118 MAGNETISATION OF LIGHT [CH. XL 



It must be observed that the light emitted by the 

 oscillation-components, above spoken of, will be all 

 of one definite kind, due to vibrations in one definite 

 direction, and will therefore be polarised. The kind 

 of polarisation must depend on the aspect from which 

 the light is seen. If seen at right angles to the axis 

 of precession, all three lines should be plane polar- 

 ised the middle line at right angles to the other 

 two. If, however, it be looked at along the axis of 

 precession, then there should be no middle line ; 

 because the axial vibration would then be end- on, in 

 which direction it produces no optical effect ; and the 

 two side lines would be circularly polarised. 



Fig. 15 consists of diagrams illustrative of the 

 changes caused in a spectrum line by application of a 

 powerful magnetic field to the source of radiation. 



/ represents a specially simple case. The cad- 

 mium line A, seen by rays travelling along the 

 lines of force, resolves itself into two lines B and C 

 which are circularly polarised in opposite directions. 

 This is due to the acceleration of one circular 

 component of the rectilinear or elliptical vibration 

 and the equal retardation of the other component. 



// represents the same simple line seen by 

 light travelling across the lines of force. In that 

 case the line becomes triple ; and if A had been 

 plane polarised, B and C are polarised in a plane 

 at right angles to that of A'. This is due to a 

 precessional movement of the plane of the orbital 

 motion, the axial vibration continuing unchanged, 

 and the two at right angles being one accelerated 

 and the other retarded. 



/// and III1 represent the effect of a magnetic 

 field, applied to a sodium source, on the con- 

 stituents of the yellow sodium double-line. D x is 



