146 Prof. J. Zeleny on the Ratio of the Velocities of the 



can enter M which cross AB in Ax, as those crossing in Bx 

 reach the plate P. 



The point x moves towards A as F is increased. When 

 this state is reached, the number of ions crossing AB is, in 

 our case, still increasing with F, and so the amount obtained 

 is dependent upon the two opposing actions, i. e. the increase 

 in the number crossing AB, and the decrease of Ax, the area 

 from which ions can reach the slit. This correponds to the 

 maximum part B of the curves. When F becomes so large 

 that a saturation-current is obtained, the number of ions 

 crossing AB is constant, and as F increases, the number that 

 reach M is proportional to the decreasing Ax. Since 

 F. A<r = a constant, the curve representing F and the number 

 of ions now approaches an hyperbola, as is shown by the part 

 G of the curves in the figure. Finally, for high values of F, 

 no matter how many ions enter the blast, none are carried 

 out. For the sake of brevity, the predominating action has 

 alone been outlined, and various factors which take part have 

 not been considered, but from what has already been said the 

 effect of these can be followed. 



For the same velocity of the blast, the nature of the curves 

 obtained depends also upon the distance between the plates. 

 When these were moved nearer together, the saturation - 

 current was obtained while all of the ions crossing AB were 

 still being blown into M. In that case a curve similar to the 

 above has a maximum which turns sharply, and is not so 

 blunt as those in the figure. 



In other cases increased ionization at the metal surface 

 plays an important part, and this is especially so when the 

 plates are close together, so that the surface-ionization is 

 a large part of the total ionization between the plates. Here 

 complex curves are obtained, due to the opposing and at 

 times even predominating influence of this surface-ioniza- 

 tion. With the rays striking the plates which were close 

 together, the sign of the charge blown out was the same 

 as that of the plate along which the current of air passed, 

 or the reverse of what was obtained above, for here nearly all 

 of the ionization was near the metal surfaces. 



For a somewhat larger distance the surface-ionization pre- 

 dominated for the smaller voltages only, and as more ions 

 were drawn from the space between the plates by the higher 

 voltages the sign of the charge obtained was reversed. 



These facts show in what a radical manner the electrifica- 

 ti'm, obtained by means of a current of air, is dependent 

 upon the velocity of the blast, upon the velocity and distribu- 

 tion of the ions, and upon the arrangement of the apparatus. 



