Also 



31 



*S', V'232 + L>^ar) 23-2 



S, ^'239 + 1^16 26-2 



assuming the square root law (Bragg & Kleeman, "Phil. Mag.," 

 September, 1905) to hold for uranium and thorium. 



Hence, finally, 



JV 23-2 



—'' = •234 x-916 X — - =-190. 



N^ 26-2 



This result may be a little too small, since the range of 

 the a particle of thorium may be slightly over-estimated. 

 The square of the range enters into the formula of compari- 

 son, but on the other hand any a rays of long range which 

 have not been removed from the thorium would make I^ 

 too large. On the whole, therefore, the actual value cannot 

 be far from *20, i.e., the uranium atoms break down very 

 nearly five times as fast as the thorium. 



I have preferred to make the method one of comparison 

 of ranges rather than of absolute determination. For there 

 are two or three difficulties in using it for the latter purpose. 

 In the first place, as already said, it is not easy to make the 

 thin aluminium leaf lie very close to the radiating surface, 

 and the layers of air close to the surface contribute a rela- 

 tively large number of ions. To make this error uniform I 

 have used a net of very fine wires, with a mesh of | of an 

 inch, to keep the foils down. The net was, of course, placed 

 over the bare surface also, when I was measured. Again, 

 there is a disturbing effect due to the secondary ionisation of 

 the absorbing sheet. Mme. Curie has called attention to effects 

 of this kind (Rutherford, ''Radio-a<?tivity," 1905, p. 189). 

 T find that there is slightly more ionisation when, of the two 

 layers of foil, Al. and Sn., the latter is on top. Using tin- 

 foil, the range always comes out rather larger than when 

 aluminium foil is employed : f ./;., the range of RaC when 

 tinfoil was used was found to be 7*4 cm., and when aluminium 

 foil was used 6'5 cm. The range of Ra, as found bv the aid 

 of aluminium foil, was 3*26, which is half the range of RaC, 

 as it should be. I had no tinfoil thin enough to give an 

 accurate measurement of the range of the a particle of Ra. 

 Both measurements with aluminium foil are too low, and the 

 one with tinfoil is too high. The tinfoil lies flatter on the 

 surface than the aluminium, which may help to explain rhe 

 difference, but it seems more probable that it is mainly due 

 to the secondary ionisation. 



One other difficulty lies in the way of an accurate deter- 

 mination of the range in air by this method. As has already 

 been mentioned bv Kleeman and myself ("Thil. Mag.," Sept.', 



