﻿Low-Speed Positive Yl-Ions in Hydrogen, 



817 



nickel-foil screen partly round the filament, but this was found 

 unsatisfactory, apparently owing to the diffusion of (he ions. 

 With the negative end of the filament connected directly to 

 the gauze G x so that x = 0, it was found that many thermo- 

 electrons shot out into the space between the gauzes. To 

 prevent this, Gi was made negative with respect to the nega- 

 tive end of the filament, a retarding P.D. of x volts being 

 applied. With #=3 volts as a minimum, this effect could 

 no longer be detected. 



The ionization produced was measured by Lenard's method, 

 using the gauge G 2 as a collecting-plate instead of D to 

 shorten the distance the positive ions had to travel before 

 ionizing by collision. A retarding field of 50 volts for 

 positive ions was applied between Gi and G 2 . The arcing 

 potential V was kept constant, and x increased to increase 

 the total accelerating P.D. of x+ v volts. The ionization 

 current plotted against the total accelerating voltage is 

 shown by a typical curve in fig. 3. The readings were taken 



Vi 



* 300 



250 



200 



150 



Fig. 3. 

 Retarding Field. = 50 volts. p= "008 





18 23 28 33 38 



Total Accelerating Voltage (x+Vvolts.) 



43 



in rapid succession to avoid any alteration in the ionizing 

 current. Values of the accelerating voltage below 19 volts 

 were not possible, and readings to repeat were difficult to 

 obtain because of the variation in the ionizing current which 

 depended upon three variables^ the temperature of the fila- 

 ment, the pressure of the gas, and also to some degree on 

 the value of .r, the " drawing-out P.D/' A second method 

 was tried to detect ionization by applying a cross-field of 



