"936 Prof. Horton and Miss Davies on Electron Velocities 



The extension of the glow right up to the grid when it 

 appeared at 25 volts in this experiment, even though the 

 velocity of the electron stream on reaching the grid was not 

 enough to produce ionization, confirms in a striking manner 

 the view that luminosity is produced at the recombination of 

 positive ions and electrons. The positive ions are driven 

 upwards by the electric field between the grid and anode 

 and become neutralized, by electrons from the filament, 

 in positions nearer to the grid than those in which they are 

 formed. As the potential difference between the grid and 

 anode is reduced, the positive ions being now produced 

 nearer the anode recombine with electrons before they get 

 near the grid, and so the luminosity no longer reaches up to 

 that electrode. ISimilar evidence was obtained from experi- 

 ments at lower pressures. In all of these the potential 

 difference between the filament and the grid was maintained 

 at a value well below the minimum ionization velocity. The 

 luminosity made its first appearance at the anode, and, 

 as the total potential difference between the filament and 

 the anode w T as increased, the lumnous column gradually 

 became longer and ultimately reached the grid. In one 

 instance the potential difference between the filament and 

 the grid was only 4 volts when a luminous column extending 

 right up to the grid was obtained. Thus the luminosity at 

 the grid could not be due to ionization, but must have 

 resulted from the recombination of the positive ions with 

 electrons from the filament. 



Discussion of Results and Conclusion. 



The experiments described in this paper lead to the 

 -conclusion that there is no luminosity produced in neon 

 as a result of the first type of ionization,- shown in our 

 earlier paper to begin at 16*7 volts. Since it is generally 

 accepted that when ionization and subsequent recombination 

 occur, a complete line spectrum corresponding to the parti- 

 cular ionization in question is produced, this absence of 

 luminosity in the case of the 16 7 volts ionization in neon 

 suggests that the corresponding spectrum may consist 

 entirely of lines outside the visible region. 



It has also been shown in the course of the paper that 

 the lines contained in the spectrum of the glow when first 

 produced, at 20 volts, are all of the (1*5, s — m, p), or 

 principal series, type according to Pascheir's classification ; 

 while the lines which do not appear until the ionizing 

 Telocity 22-8 volts has been reached, are all of the 



