﻿Ultra-violet Light on Chlorine. 7G9 



The plate of gypsum, examined in the quartz spectrograph, 

 was still very transparent to wave-lengths of 185/-6//, ; the 

 calc-spar had a sharp boundary, being transparent to wave- 

 length 206 /ju/jl, but opaque to 202 /j,{i ; the ultra-violet crown 

 glass absorbed all shorter than 290 fi/n. The evidence is 

 sufficient to warrant the conclusion that, for any considerable 

 ionization, wave-lengths shorter than 200 fi/u are required, 

 and if any ionization is caused by light which can pass 

 through any considerable thickness of glass, the amount can 

 hardly be within our means of measurement. 



Schumann Violet. 



In all the experiments hitherto described, the light traversed 

 at least two centimetres of air and a quarter millimetre of 

 quartz before reaching the gas. In order to see whether any 

 effect was produced by rays of still shorter wave-length, such 

 as air and quartz powerfully absorb, a different illumination 

 vessel was used, made of brass and closed by a plate of fluor- 

 spar. The spark-gap was placed within two or three milli- 

 metres of the window, from which it was separated by a 

 shield of fused quartz pierced with a small hole about the 

 same diameter as the window; the condenser cylinder and 

 rod were of graphite, and the sensitiveness of the electro- 

 meter was reduced to about one half. The following series 

 of results was obtained. : — 



Deflexions cm. 

 + - 



(1) Air alone at 5 litres per 



minute 12, 13 29, 2S 



(2) Air + about 10 per cent. 



chlorine, 5 litres per 



minute 1, 1 —2,4,4: 



(3) Pure chlorine, 3 litres 



per minute 



(4) Air alone, 5 litres per 



minute 12, 14 29, 20 



(5) Air + 50 per cent. 



chlorine 



(G) Pure chlorine, 3 litres 



per minute 



(7) Air alone, 5 litres per 



minute 11 21, 17, 10 



This was a long series of experiments, and the diminution 

 in the effect observed with pure air towards the end of the 

 series is easily explained by the fact that it is impossible to 



Phil. May. S. 0. Vol. 23. No. 137. May 1912. 3 E 



