404 3fr. Fredericl Soddy [March 15, 



the only ones to choose between, if radium is produced by the known 

 process in the disintegration of one element, rather than by the at 

 present entirely unknown process consisting of a synthesis of several 

 together. Of these two, since Mme. Curie discovered radium in 

 minerals containing large quantities of uranium, it was natural to 

 select uranium as the probable primary parent in the disintegration of 

 which radium is produced. If uranium were the direct parent of 

 radium, 100 grams of uranium, purified chemically from radium, would 

 generate a fresh quantity cf radium easily detected by the instrument 

 shown after the lapse of only a few hours. 



In preliminary experiments started nine years ago, a kilo of 

 uranyl nitrate ( = 500 gr. of uranium) was carefully purified from 

 radium by precipitating sulphate of barium in its solution. The 

 solution was preserved in a closed vessel, and tested from time to 

 time. After 500 days it was found that the quantity of radium had 

 increased by an amount, which although excessively minute, was 

 about one hundred times as much as the radium initially present. 

 The methods of testing for radium were not then so accurate or 

 refined as those later introduced by Strutt and Boltwood, but there is 

 no reason to doubt their general trustworthiness. Tlie observed 

 growth of radium was, at most, only one thousandth part of wliat, 

 according to existing data, should have been produced if uranium 

 were the direct parent of radium. Small as it was, however, this 

 was the first observation of a ])roduction of radium. It indicated the 

 existence in commercial uranium compounds of a substance capable 

 of generating radium with the lapse of time. In the meantime the 

 character of the pro])lem liad changed. As a sim])le consequence of 

 the theory of atomic disintegration, it follows that, if uranium is the 

 parent of radium, proportionality must exist between the quantities 

 of radium and uranium in all radioactive minerals, which have not 

 been altered by external agencies whilst contained in the earth's 

 crust. The quantity of radium after the lapse of some tens of 

 thousands of years will come into equilibrium with that of the 

 uranium when as much radium is produced as disintegrates in each 

 unit of time. Then Ap P = Xq Q of the differential equation, so that 

 the ratio between the quantities must be the same as the ratio be- 

 tween the periods of average life of the two elements. Researches 

 carried out by McCoy, Strutt, and Boltwood, particularly the last 

 two, proved that this was the case. The ratio of radium to uranium 

 in all unaltered minerals is, on present data, 3-1 x 10"'^ to 1, that is, 

 there are 310 mg. of radium per ton of uranium. From this ratio, 

 the period of uranium calculated from that of radium is 8,000,000,000 

 years. Thus was obtained fairly conclusive, though indirect, evi- 

 dence that uranium is the primary parent of radium. Unfortunately 

 it still remains the only evidence available. To account for the ex- 

 cessively slow growth of radium in the first uranium preparations 

 studied it was necessary to suppose that between the uranium and 



