Radiation and Ionization of Helium. 755 



section and 2'2 cm. in diameter. The lower end of the side 

 tube forms one limb of a glass U-tube containing carbon 

 which can be cooled in liquid air, and through which the 

 helium gas is admitted to the apparatus. An opening nearer 

 the top of the main tube leads through a second carbon 

 purifying tube to the pumps and pressure-gauge, and to a 

 long vertical tube some 4 cm. wide containing mercury the 

 level of which can be adjusted so as to vary the pressure of 

 the gas in the apparatus. Mercury vapour was prevented 

 from passing over into the essential parts of the apparatus by 

 means of a U-tube on the pump side of a wide- bore stop- 

 cock, which was only opened when the U-tube was cooled in 

 liquid air or in solid carbon dioxide. All the precautions 

 described in our earlier paper were again taken to prevent 

 the evolution of gas from the metal and glass parts of the 

 apparatus during the experiments, and the "baking-out" 

 treatment was continued for more than a week before 

 experiments were begun. 



In order to be sure that the helium used in the experiments 

 did not contain hydrogen, copper spirals coated with oxide 

 were electrically heated to about 500° C. in the gas for some 

 500 hours. This process was performed by maintaining an 

 excess of electrolytic oxygen in the gas and by replacing 

 each spiral when it burned through. The helium was then 

 purified from excess of oxygen and from other gases by 

 standing for several hours over charcoal cooled in liquid air. 

 It was stored in a orlass olobe from which it could bo allowed 

 to pass into the apparatus by opening two stop-cocks, one 

 at each end of a fine capillary tube. The gas passed very 

 slowly through this tube and entered the apparatus at the 

 bottom of the side tube, through the U-tube containing 

 carbon cooled in liquid air. In most of the experiments the 

 pure helium was steadily streaming through the ionization 

 tubes during the observations, so that the pressure rose slowly; 

 but, with an initial pressure of 03 mm. or more, the increase 

 of pressure during the course of a single experiment was 

 not important owing to the large volume included in the 

 pumps &c. 



The apparatus represented in fig. 1 was designed so that 

 radiation pro luced in the space between the gauzes K and L 

 by the bombardment of helium atoms by electrons from 

 the side filament could not illuminate the platinum-covered 

 cylinder E, either directly or by retlexion from the glass or 

 metal parts of the apparatus. This precaution was very 

 necessary, since the apparatus was to be used to tost whether 



