Velocities of a. Particles from Radioactive Substances. 563 



the rays from the slit to the photographic plate was shielded 

 from the stray electrostatic field by a long metal cylinder 

 CC, closed at the end next to the slit except for a narrow 

 opening which just permitted the passage of rays from 

 the slit. 



The flat end of the cylinder was at a distance of 0*5 mm. 

 from the ends of the silvered glass plates. Approximate 

 calculations of the effect of the stray field were made by 

 methods similar to those given by Kaufmann *. It was 

 deduced that the correction due to the stray field under these 

 conditions was almost negligibly small. No correction was 

 required for the end near the source, as the source was 

 between the plates, and a disturbance of the field at this 

 point has a relatively unimportant effect on the path. The 

 preliminary experiments with scintillations had shown that 

 the introduction of the metal screen reduced the deflexion 

 produced by a given field by nearly 1 per cent. A difference 

 of potential — generally 1000-2000 volts — was maintained 

 between the glass plates by batteries of storage and small 

 cadmium cells. This voltage was measured by a Nalder- 

 Thompson electrostatic voltmeter reading from 600-1500 volts, 

 which had a very open scale; this instrument was carefully 

 calibrated by direct comparison with two standard Weston 

 cells. The batteries were arranged so that practically any 

 desired voltage could be used, and had contacts at various 

 points dipping into insulated mercury cups, so that any part 

 o£ the battery could readily be tested on the voltmeter. A 

 high-resistance potentiometer, connected up with a small 

 40-volt battery of accumulators, and arranged in series with 

 one o£ the batteries, served to keep the applied potential 

 constant to within a volt throughout the course of the ex- 

 periment. The experiment was carried out in a darkened 

 room, but the battery and voltmeter were kept in an adjoin- 

 ing room, and well insulated leads taken through the walls 

 in tubes of paraffin wax. The battery itself and all the leads 

 were well insulated. A large reversing key made up of 

 mercury cups in a paraffin block, and an arrangement for 

 short circuiting the plates when required, completed the 

 electrical connexions of the apparatus. The exhaustion was 

 carried out by Fleuss and Gaede pumps, and finally by 

 ■coconut charcoal in liquid air. A small discharge tube con- 

 nected to the apparatus, worked by an induction coil, served 

 to indicate the state of the vacuum and enabled us to avoid 



* Kaufin.ann, Phxjs. Zeitschr. viii. p. 75 (1907). 

 202 



