202 Scientific Proceedings, Royal Dublin Society. 



Since every motion of electricity produces an electro-magnetic 

 disturbance in the surrounding aether, the motion of the electron in 

 any one of these partials would give rise to a line in the spectrum, 

 and in this way the successive partials furnish a definite series of 

 lines. 



There is one case in which a further analysis is practicable, and 

 the Paper deals especially with this case. It is where the forces 

 that determine the path of the electron are some of them strong 

 and some feeble. We may then regard these latter as perturbating 

 forces ; and, distinguishing the undisturbed orbit — that which the 

 electron would have pursued if the strong forces alone had acted 

 on it — from the actual orbit, we can represent the latter by 

 applying perturbations to the undisturbed orbit. The character 

 of the perturbations likely to occur, is known from the elaborate 

 investigations which astronomers have made into the cause of the 

 motions that go on within the solar system, and one of the most 

 familiar of them is an apsidal motion affecting the elliptic partials 

 into which the undisturbed orbit may be resolved. The undis- 

 turbed orbit will in general be such as would give rise to some 

 definite series of single lines in the spectrum ; and it is shown in 

 the Paper that the consequence of some or all of its partials being 

 exposed to the above-mentioned perturbation is to cause the corre- 

 sponding lines of the series to become double. 



In working out the details of the investigation it is shown 

 that, from observations on the positions and intensities of the con- 

 stituents of each pair of double lines, full information may be 

 elicited about the form of the elliptic partial from which they 

 arise, the 2^eriodic time of the motion of the electron in it, and the 

 direction and jjeriodic time of the apsidal perturbation affecting it. 

 In this way it is ascertained from observations on the principal 

 double line of sodium, that which gives rise to the D lines in the 

 solar spectrum, that it is caused by (or, at all events, is due to an 

 event which follows the same mathematical laws as) the motion of an 

 electron within each molecule of the vapour, in an ellipse of which 

 the axes have to one another a ratio lying somewhere between 11 

 to 1 and 13 to 1 ; round which ellipse the electron revolves 169"637 

 times in each jot^ of time ; the ellipse meanwhile being slowly 



^ The jot of time is the time that light takes to advance one tenth of a millimetre 

 in vacuo. It is about the third of the millionth of the millionth of a second. 



