TRANSACTIONS OF SECTION A. 531 



Department I. — Physics. 



The following Report and Papers were read : — 



1. Report on Electrical Standards. — See Reports, p. 31. 



2. Note on a Comparison of the Deposits in Silver Voltameters with 

 different Solvents. By S. Skinneb, M.A. — See Reports, p. 32. 



3. The Discharge of Electricity through Mercury Vaj)our. 

 By Arthur Schuster, F.R.S. 



The experimental investigation of the passage of electricity through mercury 

 vapour is of interest on account of the metallic nature of the element, the 

 monatomic character of the vapour, and the purity with which it can be 

 obtained. Previous results of the author had led to the conclusion that the 

 discharge of electricity through mercury vapour differed fimdamentally from that 

 taking place through other gases, but these results have been called in question by 

 other experimenters. 



The work now described has extended over two years, but did not lead to 

 results which may be said to be decisive on the account of the extreme difficulty 

 of excluding small traces of moisture. Though the mercury vapour experimented 

 upon no doubt was much purer than that obtained by any previous observers, it 

 was not absolutely free from some other gas, which, probably, was aqueous 

 vapour. The width of the well-known dark space round the kathode observed 

 ■was ten times larger than in air. This dark space, however, may possibly be due 

 to the small remnant of impurity which, as has been pointed out, could not be 

 excluded. 



Sur les Effets magne'tique de la Convection e'lectrique. 

 Par Dr. V. Cbemieu. 



5. Photoelectric Cells. By Professor G. M. Minchin, M.A., F.R.S. 



During the past summer I have been engaged on the study of the photoelectric 

 cells with which I had measured the voltage produced by the light of the stars. 

 The object of this investigation was to discover whether the life of a cell could be 

 in any way prolonged or not, and also to find the best liquid that could be 

 employed. 



In these cells the surface, which is sensitive to light, is a thin layer of selenium 

 spread on the end of an aluminium wire, the selenium layer being heated gradually 

 after it has been spread as a black viscous liquid on the end of the wire imtil it 

 assumes the brownish grey colour which characterises the state in which it is 

 sensitive to light. The aluminium wire is contained in a glass tube, which the 

 wire should so completely lit as to prevent the ascent of a liquid into the tube— 

 a condition which it is impossible to fulfil, as the aluminium wire cannot be 

 sealed into a glass tube. It is essential for complete success that only the 

 selenium layer on the end of the wire should come into contact with the liquid. 

 Let us imagine two such selenium-coated aluminium wires immersed in a small 

 glass tube containing a liquid, one of the wires being completely screened from 

 light, while the selenium on the end of the other can be exposed to light. In the 

 dark there should be no voltage exhibited by this cell when its two wires are 

 connected with an electrometer. If the wire to be exposed is left completely 

 naked — i.e., in free contact with the surrounding liquid — no voltage (or almost 

 none) will be developed when its selenium end is exposed to light. This result is 

 undoubtedly due to short-circuitnig in the cell itself when the light acts ; but if 



