﻿Ultra-violet Light on Chlorine. 771 



The following table contains the result of the experiments, 

 which were carried out in the sequence given. 



Diam. of nuclei, 

 uini. 



(1) Air alone without 



A very feeble effect. 





sparking. 







(2) Air illuminated. 



Very considerable 

 effect, grey. 



6xl0" 3 



(3) „ + CS 2 . 



Reddish brown 

 cloud. 



4xl0- 5 



(4) „ +C$ 2 + C1 2 . 



Scarcely any. 





(5) „ +OS 2 + H a O. 



Powerful effect, 

 grey. 



6xl0- 6 



(6) „ + CS 2 fH 2 0-f-Cl 2 . 



Scarcely any. 





(7) Pure Cl 2 . 



Very slight. 





(8) Air again. 



Powerful effect, 

 grey. 



6xl0- 6 



(9) „ + C1 2 (2). 



Scarcely any. 





(10) „ +Cl 2 -fCS 2 . 



Scarcely any. 





Thus it is seen that the results with the steam-jet fall into 

 line exactly with those obtained by means of the electro- 

 meter, and indicate that chlorine forms neither ions nor 

 nuclei, but that its presence is destructive of those otherwise 

 produced in air. In this they confirm the experiments of 

 Burgess and Chapman * as opposed to those of Bevan. 



Conclusions. 

 The following effects have been confirmed or established : — 

 That under the influence of ultra-violet light, 



(1) There is very little ionization produced, even in 



ordinary air, by light of wave-length longer than 

 200 w. 



(2) That pure dry air is not greatly ionized by wave-lengths 



longer than 180 ft/i, but with shorter wave-length 

 the ionization is considerable. 



(3) That the addition of a trace of water vapour to air 



enormously increases the amount of ionization. 



(4) That the addition of very small quantities of chlorine 



to ordinary air increases the amount of ionization, 

 but that with more chlorine the reverse takes place. 



(5) That pure chlorine is not ionized appreciably. 



(6) That pure chlorine does not form condensation nuclei. 



(7) That chlorine in considerable quantity destroys the 



ions and the nuclei otherwise formed in air. 



* J. C. S. Trans, vol. xc. p. 1399 (1906). 

 3 E 2 



