648 TRANSACTIONS OF SECTION A. 



about ^-^rslssTirs of a gramme of elemental arsenic, that amount being sufficient to 

 ]iroduce a well-detined mirror of arsenic on the drawn out tube. 



The apparatus previously described by me,^ which consists of a porous pot 

 with platinum foil outside as anode and the element under examination as 

 cathode inside the porous pot, was used in these experiments and the following 

 elements were tried as cathodes : lead, zinc, cadmium, tin, silver, graphite, iron, 

 platinum, aluminium, gold, cobalt, nickel, and palladium. 



The above order represents their efficiency in removing arsenic from solutions 

 of avsenious acid, the first being the most efficient, but this order does not hold 

 good for efficiency in removing arsenic when it exists in the form of arsenic acid, 

 for whereas the metals lead, zinc, cadmium, tin, and silver all reduce arsenious 

 acid with the same velocity, only 41 per cent, of the arsenic acid present in the 

 cathode chamber is converted into arseniuretted hydrogen by a zinc cathode in 

 125 minutes, while with a lead cathode 98 per cent, is reduced in the same time and 

 under the same experimental conditions. Again, silver which with arsenious 

 acid effects the reduction with a velocity equal to the first four mentioned metals 

 fails when used as a cathode to liberate any arsenic as AsH., from arsenic acid. 

 The supertension hypothesis put forward by various writers utterly fails to ex- 

 plain these results. Palladium with a comparatively high supertension (0'40 volt) 

 stands at the bottom of the series both as regards arsenious and arsenic acids, 

 while silver which has a low supertension (0"15 volt) reduces arsenious acid 

 with a velocity equal to that of lead or zinc. Iron in this respect is remarkable 

 with the lowest but one supertension (0'08 volt). It efi'ects the reduction of 

 arsenious acid with a velocity nearly equal to lead and stands next to lead in its 

 capacity for reducing arsenic acid. 



Lead has a slightly lower supertension than zinc, yet lead, as already men- 

 tioned, is far more efficient in the reduction of arsenic acid than zinc. 



From the estimations made by collecting the arsenic evolved in intervals of 

 two and a half minutes the first four metals gave a constant for a unimolecular 

 reaction. 



' Memoirs a/td Proceedings of the Manclwiter lAt, and Phil. Sue, vol. xlviii., 

 part iii., p. 17. 



