Electrical Properties of Selenium. 249 



surface by the hydrogen bubbles which formed upon it ; these 

 became mixed with the liberated Se and made it impossible 

 to ascertain the weight of the latter. 



Exp. 23. — In order to obtain a more coherent substance, 

 two parts of Se were thoroughly incorporated with one of 

 Cu 2 Se ; the mixture was then fused in an earthenware mould, 

 and thus formed into a rectangular plate weighing 1*0731 

 gramme. Se itself being a very bad conductor, it was believed 

 that practically the whole conduction would take place through 

 the selenide. The plate was immersed in water with a strip 

 of Pt, and a current averaging O023 ampere was passed for 

 ten minutes. As soon as the circuit was closed the black 

 plate began to give off dense red clouds ; and when it was 

 dried and weighed after the experiment, it was found to have 

 lost 0*0250 gramme. If the decomposition had been primarily 

 electrolytic, the weight of liberated Se, assuming its electro- 

 chemical equivalent to be 0*00082 *, would have been no more 

 than about half that amount, namely 0*0113 gramme. The 

 red Se was filtered out of the water and examined for particles 

 of debris. None were detected by the unassisted eye, but 

 under the microscope many became visible, though whether 

 in sufficient quantity to account for the observed excess of 

 weight is somewhat doubtful. 



Exp. 24. — This was a repetition of the last, except that the 

 plate contained six parts of Se to one of Cu 2 Se (instead of 

 two to one) and weighed 1*0223 gramme. The current, 

 which averaged 0*016 ampere, was left on for 30 minutes. 

 The actual loss of weight was 0*0386 gramme ; the cal- 

 culated loss, according to the electrolytic law, being 0*0236 

 gramme. In this case, therefore, with a weaker current and 

 a less friable mixture, the observed exceeded the calculated 

 loss by only 64 per cent, instead of 121 per cent, as in the 

 previous instance. The microscope again revealed the presence 

 of numerous small black particles. The closer approximation 

 is in favour of the possibility that the excess might have been 

 due to the disruption of fragments of the selenide. Further 

 experiments, which need not be detailed, leave the point still 

 doubtful. 



This effect of electrical action may be exhibited in a striking 

 manner as follows : — A sheet of moistened filter-paper is laid 

 upon a metallic plate, preferably of Pt, which is in connexion 

 with the positive terminal of a 50-volt battery ; a pencil 

 formed of one part of Cu 2 Se and six parts of Se is connected 



* The electrochemical equivalent of hydrogen multiplied by the atomic 

 weight of Se =0000010384 x 79=0-000820336. Quite possibly, how- 

 ever, the E.C.E. of Se is only half this value. 



Phil. Mag. S. 5. Vol. 40. No. 244. Sept. 1895. S 



