[ 143 ] 



VIII. The Magnetic Rotation of Sodium Vapour at the 

 D Lines. By R. W. TTood, Professor of Experimental 

 Physics, Johns Hopkins University *. 

 [Plate VIII.] 

 ri^HE magnetic rotation of the plane of polarization o£ 

 JL light in the vicinity or the D lines was discovered bv 

 Macaluso and Corbino, who have published numerous papers 

 on the subject. A more extended study of the phenomenon 

 was made by Hallo, working in Professor Zeeman's labora- 

 tory. In the latter investigation quantitative data were 

 obtained, which were used for a rough test of Becquerel's 

 formula. 



Previous to the appearance, at least in this country, of 

 Hallo's paper I completed a series of measurements of the 

 magnetic rotation of the vapour, which were described in the 

 Philosophical Magazine for October 1905 (Magneto-Optics 

 of Sodium Vapour). Whereas all previous investigators had 

 confined their observations to flames coloured by salts of 

 sodium, the medium investigated in this case was the non- 

 luminous vapour of metallic sodium confined in tubes of steel 

 or porcelain. The observations could be extended over a 

 much wider range of wave-lengths by employing very dense 

 vapour, and a much better test of the dispersion formula 

 made in consequence. For purposes of comparison of the 

 two methods, the statement will suffice that in Hallo's experi- 

 ments the maximum rotation observed at a point midway 

 between the D lines was only 15°, while in the work with the 

 vapour of the metal rotations as great as 1350° at this point 

 have been actually observed. 



In the paper referred to, I showed that the enormous 

 rotations of the plane of polarization in the vicinity of the 

 D lines should, if correctly determined, enable us to test in a 

 most rigorous manner any formula intended to express the 

 rotatory power of a magnetized absorbing medium. 



Though results of considerable accuracy were obtained in 

 the earlier experiments, it seemed best to repeat the work 

 under more favourable conditions, eliminating certain sources 

 of error which were suspected to exist. 



Great difficulty was found in the preliminary experiments 

 in keeping the vapour at a uniform density during the time 

 necessarv to secure a set of readings, even after the expedient 

 of electrical heating had been adopted. To remedy this 

 irouble, a small electrical oven was ordered from Heraeus of 

 Hanau, designed especially for the work. This furnace fitted 



* Communicated by the Author. 

 Phil. Mag. S. 6. Vol. 11, No. 79. July 1907.. L 



