Relation between Uranium and Radium. 281 



analysed and the content of uranium determined. The 

 quantity of uranium present multiplied by '38xl0~ 6 (the 

 constant ratio of radium to uranium in minerals) gives the 

 quantity of radium in the uraninite used, and in this way 

 quantities of radium of the order of 10~ 9 gram may be 

 weighed out directly. 



The uraninite is introduced into a flask of about 100 c.c. 

 capacity, in the neck of which bulbs are blown, and a small 

 tube also blown into bulbs is sealed to the neck to act as a 

 trap and prevent loss of the solution by bumping during 

 boiling. The flask is exhausted and a quantity of aqua-regia 

 drawn in and the neck sealed. When solution is complete the 

 seal is broken under a clamped rubber tube and a quantity of 

 water is drawn in. The emanation is extracted by connecting 

 the rubber tube to a small gas washing-bottle, to serve as a 

 trap, and a larger similar bottle, both previously exhausted 

 on the water-pump and closed. Between the flask and the 

 first bottle is a T-tube connected to a water-supply. The 

 uraninite solution is boiled vigorously on the water-bath, until 

 all gas is expelled, and it is then removed and sealed up for 

 the next test. The gas expelled is driven into the larger of 

 the two bottles by flowing in water, and in the same way 

 sent into the electroscope through the soda-lime tube, the gas 

 in the latter being displaced as before by a small additional 

 volume of air. 



This method of working is so convenient and rapid that it 

 is being gradually used in all the tests. The preparations 

 sealed to the mercury pump are, as opportunity offers, being- 

 cut down and transferred to flasks with inlet and outlet tubes 

 sealed into the neck. The tubes are kept sealed, and are 

 broken under rubber before, and resealed after, each test. 

 As already mentioned, very similar methods have been 

 described by Boltwood and Strutt. The possible objection 

 to the method, that some of the emanation may redissolve in 

 the water used, must be tolerated in view of its manifold 

 advantages over the other methods tried. The error, if any, 

 is likely to be bat small. 



The uraninite employed was pitchblende from Joachimsthal, 

 of activity 4' 2 times uranoso-uranic oxide. The activity was 

 determined by sprinkling some 10 mg. of the oxide and some 

 4 rcg. of the finely powdered mineral, on two similar metal 

 disks 2" 5 cm. in diameter, and comparing the leaks produced 

 in an electroscope. An analysis of the mineral showed the 

 presence of 53 per cent, of uranium. Each milligram of the 

 mineral therefore contains 0"53 x 0'38 X 10~ 9 =2 X 10~ llJ gram 

 of radium. The first test w T ith the solid mineral immediately 



Phil. Mag. S. 6. Vol. 14. >'o. 80. Aug. 1907. U 



