222 Mr Purvis, The influence of a strong magnetic field on the 
X 
Inten- 
sity 
dX/X 2 
2668-90 
3 n. 
+ 2-26 s 
+ 1-00 p 
0 
- 1-00 p 
-2-26 s 
The four components were very close to each other ; 
1 but they appeared to be of equal strength. Their 
[ weakness and diffuseness made the measurements 
rather doubtful. 
The following lines became triplets ; and the middle con- 
stituent of each was stronger than the two outside ones. 
3267-60 
1 
+ 1-17 s 
0 p 
-1-19 s 
1 The unaffected line was a little stronger than X 2770-08 ; 
r otherwise they were very similar in appearance and 
) values of the divided constituents. 
3231-60 
1 
+ 1-76 « 
0 p 
-1-75 s 
| Compare Bi 2938-41 and Bi 2898-08. 
3029-90 
1 
+ 1-95 s 
0 p 
-2-00 s 
| Compare with Sn 3009. 
2770-08 
2 
+ 1-20 s 
0 p 
- 1-20 s 
J Compare X 3267-60, Bi 3397 and Sn 3801. 
2598-15 
2528-62 
1 r. 
1 r. 
+ 1-60 s 
0 p 
- 1-60 s 
+ 1-59 s 
0 p 
-1-59 s 
' 
The constituents were weak and diffuse and very close 
„ to each other. They were only seen well separated 
i on analysis by the calcite. They appear to be fairly 
j comparable. 
Bismuth. 
The line 4722‘80 
were only seen well 
measurements were 
was divided into six constituents ; but they 
separated on analysis by the calcite. The 
between corresponding constituents. 
4722-80 
1 
+ 2-18 s 
+ 0-92 s 
+ 0-65p 
0 
- 0-65 p 
- 0-92 s 
-2-18 s 
J The two constituents vibrating parallel were sharper 
l and stronger than the other four; but none of the 
six were very sharp, and they were all very diffuse. 
The following lines were divided into four. The constituents 
were only seen well separated on analysis ; and the measurements 
were between the corresponding ones. 
