336 Prof. Nagaoka and Mr. T. Takamine on Anomalous 



The appearance of satellites was feebler in low than in 

 high fields, so that the time of exposure varied with the 

 strength of the field and the intensity of lines. 



To clear up some doubts as regards the existence as well 

 as the order of spectrum of lines, we interposed the Lummer- 

 Gehrcke plate or the Fabry-Perot air-plate before the photo- 

 graphic objective, and obtained crossed spectra, which in 

 nearly all cases decided the question, provided the lines 

 w r ere sufficiently strong. The interference points for the 

 green line obtained by crossing the echelon with the Lummer- 

 Gehrcke plate are shown in PI. V. fig. 1, and those by com- 

 bining the echelon with the Fabry-Perot plate in PL Y. 

 fig. 2. 



In examining the Zeeman effect of the satellites, a series 

 of photographs in successive stages of magnetic fields were 

 taken with the echelon grating, in which the principal line 

 appeared either in two consecutive orders of spectra or as a 

 single line. One * of us has already discussed the separation 

 of the green line into a nonet, and its relation with the 

 strength of the field, and found that the change in wave- 

 length is proportional to it in all the different branches. By 



applying Runge's rule, the value of — is 1*780 x 10 7 e.m.u., 



which differs from that found by Cotton and Weiss for the 

 same quantity (=1*771 x 10 7 ) only by 0*5 per cent. Instead 

 of measuring the field separately, it was deduced indirectly 

 from the amount of separation of the principal lines. 



The numerous photographs of the lines were examined 

 micrometrically, and the position of the satellites in different 

 fields plotted on a section paper, and the successive points 

 joined by continuous lines. The process, though somewhat 

 tedious, gave a rough delineation of the courses taken by dif- 

 ferent satellites, but it was not entirely free from ambiguity as 

 to which branch of the separated satellite the points belong to, 

 especially when different types of separation appeared, as in 

 the satellites of the green line of mercury. 



In addition to the photographs taken in uniform fields, it 

 was thought desirable to have them also in heterogeneous 

 fields, so that we can easily follow the course of the curve 

 taken by lines of satellites in a continuous manner, the field 

 strength being measured by the amount of separation in 

 corresponding places, as already described. Evidently the 

 points found from photographs in different uniform fields 

 must agree with the course indicated by the photographs in 

 heterogeneous fields ; by following these two processes of 

 * Nagaolia, Phys. Zeit. xi. p. 789 (1910). 



