residual gases shrank almost to zero, in another the initial volume increased to about ten times its initial value. In the latter experiment a brilliant spectrum of helium was observed in the residual gas. This difference of behaviour is probably due to different degrees of absorption of helium by the glass tubes.
If the α particles are helium atoms, we may expect that a large proportion of the helium, which is produced in a tube containing the radium emanation, will be buried in the wall of the glass tube; for the α particles are projected with sufficient velocity to penetrate some distance into the glass. This helium may either remain in the glass, or in some cases rapidly diffuse out again. In any case, a fraction of the helium would be liberated when an intense electric discharge is passed through the tube. Ramsay and Soddy have in some instances observed that a slight amount of helium is liberated on heating the walls of the tube in which the emanation had been stored for some time.
The volume of helium produced per year by 1 gram of radium can easily be calculated on the assumption that the α particle is in reality a helium atom.
It has been shown that 2·5 × 10^{11} α particles are projected per second from 1 gram of radium. Since there are 3·6 × 10^{19} molecules in one cubic centimetre of any gas at standard pressure and temperature, the volume of the α particles released per second is 7 × 10^{-9} c.c. and per year 0·24 c.c. It has already been pointed out that, on this hypothesis, the volume of helium released by the emanation is three times the volume of the latter. The amount of helium to be obtained from the emanation released from 1 gram of radium in radio-active equilibrium is thus about 3 cubic mms.
Ramsay and Soddy have tried to estimate experimentally the probable volume of helium produced per second by one gram of radium. The helium, obtained from 50 mgrs. of radium bromide, which had been kept in solution in a closed vessel for 60 days, was introduced into a vacuum tube. Another similar tube was placed in series with it, and the amount of the helium in the latter adjusted until on passing a discharge through the two tubes in series the helium lines in each tube were of about the same brightness. In this way they calculated that the amount of helium
