305 
in producing Electric Conductivity in Gases. 
produced at most of the collisions. The current when the wire is 
positive increases rapidly at the higher voltages. 
The experiments at this pressure and at lower pressures would 
indicate that a smaller voltage is sufficient to produce ionisation 
by the collisions of the negative ion than what is required to give 
the necessary velocity to the positive. But they show that the 
positive ion we are dealing with can be given the velocity necessary 
to produce ionisation. 
7. In experimenting with the ionisation produced by an 
incandescent wire in a gas at low pressure, there are many in- 
dications that we are dealing not only with the ions produced 
from the molecules of the gas but that we are actually getting 
ions off from the hot wire itself. 
When the gas is at atmospheric pressure the number of 
positive ions is in excess of the number of negative except when 
the temperature of the wire is very high, when the amount of 
positive and negative is approximately the same. When the 
pressure is reduced the negative ions are greatly in excess, even at 
moderately high temperatures. Using the apparatus described, 
the current when the wire is negative may be 50 times what it is 
when the wire is positive, the pressure being 1 mm. or less. Such 
a difference from what is observed at atmospheric pressure sug- 
gests that at low pressure we get a copious supply of negative 
ions from the wire itself. 
Again, the current when the wire is negative varies little with 
the pressure while the pressure is very small ; for example, the 
current was practically constant when the pressure was varied 
from gV mm. to mm * This suggests that at such pressures 
the ions resulting from the ionisation of the gas molecules were 
small in number compared with those coming from the wire. 
Possibly the positive ions which are active in producing 
secondary ionisation also come from the wire, which may account 
for the apparent difference between them and the positive ions 
investigated in other cases of ionisation. 
However, more experiments are required before entering into 
any further discussion of this nature. The wire may produce 
ionisation at considerable distance from itself, in which case even 
when the wire is positive, the negative ions would travel for some 
distance through the gas and in this way produce secondary 
ionisation, but this action could scarcely explain the effects 
obtained. 
It is intended to investigate more completely some of the 
effects described in the paper. I desire to thank Prof. Thomson 
for kind advice during the course of these experiments, which were 
carried out at the Cavendish Laboratory some time ago. 
