442 SCIENCE PROGRESS 



both vessels was the same. Some gases had no appreciable 

 effect on the rate of decomposition of the ozone, whilst the rest, 

 even when present in very small quantities, were found to 

 accelerate the destruction of the gas enormously. This very 

 marked difference in the behaviour of the various impurities is 

 of interest because it points to the conclusion that the catalytic 

 effects observed are the result of chemical action ; for, if they 

 originated in a purely physical influence, we should expect every 

 substance to possess in some small degree the property of 

 increasing the rate of the chemical transformation. Moreover, 

 the gases chlorine and nitrogen peroxide, which have been 

 shown to be powerful catalytic agents in determining the recon- 

 version of ozone to oxygen, almost certainly form with ozone 

 unstable addition compounds. The latter conclusion is based 

 upon two facts. Firstly, iodine — an element closely allied to 

 chlorine — is converted by ozone into iodic acid in the presence 

 of water; secondly, nitrous acid is completely oxidised to nitric 

 acid in the same circumstances. It is possible, therefore, that 

 dry ozone forms with chlorine and nitrogen peroxide unstable 

 ozonides which are very rapidly destroyed and give rise to 

 oxygen. That the destructive effect of alkalies on ozone arises 

 from a similar cause is undoubted ; the following facts admit of 

 no other interpretation. When ozone is passed into a con- 

 centrated solution of potash at a low temperature the colour of 

 the solution changes to brownish yellow and no ozone escapes 

 through the solution. The brown solution is extremely unstable 

 and when it decomposes only potash and oxygen are formed. 

 Moreover, if the brown solution be rapidly neutralised with 

 acid, the temperature being kept low, not a trace of hydrogen 

 peroxide can be detected in the solution. A concentrated solu- 

 tion of soda behaves in a precisely similar manner towards 

 ozone. The brown solutions are therefore unstable ozonides. 

 Further, the facts of organic chemistry speak eloquently in 

 favour of the view that molecules can become more unstable or 

 chemically active when united with other molecules. This view 

 of catalysis in idiochemical changes is of course not novel, though 

 direct evidence that catalysis in gaseous changes can arise from 

 such a cause has hitherto been peculiarly defective. The 

 deficiency of evidence in support of such a view has been the 

 chief cause of Armstrong's postulation of the formation of 

 addition compounds, in his theory concerning the influence of 



