60 EEPOET— 1889. 



ground, minute quantities of chlorine liberated gave a green tinge to the 

 gas. The following is the order in time in which a green tint was first 

 observed in the different tubes : — 



After 1 day moist ozone and moist hydrogen chloride. 



„ 3 days „ oxygen „ _ „ 



„ 5 „ partially dry ozone, partially dry hydrogen chloride. 



,. 96 „ „ „ oxygen „ „ „ „ 



Ozone and hydrogen chloride were dried over phosphorus pentoxide 

 and exposed to light for 96 days (May 24 to August 28) ; the gas in the 

 tube was then drawn through a solution of potassium iodide, but no trace 

 of free chlorine could be detected. 



Similar tubes containing dry and moist hydrogen chloride and ozone 

 were kept in the dark for 96 days. After that time no trace of green 

 colour could be detected in the tubes, and on the addition of solution of 

 potassium iodide to the contents of the tube, chlorine and ozone were 

 proved to be absent. Although gaseous hydrogen chloride is not de- 

 composed by ozone in the dark, yet it was found that when ozone was 

 allowed to pass through a 30 per cent, solution of the acid (in the dark) 

 a decided smell of chlorine was observed, especially after allowing the 

 solution to stand for some time in a stoppered bottle. A solution of 

 hydrogen dioxide added to a 15 per cent, solution of hydrochloric acid in 

 the dark also liberated chlorine, which was at once detected by the smell. 

 Experiments in this direction are still being made. 



In a previous report it has been pointed out that dry hydi-ogen chloride 

 and dry oxygen are unchanged even after long exposure to sunlight, but 

 that the presence of liquid water very much hastens the decomposition of 

 the acid, the oxidation being most rapid when water is present sufficient 

 to form a saturated solution of the gas. 



In this, as in many other cases, it is the combined influence of water 

 and oxygen in presence of sunlight which causes oxidation, and in seeking 

 for an explanation of the part played by water in promoting decom- 

 position the possibility of the formation of hydrogen dioxide suggested 

 itself. In order to decide this point a very large number of experiments 

 have been made, only a few of which need be quoted. It is quite obvious 

 that in the absence of any substance upon which the peroxide can act, 

 minute quantities which may be formed in pure water will readily be 

 decomposed and so escape detection. In the first experiments varying 

 quantities of pure ether were added to the water, as it is known that the 

 presence of this substance increases the stability of the peroxide without 

 itself being acted upon. 



100 CCS. of water were mixed with 25 ccs. of pure ether ; this liquid 

 was exposed to light in an atmosphere of oxygen for 66 days (Dec. 10 to 

 Feb. 4). After this time some of the ether was shaken with a solution 

 of potassium bichromate, when the presence of hydrogen peroxide was at 

 once shown by the decided blue colour imparted to the ether. Tubes 

 containing pure ether and water were exposed to sunlight in presence of 

 oxygen from Feb. 7 to May 16 (98 days) : these solutions after exposure 

 readily liberated iodine from a 5 per cent, solution of potassium iodide ; 

 this was estimated by ^'"^^^ solution of sodium thiosulphate. After addi- 

 tion of the iodide, the solution was shaken and allowed to stand in the 

 dark for some hours before titration with the thiosulphate. 



Before giving the results of these experiments it may be mentioned 



