300 Mr McClelland , On the Action of Incandescent Metals 
pressure of the gas is atmospheric 1 , but after a certain potential 
difference between the terminals has been reached the current 
increases rapidly for further increases of potential. The curve 
shows that the current at first increases proportionately with the 
e.m.f., then shows signs of reaching a maximum and afterwards 
increases rapidly. 
This second rapid increase of current can easily be explained 
if we assume that the negative ions travelling from the negatively 
charged wire to the cylinder C have the power of producing ion- 
isation when they collide with molecules, provided the velocity of 
the ion is sufficiently great. In the case we are dealing with the 
E.M.F. has been sufficient to produce such a velocity in the air at a 
pressure of § mm. When the potential difference is sufficiently 
great to produce the required velocity the secondary ionisation 
begins and then we have the rapid increase of current. At a 
pressure of § mm. the negative ion has not the large mass 
travelling with it that it has at atmospheric pressure and again 
it is moving in a rare medium, so that it acquires a high velocity. 
Prof. Townsend 2 has used the above theory to explain similar 
results obtained with the ionisation produced by Rontgen radia- 
tion in a gas at low pressure. The numbers given above refer to 
an experiment in air at a pressure of | mm. We have selected 
the results at that pressure because the secondary ionisation is 
then more marked. At higher pressures the ions do not so easily 
attain the necessary velocity, and at very much lower pressures 
the secondary ionisation is smaller compared with the initial 
ionisation produced direct by the wire. 
If we use a smaller current to heat the wire the secondary 
ionisation is even more apparent than in the curve given above. 
The negative ions have then a greater velocity, as we have seen 
above. 
The following numbers refer to such an experiment, the wire 
being at a lower temperature and a much smaller capacity joined 
up to the electrometer. 
Potential of the Current in arbitrary 
wire in volts units 
40 12 
80 25 
120 45 
160 91 
200 145 
280 430 
1 Proceedings Cambridge Phil. Soc. Vol. x. Pt. iv. 
2 Phil. Mag. Feb. 1901. 
