Stoney — Cause of Double Lines in Spectra. 583 



1874, and printed in the Scientific Proceedings of the Royal Dublin Society of 

 February, 1881, and in the Philosophical Magazine for May, 1881 (see pp. 385 and 

 386 of the latter). It is there shown that the amount of this very remarkable 

 quantity of electricity is about the twentieth et (that is, 1/10"°) of the usual electro- 

 magnetic unit of electricity, i.e. the unit of the ohm series. This is the same as 

 three- eleven thets (3/10'^) of the much smaller G.G.S. electrostatic unit of quantity. 

 A charge of this amount is associated in the chemical atom with each bond. There 

 may accordingly be several such charges in one chemical atom, and there appear 

 to be at least two in each atom. These charges, which it will be convenient to 

 call electrons, cannot be removed from the atom; but they become disguised when 

 atoms chemically unite. If an electron be lodged at the point P of the molecule, 

 which undergoes the motion described in the last chapter, the revolution of this 

 charge will cause an electro-magnetic undulation in the surrounding aether. The 

 only change that has to be made in our investigation to adapt it to this state 

 of things is to change 6t into [dt — 73-/2), i.e. a mere change of phase. We, in this 

 way, represent the fact that it is the tangential direction and velocity of the 

 motion of P, not the direction and length of its radius vector, which determine the 

 direction and intensity of the electro-magnetic stresses in the surrounding sether. 

 "We have further to correct for the change of phase (about one-fourth of a vibration 

 period) consequent upon what takes place in the immediate vicinity of the moving 

 charge. 



Within the molecule itself the oscillation of the permanent charge probably 

 causes electric displacements in other parts of the molecule ; and it is possible that 

 it is to the reaction of these upon the oscillating charge that we are to attribute 

 those perturbations of which the double lines in the spectrum give evidence. 

 They obviously may, however, have some other source. 



Beside the irremoveable electric charges which electrolysis has brought to 

 light, and which establish the fact that some parts of the molecule behave as 

 perfect non-conductors, there may presumably be temporary charges in such 

 other parts of the molecule as conduct. This probably happens by direct electri- 

 fication of the molecule when the luminous condition of the gas is produced by 

 the passage of an electric current through it, and it would seem that it must also 

 be brought about indirectly in cases of combustion, owing to the combinations and 

 decompositions which then occur during which some of the permanent charges 

 become disguised or cease to be disguised ; in either case having the effect of 

 charging the molecule with free electricity, positive or negative. 



Now, molecules whether electrified in these ways, or by the motions set up 

 within the molecule developing electricity as in an influence machine, must 

 be expected to discharge into one another when they collide, and hence will arise 



