Ront gen Radiation from Gases and Vapours. 657 



with an aperture of the exact size and shape o£ the window, 

 was made, and screw-holes made through it, and in the 

 side of the box so that the frame could be screwed firmly 

 into position. The frame and the box were made sufficiently 

 hot to melt sealing-wax which was applied to each. The 

 aluminium-foil of the window was then stretched between 

 the two, and the whole was screwed down firmly into place, 

 while the sealing-wax was still quite fluid. In this way a 

 perfectly air-tight joint could easily be made, and as the 

 aluminium-foil was tightly gripped along the whole length 

 of its four edges, there was no strain on the sealing-wax, and 

 the tendency to crack away was therefore eliminated. More- 

 over, as the foil was tightly held on all sides it was possible 

 to use much thinner foil than would otherwise have been the 

 case. 



The ionization-chambers B, B were of the ordinary cylin- 

 drical type, with a central wire electrode insulated by sulphur 

 and protected from direct leakage from the charged outer 

 cylinder by an earthed guard-ring. The lower end was closed 

 with thin aluminium-foil. 



The electroscope was one of the Wilson inclined type, and 

 considerable trouble was taken to render it as sensitive as 

 possible. By a very careful adjustment of the position of the 

 gold-leaf suspension, and the potential of the charged plate, 

 it was found possible to get deflexions of 10 microscope eye- 

 piece divisions for l/50th of a volt. The capacity of the gold- 

 leaf system with the electrodes was 7 cms. ; and using the 

 null method, a change of 1 division in 10 minutes could be 

 easily detected. This would correspond to a change in the 

 current through the ionization-chamber of about 3xl0~ 17 

 amperes. 



This sensitiveness enabled readings to be made with con- 

 siderable accuracy, and it was found in practice that an 

 alteration of two or three millimetres of mercury in the 

 pressure of the air in the standard chamber, was sufficient to 

 cause a perceptible change in the rate of movement of the 

 electroscope-leaf. This amounted, in general, to an accuracy 

 of about 1 per cent. 



The electroscope was operated from a distance with a key 

 of the usual description, and the wires leading from it to the 

 electrodes were shielded in a carefully earthed zinc box. The 

 X-ray bulb was enclosed in a thick lead box, the rays emerging 

 through a suitable aperture. The ionization-chambers and 

 the electroscope were carefully shielded from the direct rays 

 of the bulb, and also from any stray radiation, by lead screens 

 of suitable shape. Some of these are indicated in figure 2. 



