The Discharge of Electricity through Gases. 54T 



Nature of particle. Telocity of mean square. 



Hydrogen atom 26 X 10 4 



Hydrogen molecule 18 x 10 4 



Nitrogen atom 7 x 10 4 



Mtrogen molecule 5 X 10 4 



As we are only dealing now with the order of magnitude, the 

 temperature need not be taken into account, and we may take v to be 

 10 5 ; we thus obtain 



— > 10 s . 



m 



The actual value of e/ra is 10 4 for hydrogen and 0*7 X 10 3 for 

 nitrogen, if we imagine each atom of nitrogen to carry the same 

 charge as the atom of hydrogen in water ; but, as nitrogen may unite 

 with these atoms of nitrogen we must assume three charges at least 

 to be carried, which would make e/m equal to 2 X 10 3 . 



It thus appears that there is nothing in the actual facts which is 

 in any way not in harmony with the theory. The lower limit for e/m 

 comes very near the actually observed values, and it is not astonishing 

 that the upper limit yields so great a value. It will be seen that in 

 equation (3) the radius of curvature enters as the square. I think I 

 may take the experiments hitherto recorded as a confirmation of the 

 theory. Assuming the theory to be correct, they show that in the 

 glow the particles are quickly reduced to a velocity of the same order 

 of magnitude as the mean velocity in an unelectrified medium. If 

 the particles do not carry fixed charges, they must become electrified 

 by contact at the electrodes. This is the view generally taken, and it 

 is interesting to trace its consequences. The change e of a sphere 

 touching a plane charged with surface density a is* given by 



e = § A 3 *?- = 20 a 2 <r approximately. 



Substituting for a the highest value which I have obtained in my 

 experiments, about 2*5 electrostatic units, e would be numerically 

 equal to 50a 2 . If for a 2 we take about 5 X 10~ ]0 , which is the 

 molecular range obtained from experiments in gases, we should be 

 above the mark. This, when substituted in the above, gives for e 

 the value 10~ 17 in electrostatic units, instead of 10~ 12 , that is, the 

 charge would be about 100,000 times less than according to our 

 theory. Applying the equation (4), I calculate that, according to' 

 the hypothesis of electrification by contact, the average velocity of 

 the molecules would only have been 2 cm. a second, which is a reductio 

 ad absurdum. 



* Maxwell, vol. 1, p. 257. 



