100 LETTEES OF BEEZELIUS 



oxygen modify it in such a way that it unites at 

 ordinary temperatures with many bodies, on which 

 pure oxygen, under the same conditions, has no action. 

 Such substances are for example HO, PbO, MnO, 

 AgO, which, when they combine with oxygen, so 

 change it that it decomposes potassium iodide, 

 converts -potassiuiq ferrocyanide into fsrricyanide, 

 colours guaiacum resin blue, etc. ; in other words it 

 acts like ozone. 3. When exposed to the action of 

 electric sparks for some time oxygen gas prepared 

 from melted potassium chlorate does, it is true, turn 

 starch paste containing potassium iodide blue, and 

 affects the organs of smell, but it is not, according 

 to my experiments, dissolved to any great extent 

 by solutions of potassium iodide or potassium ferro- 

 cyanide, or oxidizable metals ; and the residual gas 

 behaves like ordinary oxygen. Now I should think 

 that a cubic inch of dry oxygen gas if exposed for a 

 sufficient length of time to the electric current, 

 would be completely ozonized, and thus be entirely 

 taken up by the abovementioned solutions. 



4. But infinitesimal traces of water vapour, such 

 as might be present in oxygen supposed to be dry, 

 suffice to produce a perceptible quantity of ozone, and 

 so it seems to me that some such view as mine is 

 less improbable than yours. 5. The destruction of 

 ozone on heating is readily explained by the assump- 

 tion that at higher temperatures it is split up into 

 water and oxygen, like so many other superoxides, 

 especially The'nard's oxidized water. According to 

 your view we must assume that heat restores the 



