The Discharge of Electricity through Gases. 529 



though the want of symmetry is not very marked. The important 

 measurements of Wiedemann and Riihlmann* are generally taken to 

 prove a more easy escape of negative electricity under reduced 

 pressure ; but when a number of sparks are taken in rapid succession, 

 as they were in these experiments, recent researches seem to indicate 

 that the gas and electrodes may not return to their original condition 

 between the discharges. The consequence is that such a rapid 

 -succession of sparks shows some of the phenomena of the continuous 

 current, and the want of symmetry is therefore of the same nature 

 as that which belongs to the discharge, and has nothing to do with 

 the circumstances which determine whether a discharge shall pass or 

 not. The same remarks apply to Lichtenberg figures and Priestley's 

 rings; they are complicated cases of phenomena which are best 

 studied by means of the continuous current. I pass on to consider 

 the cases where a current may pass between two electrodes, the 

 ■difference of potential being small. 



We have four different ways of converting the gas into a con- 

 ductor, and our subject must be subdivided accordingly. If the elec- 

 trodes are either (1) heated to redness or (2) illuminated by ultra- 

 violet light, the discharge passes even when the difference of potential 

 does not amount to more than a few volts. Further (3), it has long 

 been known that flames do not behave as dielectrics ; and, finally (4), 

 gases through which a discharge is passing have been shown to 

 conduct freely. 



(1.) Discharge from Glowing Electrodes. 



Edward Becquerel,f as far as I know, was the first to discover that 

 air between red-Lot platinum electrodes ceases to insulate, and he 

 points out that when the two electrodes are of different sizes, the 

 ■equalisation of potential takes place more quickly if the larger 

 electrode is negative. Guthrie, J by a different method of experi- 

 mentation, found that red- or white-hot iron balls discharged elec- 

 tricity more easily when positive than when negative. 



Looking at the question from the point of view of the theory of 

 electrolytic convection, it seems of the greatest importance to decide 

 whether the discharge from hot bodies is accompanied by any 

 chemical action or not. Mr. Arthur Stanton has conducted in my 

 laboratory a research intended to throw light on the question. The 

 experiments were planned and independently carried out by him, and 

 his own account of them appears (p. 559) as an Appendix to this 

 paper. But I must briefly allude to his results, which I consider to 

 be of great importance. 



* < Poggendorff's Annalen,' vol. 145, pp. 235, 364 (1872). 

 f ' Annales de Chimie,' vol. 39, p. 355 (1853). 

 % 'Phil. Mag.,' vol. 46, p. 257 (1873). 



