1904.] On the Production of Helium from Radium. 351 



whole of the emanation was washed into the spiral by admission of a 

 little pure oxygen from A ; on exhaustion with the pump the gas was 

 not luminous, showing that the emanation had been almost completely 

 retained in the spiral. Mercury was then allowed to rise in the trap 

 until the bulb F was filled ; the connection to the pump was then 

 sealed at H and the spiral allowed to warm up. The emanation in the 

 vacuum tube showed a bright green spectrum, but on filling the spiral 

 with mercury and sealing off the vacuum tube, the spectrum of carbon 

 dioxide became visible ; D 3 was not seen. 



Next day this line was seen, but very feeble ; its strength increased 

 from day to day, and in 5 days the yellow, green, and two blues were 

 visible as well as the violet ; their identity was proved by means of a 

 comparison spectrum. 



Subsequent experiments were made in which the heated spiral of 

 copper was replaced by a tube containing a fragment of phosphorus ; 

 the emanation was washed out of the condensing tube by a few 

 bubbles of oxygen. The bulb of potash solution was retained, but 

 the solid potash was replaced by solid barium hydroxide. This plan 

 was not so effective in removing carbon dioxide, yet on keeping the 

 tube for 3 days, and condensing the carbon dioxide with liquid air, 

 D 3 was easily visible, although weakened by the spectrum of carbon 

 monoxide. 



On two subsequent occasions the gases evolved from both solutions 

 of radium bromide were mixed after 4 clays' accumulation, which 

 amounted to about 2*5 c.c. in each case, and were examined in a 

 similar way. In this case the non-condensable gases alone were 

 examined, the emanation being retained. Whereas with the emana- 

 tion almost the whole can be introduced into the vacuum tube, with 

 permanent gases only about one-twentieth part is available for the 

 purpose of the spectrum. The D 3 line of helium could not be 

 detected. 



The vacation now intervened, and the bulbs containing the dissolved 

 radium bromide were connected with a mercury reservoir and with a 

 gauge, so that the pressure should not rise and burst the bulbs. The 

 gas accumulated during 60 days • its composition was : Hydrogen, 

 19-48 c.c. ; oxygen, 10-37 c.c. ; nitrogen 1*02 c.c. = 30*87 c.c. 



The nitrogen was manifestly derived from leakage ; after deducting 

 one-fourth of its volume of oxygen, the remaining gas has practically 

 the composition of electrolytic gas. The rate of accumulation is about 

 J c.c. a day. 



The object of the experiment, of which an account will now be 

 given, was to form an estimate of the amount of helium produced by 

 comparing the intensity of its spectrum with that of a known quantity 

 of helium at a known pressure. 



Experiment 3. — This gas was exploded and left a residue of nitrogen 



