1018 Prof. Wood and M. L. Dunoyer on Separate Excitation 



the band of absorption, but merely an alteration in velocities 

 of right and left handed circular vibrations. 



(2) Rotations in which the sign is the same on opposite sides 

 of the absorption band, as at the D lines of sodium and the 

 iodine lines. Rotations of this type are explained by the 

 division of the line into a Zeeman doublet by the magnetic field. 



It seems probable, therefore, that there is a small Zeeman 

 effect for the iodine, but it is doubtful if it can ever be 

 detected as it is of the order of magnitude of the width of 

 the lines, probably much less in fact. 



The study of the magnetic rotation of the vapour of 

 sodium by the improved methods outlined in the present 

 paper will undoubtedly give much more satisfactory results, 

 as the rotatory power of this vapour in the red and orange is 

 certainly ten times as great as that of iodine. 



Ecole Xornial Superieur, 

 Paris. 



CVIII. The Separate Excitation of the Centres of Emission 

 of the D Lines of Sodium. By R. W. Wood and L. 

 Dunoter *. 



[Plate XVII. figs. 1-7.] 



THE experiments of one of us on the excitation of metallic 

 vapours by monochromatic light have shown that the 

 centres of emission of many spectrum lines are probably in 

 some sort of mechanical or electrical connexion. For example, 

 the excitation of mercury vapour by the light of the cadmium 

 spark showed that the vapour emitted the ultra-violet line 

 of wave-length 2536 when stimulated by light of wave-length 

 shorter than any given in the tables at the time, i. e, less 

 than 2000. In the case of the resonance spectra of sodium 

 and iodine we have innumerable examples of associated lines 

 which spring into existence when the vapour is excited by 

 light of frequency synchronous with one of them. 



In a paper on the Resonance Radiation of Sodium 

 Vapour, published by one of us in 1905 t, it was shown 

 that the vapour of metallic sodium in an exhausted glass 

 tube emitted its characteristic D-line radiation when the 

 image of a sodium flame was thrown upon it by means of a 

 large condensing lens, a cone of yellow light marking the 

 path of the exciting rays through the vapour. If the vapour 

 density was increased the luminosity was restricted to " a 

 thin skin of yellow light which lined the inner wall of the 

 tube/' owing to the failure of the exciting radiations to 



* Communicated bv the Authors. 

 t Wood, Phil. Mag. Nov. 1905. 



