tlii'OiKjli Rarefied Gases. 357 



face, and the aperture of which never amounts to more than a 

 few degrees. If the electrode, for instance, is formed by a 

 thin wire, the negative hght surrounds it in the form of a 

 spheroid; through the spheroid a small assemblage of cones 

 mav be seen more luminous than the rest. The ne2:ative lii^'ht 

 extends therefore in straight lines from the electrode, and 

 from each element of the electrode, chiefly in a direction near 

 the normal. 



12. Let us now^ assume that the pressure and intensity of 

 discharge are such that the wall surrounding the negative elec- 

 trode shows the phosphorescent light. If an additional wire, 

 not connected with any source of electricity, is introduced 

 parallel to the electrode, its shadow will appear on the sur- 

 rounding wall. The shadow will consist of a small and sharp 

 dark line on the phosphorescent glass. As soon, however, as 

 we connect the second wire with the electrode, so that we have 

 a double negative pole, two dark and well-defined fields will be 

 seen to replace the linear shadow. These fields are bisected 

 by a plane passing through the electrodes. The form of these 

 fields is that of a rectangle the longer sides of which run pa- 

 rallel to the electrodes, and the shorter sides of which are re- 

 placed by a curve convex towards the exterior. These and 

 some other phenomena, to be described in another communica- 

 tion, are explained by the following experimental facts. 



The negative electrode is the seat of a repulsion, which de- 

 flects every line of flow passing near it away from the elec- 

 trode. I have convinced myself by special experiments that 

 these lines of flow are straight lines until they nearly reach 

 the electrode ; they are then almost suddenly deflected, so that 

 they assume the form of a branch of a hyperbola. 



13. The phosphorescent light of the giass produced by the 

 negative discharge is but seldom of uniform intensity, and 

 sometimes shows very strange patterns. One can always show 

 that the light depends on the form of the surface of the elec- 

 trode. The cause of this effect of the electrode is the deflection 

 of lines of flow which I have just mentioned. Lines of flow 

 coming from one part of the surface and passing near other 

 elements of the surface are deflected by it, and cause the ap- 

 pearance of a pattern on the glass tube. This explains the fact 

 that patterns stamped into the electrode show themselves in 

 the phosphorescent light on the glass wall. I mention, for 

 instance, the head of a coin which was used as a negative elec- 

 trode, and which appeared with every detail on the glass wall 

 in the phosphorescent light, even if the wall was at a distance 

 of several centimetres. 



I now pass to a few observations on stratifications. 



