﻿331 Prof. R. W. Wood on the Magnetic Rotation 



region were too Hue and close together to be completely 

 resolved, which makes it difficult to draw conclusions. In the 

 region around wave-length 5200, the bright magnetic lines 

 usually coincide with dark absorption-lines, though theie 

 appears to be no relation between their relative intensities. 

 Above wave-length 5300 the spectra at first sight appear to 

 be the same. This is only true, however, i£ we confine 

 ourselves to the general appearance o£ the fluted bands, careful 

 inspection showing that che bright lines coincide with the dark. 

 Theoretical considerations make it seem probable that all 

 of the magnetic lines are close doubles, since the rotation of 

 the plane of polarization has its maximum value on each side 

 of every absorption-line. The double lines have only been 

 observed, however, in the case of the D lines of sodium. 

 Photographs of the magnetic effects seen with sodium vapour 

 of varying density were reproduced in connexion with an 

 earlier paper (" Magneto-Optics of Sodium Vapour "*"). It is 

 clear that two absorption-lines very close together may under 

 certain conditions give rise to a magnetic triplet, the rotation 

 midway between the absorption-lines having a value of 90°, 

 with a similar condition just above one absorption-line, and 

 just belowthe other. I have been unable to find any evidence 

 of such an effect in the iodine spectrum, though it can be 

 observed without difficulty at the I) lines, the components of 

 the triplet being very broad in this case. With very dense 

 sodium vapour, as I have already shown, we may have a 

 multiplex magnetic line, the rotations at the components 

 being 90°, 270°, 360°, 450°, &c. In the case of iodine, and 

 also in the case of sodium in the region of the channelled 

 absorption spectrum, the magnetic lines appear single with all 

 resolving powers used up to the present time. Possibly the 

 employment of an echelon or some such instrument would 

 reveal their duplicity. The line 5001 of the magnetic spectrum 

 of sodium would be the best one to begin with, as it is isolated 

 and fairly bright. If the nicols were set parallel, instead 

 of crossed, it is possible that the appearance of the lines in the 

 absorption spectrum would be altered. We should expect the 

 lines to appear doubled, with two fainter and very close com- 

 ponents between them. Probably, however, this system would 

 not be resolved with a 12-foot grating, and we should notice only 

 a slight broadening of the line. Bromine gives a fairly bright 

 magnetic-iota tion spectrum, which has been photographed 

 with the three-prism spectrography, but not with the grating. 

 It does not, how ever, exhibit the slightest trace of fluorescence 

 under any conditions. 



* Phil. Mag. [6] x. p. 408. 



