26 



and placed in the fourth column. It was then possible to 

 obtain from each reading a determination of the pd of that 

 aluminium sheet which the a particles of uranium could just 

 penetrate. For example, the table shows that when 

 pd ='000949, i/I = '456. The abscissa of curve A correspond- 

 ing to this ordinate is *205. Consequently the pd of full 

 range is equal to •000949/*205 = '00463. The figures in the 

 last column show the result of this calculation in the case of 

 each observation. Their close agreement shows that the ex- 

 perimental results fit accurately a curve derived from the 

 theory given above, and is good evidence of the soundness of 

 the calculation. 



The mean of the values in the last column is '00456. 



Had ill m. 



A very small quantity of radium bromide was dissolved 

 in water and evaporated on a platinum plate. It was then 

 raised to a bright red heat, in order to expel the emanation. 

 Some RaC still remained, but this fell to a negligible value 

 in a few hours, as was shown by the fact that the ionisation 

 current due to the newly prepared layer declined to about 

 half value in that time. It was then re-heated so as to drive 

 off such fresh emanation as had been formed since the pre- 

 vious heating. It has been shown by Kleeman and myself 

 ("Phil. Mag,," Dec, 1904), that a layer so heated is very 

 nearly free from all the radio-active products of radium. 



The same aluminium foils were used as in the previous 

 experiment. The results are given in the following table : — 



There is here not quite such good concordance between the 

 figures shown in the last column as there was in the case of 

 uranium. This is not, perhaps, a m.atter for surprise. As- 

 suming that the theory of Part I. of this paper is correct, 

 then, if the observed results are to fit the calculated curve 

 exactly, the active material under consideration should emit 

 only a particles of one ranee /^/.e., of one velocity on leaving 

 the parent atom, though not, of course, of one velocity on 



