﻿242 Prof. Najvaoka and Mr. T. Takamine : Anomalous 



and \4359 in a field of 15,500 gauss. Figs. 2 a, 2 b, 2 c are 

 enlarged photograms o£ the p- and s-components taken in 

 fields of 6300, 14,600, and 20,300 gauss respectively. The 

 general appearance of the lines in heterogeneous fields is 

 shown in fig. 3 a and fig. 3b, of which the latter is more 

 enlarged than the former. Faint lines are not distinctly 

 seen in photographic reproductions, so that these figures 

 only serve to illustrate the complex appearance of the satellites 

 with reference to the principal lines. 



Principal Line. — It is generally assumed that the so-called 

 principal line is simple and has finite breadth ; it was on 

 this supposition that we have measured the Zeeman effect 

 of satellites belonging to the lines \ 5461 and X 4047. This 

 is by no means always the case. Janicki * showed that the 

 principal line of X5461 has fine structure and can be re- 

 solved into five lines. "We recently found that the principal 

 lines of X4359 and X4047 are of similar nature, and cannot 

 be treated as a single broad line. This fact will doubtless 

 have an important bearing on the Zeeman effect. It will 

 therefore not be out of place to 'give a brief sketch of the 

 result, reserving the detailed description of the resolution 

 for another place. 



By combining the echelon grating with a Lummer-Grehrcke 

 plate, both of resolving power slightly exceeding 4 x 10 5 , 

 wej" have shown that there are two satellites +17 and —23 

 m.A.U. very near the principal line X4359. The interval 

 between —23 and the principal is, however, so vague, that 

 the presence of a cluster of lines seemed to us very probable. 

 By combining an echelon grating or a Lummer-Grehrcke 

 plate with a sliding Fabry-Perof interferometer, and making 

 the air-plate about 5 cm. thick, thereby utilizing a resolving 

 power of about 2 x 10 6 , we found that the assumed principal 

 line is composed of a group of three strong lines, the con- 

 secutive distance between them being about 6 m.A.U. The 

 middle line of the triplet may be considered as the principal. 

 "When the analysing power of the interferometer is not 

 sufficient to resolve the principal line, the mean point appears 

 displaced by about 3 m.A.U. towards the side of longer 

 wave-length, from that given in our investigation con- 

 cerning the constitution of mercury lines. The reason 

 why Janicki did not observe the fine structure of the violet 

 lines must be ascribed to the low resolving power of his 

 instruments. 



* Janicki, Ann. d. Phys. xxxix. p. 439 (1912). 



t Nagaoka and Takamiuo Proc. Phvs. Sue. xxv. p. 1 (1912). 



