I 



I 

 1 

 I 



I 



1017] on Industrial Applications of Electrons 177 



diminished. If, then, the electric waves were cut up into dots and 

 dashes, the current would experience corresponding variations in the 

 current ; this method was used years ago by Professor Fleming to 

 detect radio-telegraphic messages. 



Of late years great attention had been devoted to the improve- 

 ment of detectors of this type. It is desirable to distinguish twa 

 cases : — 



1. When the vacuum in the valve is so good that the effect of 

 the residual gas can be neglected, and the discharge is purely 

 electronic. 



2. When the conization in the residual gas is appreciable. 



In the purely electronic discharge the relation between the current 

 and voltage is very interesting, and has been investigated with great 

 success by Mr. Langmuir. The results are represented hj the curves, 

 in Fig. 2 ; the different curves corresponding to 

 different temperatures of the cathode. 



The current increases with the voltage up to a 

 certain point and there becomes constant and has a 

 value which depends only on the temperature of the 

 cathode and not upon the voltage ; the higher the 

 temperature, the greater the range of voltage where Fig. 2. 

 the current increases with the voltage, and the higher 

 also the limiting current. The sensitiveness of the arrangement 

 depends upon the steepness of the curve just before they become 

 flat. This steepness can be increased by the admission of a little 

 gas when we get valves of the second type. The action of the ga& 

 may be explained in the following way. The cathode C emits elec- 

 trons which are driven towards the anode A, some of these electrons 

 accumulate, however, on or near the grid, their repulsion on the 

 electrons coming from C would tend to neutralize the attraction of 

 the anode and the emission of the electrons would be retarded. 

 If, however, a little gas is admitted into the tube, a few positive 

 ions U'ould be produced between the anode A and the grid G, they 

 would move slowly up to the cathode through the perforations in the 

 grid, and would neutralise the repulsion exerted by the electrons ; 

 the effect would be that more electrons would come from the 

 cathode and the curve representing the current (shown by the dotted 

 line in Fig. 3) would rise more steeply. 



I have investigated the theory of this arrangement, and find that 

 the valve is most sensitive when — 



potential between grid and cathode _ 4 m 

 potential between anode and cathode n- M 



where n is the number of positive ions produced by one electron, in 

 the mass of an electron, M that of a positive ion. As m/M. is verj^ 

 Vol. XXII. (Xo. Ill) n 



