 became very much greater than the normal. This effect dis- 

 appeared when the air in the vessel was replaced by fresh 

 air ; so that it appeared clear that it must be due to a radio- 

 active gas in the vessel. It was concluded (too hastily, as 

 the event has proved) that some traces of radium salt had 

 accidently got into the vessel, and that the leakage was due 

 to the radium emanation. 



In a recent paper % I showed that different samples of the 

 same material possessed different radioactivities. In order 

 to clear up this pointy I prepared some cylinders of copper §. 

 These were annealed and put in the testing apparatus. It 

 was found that they gave a considerable rate of leak, which 

 was much diminished by washing out with fresh air. The 

 explanation seemed to be that a radioactive gas was liberated 

 from the copper by annealing, and that it gradually diffused 

 from the inner parts of the metal sheet to the air contained 

 in the cylinder. This conclusion was in agreement with the 

 observations of Prof. McLennan and Mr. Burton ||, who 

 concluded that ordinary metals gave out a slight radioactive 

 emanation, similar to that due to radium and thorium. 



I next tried drawing air through a glass tube heated just 

 below redness, and containiug scrap-copper. This air was 

 found to give a leakage three or four times the normal. 



It seemed likely to be better to bubble air through a fluid 



* Communicated by the Author. 



t Phil. Macr. July 1902. 



X Nature, Feb. 19, 1903; Phil. Mag. June 1903. 



§ It was intended to try the effect of purifying the copper electro- 

 lyticaUy, but my attention has for the time been diverted from this 

 experiment. 



II Phil. Mag. June 1903. 



Phil. Mag. S. 6. Vol. 6. No. 31. Jidy 1903. I 



