428 
MK. w. j; cai;k ox the laws goverxixci 
the same, 350 volts, for all pressures. But it will be seen that he obtained, for 
pressures up to two atmospheres, values varying from 325 to 370 volts. This 
discrepancy, however, though marked, is not large and possibly is within the range 
of experimental error. 
The results which Eaehaet obtained for spark lengths shorter than 5 micra cannot 
in any way affect the validity of this explanation, for he has shown without doubt 
that the discharges in this range are governed by a law which does not apply to the 
o'as under ordinarv circumstances. 
O kJ 
TX. Spark Potentials in Different Gases. 
In a paper on the cathode fall of potential in gases, by Capstick,^ an attempt has 
been made to show that the cathode fall in a compound gas is related to the cathode 
falls in the elementaiy gases of wliich it is composed by a simple additive law. 
Experiments were made witli hydrogen, oxygen, nitrogen, ammonia gas, and water 
^'apour. The results, though not conclusive, were yet of sufficient weight to lead 
the author to the observation that the cathode fall is approximately an additive 
quantity and is probably a ])i‘operty of the atom rather tlian the molecule of a gas. 
Owing to the difficulties experienced by Capstick and others in overcoming the 
intermittence of the current in tiie case of compound gases, the effort to extend his 
investigations to compound gases other than those mentioned was abandoned, and the 
(jiiestion up to the present time has remained unsettled. 
As already pointed out in tins papei‘, experimental evidence has been brought 
forward by Steutt to show that the minimum spark potential should be equal to the 
cathode fall measured in the same gas. In view of this conclusion it seemed desirable 
to extend the experiments described in the first part of this paper to include a larger 
number of compound gases, in order to throw light, if possible, on the question raised 
by Capstick. Measurements were tlierefore carried out with the gases oxygen, 
nitrous oxide, hydrogen sul])liide, sul])hur dioxide and acetylene. A constant spark 
length of 3 millims. was used throughout in order that a direct comparison could be 
made between the results olitained with these gases and those already recorded for 
hydrogen and carljon dioxide. The results obtained with all the gases, using this 
spark length, are recorded in Table VIL, and curves corresponding to the readings at 
the critical and lower pressures are sliown in fig. 11. 
All the curves present the general characteristic of a minimum s})ark potential, 
followed, at lower pressures, by rapidly increasing spark potentials. It Avill be seen, 
also, that the critical pressures and the minimum spark potentials vaiy Avith the 
different gases. The cuiwes, too, cut each (fiher in regular order, and at the 
loAvest pressures their relatiA^e arrangement Avith regard to the ordinate axis is 
practically the liiAmrse of that assumed ly them Avith reference to the abscissa axis 
aboAm the critical ])ressures. 
* Capstick, ‘Roy. Soc. Proc.,’ a'oI. 63, p. 356. 
