Rontgen Radiation in Air. 



607 



so adjusted that the leaf when earthed hangs vertically, in 

 which position (1) its sensitiveness is greatest, (2) changes 

 in the elasticity o£ the leaf due to its motion are least liable 

 to occur, (3) the charge received by the leaf from one 

 quadrant, due to natural ionization, is balanced by that pass- 

 ing from the leaf to the other quadrant. The quadrants 

 effectually damped any air-currents which arose. The move- 

 ments of the leaf were observed by a microscope of magnifi- 

 cation about 30. The capacity of the combined electroscope, 

 ionization-chamber, and balancing-chamber was 7'3 electro- 

 static units, and the deflexion per volt was 2000 divisions. 

 This gives about double the sensitiveness of a fine suspension 

 Dolezalek electrometer with capacity about 80 and deflexion 

 per volt 10,000, while the vibrations and oscillations ex- 

 perienced with that instrument are not existent in the 

 electroscope. 



A large coil, one of Spottiswoode's, capable of giving a 

 30-inch spark, was employed to excite a Rontgen bulb, and 

 currents varying between one and three amperes were used, 

 with an Apps hammer-break. The arrangement of apparatus 

 was practically that used by Barkla*, and is shown in fig. 3. 



Fif?. 3. 



B 

 ► TO flECTROSCOPE 



*+eoo Vol rs. 



/) 



In order to compensate for the leak due to natural ionization 

 in the ionization-chamber, a balancing-chamber B was set up 

 consisting of a brass cylinder C sliding inside another. The 

 central electrode passed back into this chamber, which was 

 kept at a potential of — 200 volts, while the ionization chamber 

 was kept at -f 200. Thus the leak to the electrode was in 



* Barkla, Phil. Mag-. May 1904, p. 544. 



2 S 2 



