156 RATE OF EMISSION OF ENERGY [CH. 
hours afterwards the activity, measured by the a rays, reached a 
minimum corresponding to 25°/, of its maximum value when in a 
state of radio-active equilibrium (see section 191). The saturation 
current between parallel plates, sufficiently far apart to absorb all 
the a rays in the gas between them, was measured by a galvano- 
meter and found to be 26 x 10-° amperes. In this case the film 
of radium bromide was so thin that the absorption of the a rays by 
the radium itself was very small. Taking into account that half of 
the a radiation from the radium was absorbed in the plate, it can 
readily be deduced that the total current corresponding to 1 gram 
of radium when in a state of radio-active equilibrium is equal to 
1:2 x 10~ electromagnetic units. Taking the charge on each ion 
as 1:13 x 10-” electromagnetic units, this corresponds to the pro- 
duction of 10" ions per second per gram. 
Langevin? has deduced from the results of Townsend on ioniza- 
tion by collision, that the energy required to produce fresh ions at 
every collision is equal to the energy acquired by an ion moving 
freely between two points, which differ in potential by about 60 
volts. This corresponds to an amount of energy of 6°8 x 10 ergs. 
The total rate of emission of energy on the production of 101° ions 
per second is thus 7 x 10° ergs per second or about 60 gram- 
calories per hour. 
Method 3. The ionization produced in the gas by the pro- 
jected a particles is due to collision with the neutral molecules. 
The maximum number of ions produced per unit length of path 
will be reached when each collision results in the production of 
fresh ions. Now Townsend? has shown that the maximum number 
of ions produced by a moving electron per cm. of its path in air at 
the pressure of 1 mm. of mercury is 21. On the kinetic theory of 
gases, it can be deduced from this result (Langevin, loc. cit.) that 
the electron ionizes every molecule in a circular cylinder whose 
axis is the direction of movement and whose diameter is equal to 
the diameter of the molecule. It follows that the electron must 
be of dimensions small compared with the molecule—a result which 
is in accordance with the experimental data. In the case of the 
1 Rutherford and Soddy, Phil. Mag. May 1903. 
2 These présentée & la Faculté des Sciences, Paris 1902, p. 85. 
° Phil. Mag. p. 198, Feb. 1901. 
