130 THEORIES AS TO ACTION OF 



be a part of the catalytic reaction, it must be shown not only 

 that it occurs, but that the sum of the times, for its formation 

 and its breaking up again to form the final products of the reaction, 

 is less than the time of the direct non-catalysed reaction. Other- 

 wise it is clear that the reaction will run more quickly without 

 the intermediate body being formed. 



It is clear that this condition is satisfied in those cases in which 

 the reaction does not go at all in the absence of the catalyser, 

 or does not go within appreciable experimental limits, as in the 

 case of sulphur dioxide and atmospheric oxygen mentioned above. 

 Here it is obvious that the reaction without the catalyst runs 

 more slowly than the sum of the formation of nitrosyl-sulphuric 

 acid, and the decomposition of this to form sulphuric acid and 

 nitric oxide, without any quantitative work on the subject. Also 

 in one case of a reaction which does run at a measurable rate by 

 itself, viz. that between hydriodic acid and hydrogen peroxide 

 in which water and iodine are formed, but which is catalysed into 

 much greater velocity by molybdic acid, it has been shown by 

 Erode that an intermediate compound per-molybdic acid can be 

 isolated, and also that the sum of the times of formation of 

 (1) per-molybdic acid from molybdic acid and peroxide, and (2) the 

 oxidation of hydriodic acid by per-molybdic acid to iodine with 

 regeneration of molybdic acid, is less than the time of direct 

 oxidation of hydriodic acid by peroxide in the absence of molybdic 

 acid. 



Although a proof that the sum of the times was greater than 

 the direct time would rule out the intermediate compound, it 

 must be clearly pointed out, however, that the times being less 

 does not necessarily prove that the intermediate compound forma- 

 tion actually occurs and is the cause of the catalyses. What it 

 does prove is that the formation of the intermediate compound 

 would increase the velocity of the reaction by a definite amount ; 

 but such increase might be due to other causes, and the increase 

 by such other means might be greater than that by the path of 

 the intermediate reaction. 



Still, an experimental proof that the path by intermediate 

 reaction is quicker, is presumptive evidence that the reaction 

 probably occurs by such intermediate reaction. 



It is only in a comparatively few of the known catalytic reac- 

 tions, however, that such intermediate compounds have been shown 



