742 Prof. R. W. "Wood on Hydrogen Spectra 



appear. 'PI. XXI. h'g. 4, shows two strong silicon lines near 

 the middle of the X group. 



This circumstance makes me think that the X group may 

 be the characteristic spectrum of an unstable compound of 

 silicon and hydrogen. There appears to be additional evidence 

 in favour of this hypothesis, for it was found that if the 

 centre of the tube was packed with cotton soaked with 

 liquid air, the X group disappeared. 



With the tube in the black stage, local cooling of the 

 central portion with liquid air did not produce any appre- 

 ciable change, the purple colour remaining as before. This 

 surprised me somewhat, as Merton has found that the 

 immersion of his tube in liquid air produced a white dis- 

 charge immediately. The conditions were not precisely the 

 same, however, in the two cases. 



If no liquid-air trap is included between the tube and the 

 pump, the mercury spectrum usually develops to a greater 

 or less degree at certain stages of the operation of the tube. 

 This sometimes happens when the tube is in the white stage, 

 the hydrogen spectrum disappearing entirely ; the colour 

 changes from white to blue-white and the prism shows only 

 the yellow, green, blue, and violet mercury lines. If the 

 spectrum is photographed in this condition it is found that 

 the X group of lines is quite as strong as before the appear- 

 ance of the mercury, although the ultra-violet hydrogen 

 lines in the vicinity have disappeared entirely, further 

 evidence in favour of the hypothesis that the lines are not 

 due to hydrogen. 



It is a little difficult to see how the operation of the tube 

 with a heavy current can bring on the white stage and then 

 destroy it, which is what invariably happens, for the white 

 stage never appears for the first time until the current has 

 been on for fifteen or twenty seconds. Once produced, how- 

 ever, we may stop the current, exhaust, admit fresh hydrogen 

 through the Pd tube, and have the white stage at once on 

 closing the switch. 



Effects of a Condenser. 



With the tube in the white stage, showing no trace of 

 the Balmer lines, or only a very faint trace of H a , if a con- 

 denser is placed in parallel with the circuit, the Balmer 

 lines at once appear, and the colour of the tube changes to 

 salmon-pink. 



Current measurements showed that the introduction of the 

 condenser caused an enormous increase in the total current (its 

 integrated value) quite apart from any considerations of time. 



