for Production of Luminosity in Atmospheric Neon. 925 



sensitive moving-coil galvanometer which was situated in 

 a room adjoining that in which the neon spectrum was 

 under observation, the latter room being kept in darkness 

 for the better detection of faint lines. 



Experimental Results. 



For the purpose of determining the limiting electron 

 velocities required to produce particular lines in the 

 neon spectrum, it was necessary that the bombarding 

 electrons should suffer no change of velocity throughout 

 the space which was viewed by the slit of the spectroscope, 

 except such change as results from collisions with gas 

 atoms. To secure this condition, the grid and anode were 

 maintained at the same potential by connecting each to the 

 positive terminal of an insulated battery, the negative end 

 of which was connected to the negative end of the filament. 

 The maximum velocity of any of the electrons passing- 

 through the hole H is thus the sum of that due to the 

 applied potential difference and the velocity of emission of 

 the electrons from the filament. If the maximum velocity 

 of emission is determined, the velocity of the swiftest 

 electrons passing through H under any applied potential 

 difference is known. The maximum velocity of the 

 electrons passing through H, when any given line is first 

 seen in the spectroscope, thus gives an upper limit for the 

 critical electron velocity required for the excitation of 

 the particular line in question. The velocity of emission 

 was determined by finding what potential difference it was 

 necessary to apply between the filament and the grid in 

 order to prevent any electrons from the filament reaching 

 the grid. In taking this maximum velocity as the critical 

 velocity, it is assumed that the electrons having this velocity 

 are sufficiently numerous to produce the given line with 

 such intensity as to be visible. In our earlier paper 

 a method was described by means of which the velocity 

 of the electrons actually producing the radiation or ioni- 

 zation current under observation could be obtained. With 

 the present apparatus it was not possible to determine the 

 actual electron velocity in a similar manner ; and the 

 electron velocities stated as used in the present resear< h 

 are in all cases the maximum velocities in the electron 

 stream under the conditions of experiment. 



The first appearance of luminosity and the alterations in 

 its spectrum were carefully observed as the velocity ot the 

 electron stream was gradually raised from about 10 volts to 



Phil. Mag. S. G. Vol. 41. No. 246. June 1921. 3 P 



