746 Prof. E. Rutherford on th 



te 



In order to make certain that the observed absence of growth 

 of the radium cannot be ascribed to some chemical action, the 

 solution after 64 days was removed and again treated with 

 ammonia. The precipitate was dissolved in hydrochloric and 

 again tested for growth of radium. By this treatment 

 the initial content of radium was reduced from '14 to *04. 

 No growth of radium was observed over a period of 20 days. 

 The solution was then removed to Manchester and tested 

 at intervals over a further period of 136 days. The growth 

 in the total interval of 220 days is certainly less than *03 or 

 not greater than 1/500 of the normal growth to be expected. 



§ 9. Expenments ivith the active deposit. 

 We have seen in § 7 that the observed constant growth of 

 radium in a solution freed irom actinium X shows that 

 actinium B is not directly transformed into radium. This 

 result has been confirmed by testing directly whether there 

 was any growth of radium in the active deposit of actinium. 

 The active deposit (actinium A and B) was concentrated on a 

 platinum plate by keeping it negatively charged in the 

 presence of the emanation from an active solid preparation 

 of actinium. Four different experiments were tried in which 

 a platinum plate was exposed for 4, 7, 7, and 14 days re- 

 spectively. After exposure, the platinum plate was placed in 

 a solution of hydrochloric acid to dissolve off the deposited 

 matter, and the solution with the platinum plate in situ 

 tested for radium. The first two experiments showed a just 

 measurable quantity of radium, but with still greater pre- 

 cautions against radium contamination, the last two experi- 

 ments showed no measurable amount. It may bo of interest 

 to consider briefly a method of calculating the amount of 

 radium theoretically to be expected, if actinium B changes 

 directly into radium. Suppose as the basis of calculation 

 that the growth of radium in the solution actinium I. (§ 4) 

 is normal, and compare the amounts of actinium B in this 

 solution and on the platinum plate. The « ray activity of 

 the actinium and its products in the solution spread in a thin 

 film corresponded to 9900 divisions per minute in the a ray 

 electroscope, while the activity of both sides of the platinum 

 plate, tested immediately after removal from the actinium 

 emanation, was 3800 divisions. Now the actinium in equili- 

 brium contains four a ray products whose ranges are 4*8, 

 6'55, 5*8, and 5*5 cms. respectively, while the active deposit 

 contains only one of range 5*5 cms. Taking as a first 

 approximation that the ionization due to an a particle from 

 each product is proportional to its range in air, the solution 

 contains an amount of active deposit corresponding to an 



