316 Royal Society. 



same way upon pure mercury ; but, from the very minute quantity 

 of it which is at any time present, the experiment requires some 

 care in order that the effect may be observed. Ou passing a stream 

 of atmospheric air, which gave the usual reactions with test-paper, 

 for some hours over the surface of mercury in aU-tube, the metal 

 was distinctly oxidized at the end at which the air first came into 

 contact with it. 



This experiment, however, cannot be considered conclusive, as 

 mercury will tarnish and lose its mobility under the influence of 

 many bodies besides ozone. 



It is well known that all ozone reactions disappear when ozone 

 is passed through a tube containing pellets of dry peroxide of 

 manganese, or other body of the same class. The same thing 

 occurs with the substance supposed to be ozone in the atmosphere. 

 About 80 litres of atmospheric air were drawn, at a uniform rate, 

 through a tube containing peroxide of manganese, and afterwards 

 made to play upon very delicate test-paper. Not the slightest co- 

 loration occurred, although the same paper was distinctly affected 

 when 10 litres of the same air, without the interposition of the 

 manganese tube, were passed over it. 



But the action of heat furnishes the most unequivocal proof of the 

 identity of the body in the atmosphere with ozone. In a former 

 communication (Phil. Trans, for 1856, p. 12) I showed that ozone, 

 whether obtained by electrolysis or by the action of the electrical 

 brush upon oxygen, is quickly destroyed at the temperature of 

 237° C. An apparatus was fitted up, by means of which a stream 

 of atmospheric air could be heated to 260° C. in a globular glass 

 vessel of the capacity of 5 litres. On leaving this vessel, the air 

 was passed through a U-tube, one metre in length, whose sides 

 were moistened internally with water, while the tube itself was 

 cooled by being immersed in a vessel of cold water. On passing 

 atmospheric air in a favourable state through this apparatus, at 

 the rate of three litres per minute, the test-paper was distinctly 

 tinged in two or three minutes, provided no heat was applied to 

 the glass globe. But when the temperature of the air, as it passed 

 through the globe, was maintained at 260° C, not the slightest 

 action occurred upon the test-paper, however long the current con- 

 tinued to pass. Similar experiments with an artificial atmosphere 

 of ozone (that is, with the air of a large chamber containing a 

 small quantity of electrolytic ozone) gave precisely the same re- 

 sults. On the other hand, when small quantities of chlorine or 

 nitric acid vapour, largely diluted with air, were drawn through 

 the same apparatus, the test-paper was equally affected, whether 

 the glass globe was heated or not. 



Erom these experiments I consider myself justified in conclu- 

 ding that the body in the atmosphere, which decomposes iodide of 

 potassium, is identical with ozone. 



