CONDITIONS OF CHEMICAL CHANGE 667 



minute ; but when moist chlorine was used, no action at all 

 could be observed. Therefore, 10 to 15 cm. of moist chlorine 

 will cut out all the chemical rays and the same length of the 

 dry gas approximately eight-ninths. There is some difficulty 

 in reconciling Mellor's observation with Bunsen and Roscoe's 

 measurements. From the work of the latter authorities, the 

 natural conclusion would be that moist chlorine should cut 

 out an amount of light not differing greatly from that found 

 by Mellor for the thoroughly dried gas. The results of 

 experiments on these lines largely depend, however, on the 

 character of the light used ; hence it is unfortunate that Mellor 

 does not mention the source of light used to illuminate 

 the insolation vessel. It would appear, therefore, that there is 

 still room for — in fact a pressing need for — experimental work 

 connected with this important aspect of the subject. 



Several hypotheses relating to the interaction of chlorine and 

 hydrogen are based on the assumption that the change takes 

 place through an intermediate compound. Mellor has shown 

 that the intermediate compound cannot be chlorine monoxide 

 or hypochlorous acid, by proving that neither of these sub- 

 stances will shorten the induction period ; accordingly he 

 assumes that if an intermediate compound be formed at all, it 

 must be a complex molecule of the type xC\ 2 , ^H 2 0, 2H2, built 

 up of molecules of hydrogen, chlorine and water. The hydro- 

 gen was introduced into the formula of the compound because it 

 was supposed that chlorine by itself acquired no appreciable 

 activity in sunlight. 



Bevan, whose work was published almost simultaneously 

 with that of Mellor (Roy. Soc. Phil. Trans.), made several in- 

 teresting and important observations which led him to adopt 

 an explanation somewhat similar to Mellor's but modified in one 

 respect. Draper had found that a mixture of hydrogen with 

 previously insolated chlorine instantly gave rise to hydrogen 

 chloride on exposure to light. Bunsen and Roscoe denied that 

 this was the case but Fremy and Becquerel (Wurtz. Did. de 

 Chemie, 1879, ii. 255) confirmed Draper's statement. Bevan 

 found that previous exposure of the chlorine to light did, as a 

 matter of fact, cause a difference but the increased power which 

 moist chlorine thus acquired of acting on hydrogen is lost when 

 it is passed through water. Bunsen and Roscoe's failure to 

 confirm Draper's result was therefore due to their having 



43 



