Part Il,— Magnetic Perturbations of the Spectral Lines. 17 
case depicted in fig. 7, although, as I have said before, my observations do not 
confirm the existence of this case. 
The supposition made above to account for the doubling of the middle line, 
viz. that the amplitude of the z component vibration varies periodically, is one 
which appears to be justified when we consider the nature of the moving system 
and the forces which control it. For the revolving ion is part of some more or 
less complex system which must set in some definite way under the action of the 
magnetic field—say with its axis along the direction of the magnetic foree—and, 
in coming into their position, the inertia of the system will cause it to vibrate 
with small oscillations about the position of equilibrium, and this vibration 
superposed on the precessional motion of the ionic orbit gives the motion 
postulated above to explain the quartet. 
This, indeed, comes to the same thing, as a suggestion made by Prof. G. F. 
Fitz Gerald about a year ago—shortly after I discovered the existence of the 
quartet form (October, 1897). In Prof. FitzGerald’s view, the ion revolving 
in its orbit is equivalent to an electric current round the orbit, and therefore the 
revolving ion and the matter with which it is associated behaves as a little 
magnet, having its axis perpendicular to the plane of the orbit. The action of 
the magnetic field will be to set the axis of this magnet along the lines of force, 
and, in taking up this set, the ionic orbit will vibrate about its position of 
equilibrium, just as an ordinary magnet vibrates about its position of rest under 
the Earth’s magnetic force. 
In a similar way a periodic change in the ellipticity of the orbit produces a 
doubling of the spectral lines previously existing, while a periodic oscillation 
in the apsidal motion renders the lines nebulous or diffuse ; and by treating these 
cases in the foregoing manner the corresponding forces may be discovered. 
It is clear, therefore, that perturbations of this kind are sufficient to account 
for all the observed phenomena, and that the theory is ready to meet the 
demands of more complicated types than have yet been observed. It is 
legitimate to expect that perturbations of this kind will occur in some, at 
least, of the ionic motions, and, in fact, that they must occur, and that the 
perfect freedom required for the production of the pure precessional triplet 
cannot exist in all cases. 
The existence of all these variations of the normal triplet is a matter of great 
Interest, not only as showing that the perfect uniformity required for the 
production of the normal triplet is not maintained in all cases, but also as giving 
us a further insight into the nature of the conditions under which the ionic 
motions take place, as well as demonstrating that the causes supposed by 
Dr. Stoney, in 1881, to be operative in producing doublets and satellites in the 
natural spectra of gases may be really the true causes by which they are produced. 
E 2 
