PROCEEDINGS OF SECTION B. 363 



a quarter of the cross-section with concentrated HaSO^, and about the 

 first 5cm s. of its length was loosely packed with glass wool. The 

 forward end, C, was bent up into the same straight line with the 

 testing apparatus to prevent possible eddies in the air current. The 

 active gas then passed through D, a brass cylinder with insulated cen- 

 tral electrode, where a strong electric field removed any ions or excited 

 activity that would otherwise be carried into the testing cylinder. F, 

 the measuring vessel, consisted of a brass cylinder with a central electrode 

 insvlated from the cylinder by sulphur plugs with intermediate guard 

 rings connected to earth. The cylinder ends were capped with brass 

 gauze to prevent alteration in the volume of the field with varying 

 voltages. Glass tubing (G), of the same diameter as the measuring 

 cylinder, connected with the filter-pump by a conical glass tube. The 

 filter-pump was actuated from a cistern with a constant level overflow, 

 and, with its supply pipe, was rigidly fixed. The current was absolutely 

 constant over the whole range of time used in the measurements. 



The HsSO^ tube was found necessary to keep the insulation in the 

 testing vessel good. It did so quite efficiently, the insulation being con- 

 stantly tested. Experiments showed that its absorption (if any) was the 

 same for emanation, varying as Avidely in activity as in the test experi- 

 ments. The whole apparatus was rigidly fixed. Connections between 

 the tubes were made by rubber tubing, which was painted over after- 

 wards thoroughly with rubber solution. All parts of the measuring 

 system were screened in tinfoil-covered cardboard boxes connected to 

 earth. A key of the type described by (A) Rutherford was used. It 

 gave a constant small electrostatic effect on breaking the earth con- 

 nection, and the measuring range was taken beyond the influence of 

 this. An electroscope of the C. T. R. Wilson type was used. It was 

 adjusted to its maximum sensitiveness inclination with about 240 volts 

 P.D. between the plate and gold leaf. Earth connections were soldered. 

 A battery of test-tube accumulators was used. Two hundred and fifty 

 volts certainly produced practical saturation, and 280 volts was used. 

 Readings were taken over precisely the same points on the scale as much 

 as possible. As, however, the solubility of the thorium emanation in 

 the solutions was unknown, the different orders of activity were tested 

 against thorianite standards. Thorite, thorogummite, aeschynite, and 

 broeggerite were compared with TOOOO gram thorianite made up as 

 detailed, and the others with ■2500--l()00 grams thorianite. The mineral 

 tested and its standard were taken over precisely the same range 

 in the measurement. In addition, a known amount of thorianite in 

 solution was added to a sulphate solution of a monozite of known activity. 

 The solution was then made up to standard volume and tested. The 

 total activity was equal to the sum of the separate activities, so that the 

 solvent action of the two sets of solutions used was the same for the 

 quantities involved. The decay of the emanation gave rise to excited 

 activity in the measuring cylinder, and the activity due to this had to be 

 subtracted from the total activity. 



In practice, readings were taken with the mineral solution in, till 

 constant readings Avere reached ; then five (5) readings were taken in 

 quick succession over the chosen range. The solution tube was then 

 rapidly cleaned out hj a syphon and washed thoroughly with water. 



{h) " Rutherford Kadio-actiyity," second edition, p. 97. 



