284 Mr Campbell, 



fifth of a or b. Some estimate of the activity of the glass was 

 obtained by comparing the increase in the ionisation which should 

 be caused by the radiation from a lead plate of the size used, 

 calculated from the results given in Phil. Mag., Feb. 1906, p. 217, 

 with the observed difference between the ionisation when the lead 

 plate was in the tray and when the glass tray was empty. The 

 agreement was sufficiently close to show that the activity of the 

 glass could not be a large fraction of that of the lead. 



5. The whole of the measuring box was enclosed in a thick 

 screen of iron, sufficient to absorb nearly all the penetrating 

 radiation from outside : no appreciable amount of the total radia- 

 tion from the materials could be secondary radiation excited from 

 outside. 



6. In making the measurements the ionisation in the vessel was 

 observed when the clean glass tray was in position in the drawer : 

 the tray was then filled with the material under investigation and 

 the measurement repeated. These observations were taken alter- 

 nately several times, and the mean of the differences in the currents 

 through the gas was taken as a measure of the activity of the 

 material. The quantities to be measured are so small that it is 

 impossible to attain great accuracy. The whole increase of current 

 caused by the radiation from lead, the most active material 

 investigated, was only 10 per cent, of the whole current through 

 the vessel. The current could be measured with an accuracy of 

 1 per cent. : this is 1 per cent, of the effect caused by the lead 

 and a much larger percentage of the effect caused by many of the 

 materials investigated. The experiments require considerable 

 time for their performance to this degree of accuracy : work has 

 been much delayed by the difficulty of avoiding spurious activity 

 derived from the air of the laboratory, but it is hoped that these 

 difficulties have been overcome. 



7. The table shows the results obtained. The first column 

 shows the material investigated : the second the activity of a slab 

 of that material in arbitrary units. The third gives the activity 

 calculated on the assumptions that the activity of the metal is 

 correctly given in the second column, and that the whole activity 

 of the compound is due to the activity of the metal contained in 

 it. It will be seen that the only elements occurring besides the 

 metals are sulphur and oxygen : it is known that the activities of 

 these elements are very small — too small to be measured, at least 

 by this method : but their activity must contribute something to 

 the total, and hence it is to be expected that the observed activities 

 will be slightly too high. 



