Relation between Uranium and Radium. 283 



same direction in the early determination by Rutherford and 

 Boltwood of the ratio of uranium to radium in minerals, 

 which involved the comparison of the emanation produced by 

 known small quantities of radium and by uraninite. It was 

 ascribed by them to the partial precipitation of the radium 

 in the standard solution, and caused the value first given for 

 the ratio of radium to uranium to be twice too great. 



At this stage the data given in the last paper may be 

 profitably reconsidered. In the old experiment the radium 

 emanation from the kilogram of uranyl nitrate gave a leak 

 in the electroscope of 15 divisions after 550 days. This cor- 

 responds according to the new calibration with 2 x 10~ 10 gram 

 of radium instead of 1*6 X 10~ 9 as before stated. But owing 

 to the low value probably given by the bubbling test, this is 

 likely to be if anything too low. 



There is a discrepancy in the theoretical estimates of the 

 rate of disinte oration of uranium according to the method 

 employed in the calculation. The value in the last paper for 

 the fraction disintegrating, 6 x 10~ 17 per second = 1*9 X 10~ 9 

 annually, was derived from a consideration of the relative 

 a-ray activity of uranium and radium, and the relative 

 number of a-ray changes in each. This was the only method 

 then available, but neither factor was or is very certainly 

 known. The work of Boltwood and Rutherford on the ratio 

 of the amount of uranium to that of radium in equilibrium in 

 the natural minerals, gives directly the ratio of the rate of 

 change of radium (2*6 X 10 6 ) in terms of that of uranium as 

 unity. Rutherford takes 2 x 10 -10 (year) -1 as the fraction 

 of uranium disintegrating, which is about ten times smaller 

 than the value before used. It is to be remarked that this 

 value makes the equilibrium a-ray activity of radium about 

 15 million times that of uranium, assuming only one a-ray 

 change in the latter case and four in radium*. Possibly it 

 may turn out that more than a-ray change is concealed in the 

 change of uranium into uranium X. 



Accepting Rutherford's value in 500 grams of uranium in 

 550 days 1*5 x 10~ 7 gram disintegrates, instead of 1*5 x 10~ 6 , 

 the old estimate. The result is a rate of production of radium 

 700 times less, instead of, as stated, a thousand times less, 

 than that calculated on the assumption of a direct change of 

 uranium into radium. Owing to the fact that the solution 

 was not boiled this estimate is necessarily somewhat indefinite. 



* Thus Boltwood (Am. Jour. Sci. 1906, xxi. p. 409) states that the 

 final or equilibrium activity of radium is 5'64 times the initial, and 

 Bragg (Phil. Mag. 1906, [6] xi. p. 148) has shown that the range, and 

 therefore the ionizing power, of the a. particle from radium is the same 

 as that from uranium. Hence the a-ray activity of radium in equilibrium 

 ought to hd 5-64 X 2-6 x 10 G = 14-6 X 10° times that of uranium. 



U2 



