324 Dr. A. Schuster. Experiments on t/te. 



neighbourhood of another electrode. Goldstein has, however, de- 

 scribed some effects of negative electrodes on each other, and E. 

 Wiedemann has recently made an experiment in which some remark- 

 able effects were seen on the approach of the positive electrode to the 

 kathode. Professor Wiederaann's experiment belongs to the same class 

 as those presently to be described, but differs from mine in so far as it 

 refers to very great exhaustion at which the glow is no longer visible, 

 while I have generally worked at much higher pressures and especially 

 studied the effect on the glow. The peculiarity of the arrangement 

 which I have used consists in the large size of the negative electrode, 

 which allows a much easier observation of the parts at which the 

 negative glow chiefly settles. I shall show that it tends to accumulate 

 away from the positive electrode. 



The glow which surrounds the negative electrode is divided into 

 three layers which are, however, only clearly separated when the 

 pressure has been reduced to the fraction of a millimeter. The first 

 layer narrow, and closely surrounding the negative electrode, is with 

 new electrodes of a beautiful golden colour. Its spectrum is chiefly made 

 up of the hydrogen and sodium lines. The sodium is evidently due to 

 matter which has settled on the metal, and the hydrogen comes out 

 of the metal where it was absorbed. In time these lines disappear, 

 the layer loses its golden colour, and the spectrum is now that of the 

 positive half of the discharge. The second layer is the so-called 

 dark space. The name is a good one, although Goldstein has pointed 

 out that this layer is only relatively not absolutely dark. Its 

 luminosity depends much on the gas used. With nitrogen it seems 

 to me to be much darker than with the hydrocarbons. The thickness 

 of this dark space is according to the measurements of Puluj about 

 forty or fifty times as great as the mean free path in air at the same 

 pressure. But the thickness depends on many circumstances, and 

 would not, according to Puluj, be quite proportional to the mean 

 free path at different pressures as we should expect it to be ; but 

 measurements of this sort can be only very approximate at present. 



The third layer is the glow proper. When the curvature of the 

 electrode is small its inner boundary is sometimes defined with quite 

 a remarkable sharpness, but this also depends on the gas which is 

 used. The outer boundary of the glow extends to different distances, 

 according to the intensity of the current, the size of the electrode, 

 and partly also its shape. It would be useless at present to enter 

 further into the different causes which influence the thickness and 

 definition of these different layers. 



The glow is separated from the positive part of the discharge by a 

 non-luminous space, sometimes also called the dark space. In order 

 to prevent confusion I propose the name " dark interval " for it. 



I have used in my experiments on the glow a cylinder open at one 



