of Radium from Uranium. Ill 



I. — Preliminary experiments with a hundred grams of 

 commercial uranium nitrate showed evidence of the presence 

 of radium in noticeable amount. It was found that the 

 radium could be removed practically completely by repeated 

 precipitation in its aqueous solution of small quantities of 

 barium in the form of sulphate. The general method of 

 testing for the presence of radium may be briefly described. 

 The solution of uranium nitrate in water was kept for a period 

 of at least some days in a closed bottle, into the mouth of 

 which inlet and outlet tubes provided with taps had been 

 ground. At the end of the period, which was accurately 

 noted, a current of air was blown through the uranium 

 solution, through a drying-tube containing calcium chloride 

 between plugs of cotton-wool, into the electroscope. A 

 volume of air sufficient to displace the whole of the air in 

 the bottle, but smaller than the volume of the electroscope, 

 was used. The electroscope was of the ordinary single-leaf 

 tvpe insulated by a sulphur bead, first described by 0. T. R. 

 Wilson, and the rate of discharge of the leaf was read by a 

 microscope provided with a scale in the eyepiece. The 

 natural leak of the instrument was always determined at the 

 beginning of each experiment. The somewhat large natural 

 rates of leak in the final series of measurements are due to 

 the gradual accumulation of the later products of slow change 

 left behind by the emanation. Rutherford has shown that 

 this effect increases rather than decays with lapse of time. 

 As, however, in every case the natural leak was small com- 

 pared with the rate of leak measured, and as it remained 

 constant throughout the day's experiments, it did not introduce 

 any error into the measurements. 



A kilogram of commercial uranium nitrate was dissolved 

 in water, sulphuric acid was added and then successive small 

 quantities of barium-nitrate solution, the liquid being left to 

 stand, and filtered from the precipitated barium sulphate after 

 each addition. The most effective conditions for the remoA'al 

 of the radium appeared to be to carry out the addition of 

 barium drop by drop to the cold dilute solution of uranium 

 nitrate, with constant stirring. Preliminary experiments 

 indicated that the radium had been practically all removed, 

 and on June 13th, after a final precipitation, the uranium 

 solution was closed up in the bottle, and except for occasional 

 tests, in which air was blown through, the solution remained 

 undisturbed throughout the w T hole series of experiments. 

 The initial observation was taken on June 20th, 1903, seven 

 days after the closing of the bottle. The reading-microscope 



