The Discharge of Electricity through Gases. 549 



consequence are quite sufficient to cause a decided bending, even 

 without any rotation of the disc. 



Magnitude of some of the Quantities involved in the Discharge. 



As it is necessary to bear in mind the order of magnitude of some 

 of the quantities involved in the discharge, I may briefly note some 

 of the most important ones. 



The most probable value for the charge carried by each atom of 

 hydrogen I find to be 3 X 10 -23 electro-magnetic units, more generally, 

 say, 3k X 10 -23 where k is a numerical constant. 



Question I : — How does the energy acquired by an ion between 

 two impacts compare with the average kinetic energy of a molecule ? 



Answer : — The ratio e/ra for hydrogen is known to be 10 4 approxi- 

 mately. If a particle of mass m carries a charge e, the velocity 

 generated from rest through a range in which the difference of 

 potential is V is 



or 140 v/V. 



V w- 



If V is one volt, this would be equal to 14 X I0 5 ; a quantity 

 nearly ten times as great as the mean velocity of a hydrogen 

 molecule. In the positive part of the discharge when the velocity of 

 diffusion is uniform, the fall of potential in my experiments was, 

 roughly speaking, about 1 volt per millimeter; the mean free path 

 calculated according to the kinetic theory varied between \ mm. 

 and \ mm. Hence, on the average, the velocity generated in the 

 atom by electric forces between two encounters exceeds several 

 times the mean velocity in the stationary state. It will appear that 

 the number of atoms carrying charges is small compared to the total 

 number ; so that the actually observed rise in temperature need not 

 be considerable. At each impact the atom must give up, on the 

 average, that proportion of its own velocity which it gains during 

 two encounters; and the above numbers show that the energy 

 communicated is very considerable. Hence the luminosity of the 

 positive discharge. Even if by an impact the ions are thrown back, 

 the electromotive force will, in general, be strong enough to reduce 

 it to rest, and send it forward before the next impact. The path of 

 particles will therefore not be straight, and the velocity of the ions 

 before impact will almost entirely be in the direction in which the 

 force acts. 



Question II : — If the molecules, each charged with a quantity of 

 electricity e, approach each other with a velocity equal to the mean 

 speed in a homogeneous gas at ordinary temperature, at what 

 distance from each other will they come to rest ? 



