exposed to Entladungsstrahlen. 75 



each other. The wires can be connected to one of the pairs of 

 quadrants of an electrometer, the other pair of quadrants being 

 put to earth. A piece of wire gauze connected with earth was 

 fastened across the top of the tube CD. The outer cylinders could be 

 connected with one of the terminals of a battery of a large number 

 of small storage cells, the other terminal of which was connected 

 with the earth. The two pairs of quadrants of the electrometer 

 were first connected together and the connection then broken. 

 When no discharge passed through the upper tube, the spot of 

 light reflected from the mirror of the electrometer remained at 

 rest ; when, however, the discharge passed through the tube, the 

 spot of light was deflected, showing that conduction took place 

 through the gas between the wire and the cylinder. 



The gas through which a discharge passes is rendered a con- 

 ductor and retains this property for a considerable time ; and one 

 of the chief difficulties of the investigation was to separate the 

 effects due to the diffusion of the gas from the immediate neigh- 

 bourhood of the spark from those that might be due to any 

 radiation from it. At first I endeavoured to do this by placing 

 between the cylinders and the discharge two coaxial tubes so 

 arranged that any gas on its way from the neighbourhood of the 

 discharge to the cylinders would have to pass between the tubes ; 

 the tubes were connected with the terminals of a battery of a 

 large number of storage cells, so that an intense electric field 

 was established between the tubes — this should drive to the walls 

 of the tube the ions in any gas passing between the tubes and 

 thus free it from any conductivity due to ions which had come 

 from the neighbourhood of the discharge. This method did not 

 work well in practice, and I finally relied on the comparison of 

 the leaks through the two cylinders to separate the effects due to 

 convection from those due to the Entladungsstrahlen ; the effects 

 due to convection should not be greatly different in the two 

 cylinders (experimental evidence of this was furnished by ob- 

 servations at very low pressures, when the effects due to convection 

 are very large ; at such pressures the rate of leak was the same in 

 the two cylinders) ; while it is only the cylinder which is directly 

 under the discharge which will be affected by the 'Entladungs- 

 strahlen.' The first experiments were made with an alternating 

 arc discharge from a high tension transformer. When the 

 apparatus shown in Fig. 1 contained air at the pressure of a few 

 millimetres of mercury there was no appreciable leak in the side 

 cylinder, while there was a considerable and regular leak in the 

 one exposed to the 'Entladungsstrahlen.' At pressures comparable 

 with the atmospheric pressure, there was no appreciable leak in 

 either cylinder ; at very low pressures there was a rapid leak in 

 both cylinders due to the convection of the gas from the neigh- 



