Velocities of Two Groups of Rays and their Ahsorjjtion Coefficients. 77 
Sulphur dioxide. — From Johnson's " Pure Sodium Sulphite " and H.jSO^. 
Dried HoSO^. 
The gas flowed gently through the chamber and into a large vessel of 
about 10 litres capacity connected to its exit. This arrangement was to 
ensure that the gas in the chamber should be steadily replenished, but that the 
pressure within should not vary by more than a few millimetres of mercury 
during a run. The mean pressure was recorded. 
The ordinates of the two curves for a given gas in Fig. 2 have been sul)- 
tracted, giving the curves shown in Fig. 3. These curves represent the total 
ionisation produced in the gas at various pressures by the cathode particles 
from the gold leaf backing the chamber when silver X rays fall upon it.* 
At this point it would not be out of place to indicate the manner in which 
the absorption coefficients have been obtained from the "cathode ionisation " 
curves in Fig. 3, as in the latter part of the paper the exact process has to 
be kept in mind. 
Consider an instantaneous distribution of ions represented by Fig. 4 h. 
There is a critical pressure P at which, for a given initial speed of ejection, 
the ionising property of the least scattered cathode particle persists until the 
other side of the chamber is just reached. For this pressure let po the 
number of ions present per c.c. in a thin slab of gas next to the gold leaf. 
JSText, assuming that the number of ions produced per c.c. from any cause be 
proportional to the density of the gas, then at a pressure 2P the density of 
ions at the gold radiator will be '2po, but now the maximum distance traversed 
by the least scattered particles will be ^/2. In other words, doubling the 
pressure halves the linear dimensions of the ionised track of each electron 
without altering the shape of the track, whatever that may be. The con- 
dition is represented by Fig. 4 a. Coming now to the case in which the 
P / 
pressure is -^/2, the most normally directed electrons would ionise, were 
they not absorbed by the front face of the chamber, up to a maximum dis- 
tance of 2t from the gold leaf, and the density of the ions at the gold leaf 
— ^ 7'-- The areas of all the curves are equal. 
Reference to the cathode ionisation curves will show that condition {h) 
is represented by the point B on the air curve, {a) by A and (c) by C. In 
* More strictly it is the ionisation by the cathode particles from the gold, less 
that by those emerging from a sheet of waxed paper, plus the ionisation in the gas 
by the tertiary X rays from gold and all the effects due to it. The lower set of curves 
in Fig. 2 shows that the ionisation in the gas due to the cathode particles emerging- 
from a sheet of waxed paper can be neglected. If any did exist their properties would 
not differ greatly from those from gold. The ionisation effect due to any tertiary 
X rays from the gold leaf would be proportional to the pressure of the gas, and, as the 
curves in Fig. 3 show no variation in the ordinate beyond a certain critical pressure, 
this effect can also be neglected. 
