Irreciprocal Conduction. 143 



which produced the film before platinization are now unable 

 to do so, but whether the surface of the anode has increased 

 by platinizing in the same ratio we are unable to say. In 

 one case, in which before platinizing the film was formed 

 with a current of 70 divisions, after a few moments' platiniz- 

 ing the film would not form with currents as high as 300 

 divisions. 



3a. Experiments with J1.$0± and Carbon Electrodes . 



In these experiments we at first found considerable difficulty 

 owing to the disintegration, during electrolysis, of the various 

 kinds of carbon. Carre-prepared carbons, gas-carbons, and 

 graphite (from lead-pencils) all behaved in the same way. 

 On making circuit no gas at first comes from the anode, but 

 eventually the outer shell breaks off and gas is evolved *. It 

 occurred to us that the coherent carbon filaments used in 

 incandescent lamps might be free from this objection, and 

 this conjecture was found to be justified. At first filaments 

 from Swan lamps were used, and afterwards some kindly sup- 

 plied by Messrs. Woodhouse and Raw^son. With either of 

 these carbons we were able to electrolyze strong H 2 S0 4 

 without any disintegration or coloration of the liquid. 



Using these carbon filaments we were able to readily pro- 

 duce a film at the anode which had the same general properties 

 as that obtained with platinum. To make sure that the film 

 was not due to any coating of oxidizable material the fila- 

 ments were boiled in aqua-regia and well washed, but this had 

 no influence in preventing the formation of the film. Plati- 

 nized carbon behaves as regards the film-formation exactly 

 like platinized platinum. The insulating condition, with 

 carbon filaments, is accompanied, as in the case of platinum, 

 by the formation of a gaseous sheath round the anode. 



3 b. Experiments with H 2 S0 4 and Gold Electrodes. 



Some fine gold wire, sold to us as pure by Messrs. 

 Johnson, Matthey, & Co., and a piece of platinum foil were 

 used together as electrodes in strong pure H 2 S0 4 ; the pla- 

 tinum and gold being placed in separate limbs of a U-tube. 

 The behaviour of this combination was very peculiar. The 

 film readily formed when the gold was anode, but no sooner 



* This behaviour of carbon has been noticed by Gore and others, and 

 more recently studied by DebrayandPechard {Compt. Rend. cv. pp. 27-30). 

 These observers, using gas and other kinds of carbon purified by Ul at high 

 temperatures, found that the gas evolved at the anode during the electro- 

 lysis of HjSO,! was a mixture of CO a and O. The black powder formed 

 on disintegration (after being washed and dried) deflagrated at a tempera- 

 ture below red-heat, evolving CO and C0 2 . 



