has been shown that the activity of a catalyst for the hydrogenation of an ethylenic linking, in the presence of varying concentrations of certain inhibitants, is a linear function of the concentration of the poison. In certain cases, the main linear portion of this poisoning curve persists until more than 90 per cent. of the catalytic activity has been suppressed, whilst with further additions of the inhibitant the activity of the catalyst falls off far less steeply. It appeared desirable to

more »

... st the applicability of a linear poisoning law to other types of reaction, and with this in view the catalysis of hydrogen peroxide with finely divided platinum has been chosen for study. Bredig, Ikeda, and Reinders (2. physikal. Chem., 1901, 37, 1, 323 ; " Anorganische Fermente," Leipzig, 1901) have examined this reaction in detail, and a study of their results with iodine, bromine, hydroxylamine, and hydrochloric acid as inhibitants shows that these are not incompatible with an initially linear poisoning law. The results of Bredig and his collaborators with cyanogen iodide are apparently . inconsistent with a linear graph, whilst those obtained for other inhibitants are inadequate for the present purpose ; moreover, in any case, the number of points in positions suitable for testing this linear relationship is not large, especially in view of the relatively high experimental error. In the further experiments which have accordingly been made, the platinum catalyst was used in a non-colloidal condition, in order to avoid complications due to coagulation. Further, poisons such as hydrogen cyanide or carbon monoxide, which, by oxidation or otherwise, are changed' into products of different toxicity, were avoided, and those actually used-mercuric chloride, mercuric nitrate, and lead acetate-can be regarded as remaining constant in concentration throughout an experiment. The results obtained with these inhibitants show that, for a wide range, the activity of the catalyst is a linear function of the concentration of the poison. With still higher concentrations of the poison, the extinction of activity is far less abrupt. Divided Platinum. The Influence of Inhibitants. E X P E R I M E N T A L. In order to prepare an aqueous suspension of finely divided platinum of constant activity, an approximately N/BO-solution of