[shaver] spectra of ARGON 139 



In Plate II are shown reproductions of photographs taken with 

 accelerating voltages varying from 31 to 105 volts. Up to 30 volts 

 potential on the grid only lines belonging to the red argon spectrum 

 appeared. However at 31 volts some faint lines belonging to the 

 blue or enhanced spectrum became evident (see Plate II(o)). As the 

 voltage was increased these lines became more intense and others 

 appeared as shown in Plate II (b), which was taken with a grid potential 

 of 37 volts. Plate 11(c) shows the spectrum of the arc when the 

 accelerating potential was 40 volts, in which the number and intensities 

 of the lines belonging to the blue spectrum were still further increased. 

 These results do not exactly agree with those of Déjardin, who found 

 that the lines of the enhanced spectrum did not appear until an 

 accelerating potential of 34 volts was used. This discrepancy may 

 be accounted for by the fact that in the present experiments the 

 temperature of the filament may have been slightly greater than that 

 used by Déjardin, which would mean that the bombarding electrons 

 had a greater initial velocity and therefore required a smaller accelerat- 

 ing field to give them the same energ}^ 



Plate 11(d) shows the result with a grid potential of 54 volts. 

 It is to be noted that, according to the Bohr theory, the potential 

 necessary to give an impacting electron the required energy to remove 

 both electrons from the helium atom is 54 volts. Hence the energy 

 required to remove two electrons from any other neutral atom of 

 higher atomic number than helium, is less than that corresponding 

 to 54 volts owing to the repulsive force exerted by the remaining 

 electrons. Thus with a grid potential of 54 volts all the lines of the 

 enhanced spectrum of argon should be present. This would appear 

 to be the case, as apart from changes in intensities, the spectrum was 

 unchanged when a potential of 105 volts was used (see Plate 11(e)). 

 Some of the lines in the ultra-violet between the wave-lengths 2900 



o o 



A.U. and 2300 A.U. are very faint and do not appear in the repro- 

 duction so that the agreement between the arc spectrum at 54 volts 

 and the blue spark spectrum shown in Plate II(/) was much better 

 than the reproductions would indicate. The spectrum in Plate 11(g) 

 is that of the mercury arc which, as before, was used for wave- 

 length comparison purposes. It will be noted that the mercury line 

 of wave-length 2536 A.U. appears on several of the photographs, 

 which was due to a trace of mercury vapour in the lamp. 



V. Conclusions 



The conclusions to be drawn from these results are necessarily 

 somewhat indefinite. It seems clear, however, that with a speed 



