ELECTRIC DISCHARC4ES IN CASES AT LOW PRESSURES. 
423 
of experiments on capillary phenomena, is about 'Oo micron. Reinold and Rucker* 
found that the range of unstable thickness of a film began somewhere between '096 
and -045 micron. A value of the same order is given l)y Plateau, f who fixes the 
superior limit of the radius of the sphere of molecular action at CIS micron. 
Results of a higher order, however, were obtained by Muller-Erzbach| and 
Kayser.§ The former of these made experiments on the thickness of water and 
carbon bisulphide films, and finally concluded that the radius of the sphere of 
molecular action is at least 1’5 micron. Kayser, experimenting on condensation of 
gases on glass threads, fixed the range of molecular action at from 2 to 3 micra. 
Now the distance between the electrodes when the air film is reduced to the two 
surface layers is equal to the diameter of the sphere of molecular action, and there is 
thus strong experimental evidence from the data given above to support our adopting 
Earhart’s value of 5 micra for the smallest length to Avhich we can legitimately 
apply Paschen’s law. 
The experiments described in this pajjer have been made with a view to finding the 
relation between spark potentials and corresponding pressures for a constant spark 
length in air and other gases, but, as all the results for different spark lengths are 
connected by Paschen’s law, it Is easy to deduce curves, for any gas, expressing the 
relation between potential differences and corresponding spark lengths at selected 
pressures. Such curves, for air, deduced from those exhibited In figs. 2 and 8, have 
been plotted for a series of different pressures and are shown in fig. 9. 
Tt will be seen that these curves present a number of points of special Interest. 
* Reinold and Rucker, ‘Phil. Trans.,’ vol. 177, Part IL, p. 684, 1886. 
t Plateau, ‘Statique des Liquides,’ 1873, vol. 1, p. 210. 
I Wuller-Erzbach, ‘Wied. Ann.,’ vol. 28, p. 696, 1886. 
^ Kayser, ‘Wied. Ann.,’ vol. 14, p. 468, 1881. 
