568 Dr. C. 13. Bazzoni : Experimental Determination 



tungsten about *8 cm. in length and '08 cm. in diameter, was 

 surrounded by a coaxial cylinder of sheet copper (A), the 

 lead (E) from which was connected with a quadrant electro- 

 meter. As the currents dealt with were sometimes of con- 

 siderable magnitude, the electrometer was provided with 

 an adjustable capacity. The sensibility of the electrometer 

 with no capacity was roughly 600 mm. per volt. At P is a 

 ground cone for connecting the tube with the air-pump and 

 with the apparatus for producing and purifying the helium. 

 The joint at this cone is preferably covered with sealing-wax 

 without the use of tap-grease, but in these experiments it 

 was actually covered with soft wax, which seemed a satis- 

 factory arrangement. The heating current was taken from 

 a 50-volt storage battery, and, since it was absolutely essential 

 that the temperature of the filament should be maintained 

 constant over relatively long periods, the filament was put 

 in one arm of a Wheatstone bridge, which was kept accu- 

 rately balanced by the use of three continuously adjustable 

 resistances in parallel in the main circuit. This heating 

 circuit was insulated from the ground. The negative end of 

 F was connected to a sliding contact on a rheostat, which 

 was in series with a 200-volt storage battery. The potentials 

 were measured directly through a high-resistance Weston 

 standard voltmeter between the negative end of F and the 

 ground. This voltmeter was checked against a Clark cell 

 and a standard resistance, and found correct within 0*1 volt 

 over the range used. 



Connected to P there was first a discharge-tube for 

 examining the purity of the gas with a spectroscope ; next 

 a U tube to be surrounded with liquid air to get rid of 

 mercury ; next a mercury cut-off ; then a U tube with 

 charcoal; then another mercury cut-off ; then a discharge- 

 tube containing phosphorus pentoxide. There was next a 

 stopcock beyond which was a quartz tube containing clevite, 

 with a lateral tube containing potassium permanganate 

 crystals and a final connexion with the pump. Since the 

 ionization potential of helium is higher than that of any of 

 the substances with which it may be expected to be con- 

 taminated, it is of the highest importance to get the gas 

 absolutely pure. In this arrangement the major portion of 

 the hydrogen was removed by sparking for a long time with 

 oxygen over phosphorus pentoxide. The residual hydrogen 

 and other impurities were then taken out by the liquid air 

 and charcoal. The gas finally obtained in the experimental 

 tube was of a high degree of purity, and in the work that 

 follows showed no trace, spectroscopically,, of contamination 



