2 Hemsalech, spectrum of the Gloiv Discharge. 



of about 0'04 henry in the discharge circuit. Copper and 

 silver electrodes are particularly effective in producing the 

 glow, especially when they are placed very near together ; 

 the glow is then seen to extend to some distance up the 

 sides of the electrodes. 



But under ordinary conditions the glow is a very 

 unstable phenomenon ; sometimes it appears at one 

 electrode, sometimes at the other ; at other times again it 

 will appear at both simultaneously, and then again not at 

 all ; hence it was quite impossible to produce the glow at 

 will. 



Now, it was found that the glow was always accom- 

 panied by an enormous heating of the electrode and it 

 was thought possible that the heating might really be the 

 cause of this glow. 



An experiment made on these lines proved at once 

 successful : by simply heating the electrodes the glow 

 could be obtained on either of them at will. Also other 

 metals with which the glow had not been observed before 

 could be made to produce it. The heating may be effected 

 either by means of a Bunsen burner or electrically. 



The glow produced in this way is very brilliant and 

 sometimes extends to a distance of from 5 to 8 mm. up 

 the electrode, the latter being 2 or 3 mm. in diameter. 

 The glow is seen to be entirely confined to the surface 

 of the metal. 



Analysed by means of an oscillating mirror the glow 

 is seen to appear alternately first at one and then at the 

 other electrode in accordance with the alternations of 

 the oscillations in the discharge. Also the sharp images 

 of the glow as seen in the mirror reveal the fact that 

 the phenomenon must be almost momentary. 



The spectrum of the ordinary coil discharge in air 

 between electrodes about i cm apart is reproduced in 



