594 Profs. B. B. Boltwood and E. Rutherford on 



millimetre of the closed end. Under these conditions, by a 

 long series of comparisons, it was possible to correct the 

 readings in E and e for any slight differences due to small 

 irregularities in the bore of E and e. It was determined, in 

 this manner, that when the tube E contained gas, the pressure 

 of the gas, indicated by the relative heights of the mercury 

 columns in E and <?, could be accurately estimated to within 

 less than 0*5 mm. of mercury. 



Helium from Radium Salt, 



(First Determination.) 



The barium-radium salt (page 589) still containing a portion 

 of its water of crystallization was sealed up in the tube de- 

 scribed on page 590 (fig. 1) for eighty-three days. A small 

 bulb containing P 2 0, 5 and another containing solid KOH 

 were sealed on to the apparatus (fig. 2) at the point 0. The 

 tube b (fig. 1) was attached to the KOH tube, the radium 

 tube a (fig. 1) was inserted in this, and the entire apparatus 

 was evacuated down to the joint of a and b (rig. 1). Stop- 

 cocks /• and /were closed. The capillary tip of the radium 

 tube was broken off in the joint b and the gases in the radium 

 tube were pumped out and collected at D (fig. 2). During 

 the pumping process, the lower end of the radium tube and 

 the platinum capsule were heated to a red heat. The water 

 of crystallization remaining in the salt was driven out and 

 condensed in the KOH and P 2 O r , bulbs. The radium salt 

 was heated for about thirty minutes. A small amount of 

 electrolytic oxygen was added to the gaseous mixture, the 

 gases were introduced into B (fig. 2) through the tube L 

 and sparks were passed between the electrodes. The residual 

 gases were then slowly pumped over into the tube A, through 

 the tube U which was immersed in liquid air. The radium 

 emanation in the gaseous mixture was condensed in the 

 tube XL When the gas had been completely transferred 

 into the tube A, the charcoal tube C was dipped into liquid 

 air, the level of the mercury in A being at the point a. 

 After an interval of about twenty minutes, the mercury was 

 raised in A and the volume and pressure of the gas trapped 

 in E was measured. The level of the mercury in A was now 

 lowered to below the opening of U into A, and by means of 

 the transfer pump H, the uncondensed gases in C and its 

 connexions were pumped out and introduced into A through 

 U. With the stop-cocks b and d closed and the stop-cock a 

 opened, the liquid air was removed from the charcoal tube, 

 this was heated, and the gases previously condensed in 



