42 
Mr Parvis, The influence of a strong magnetic field 
Mg. 5173 
Cd. 4800 
sextuplets 
Mg. 5168 
Cd 4678 
triplets 
Zn. 4722 
Cd. 4800 
sextuplets 
Zn. 4680 
Cd. 4678 
triplets 
Mg. 
Cd. 
Zn. 
, -2-73 
- 2-79 
- 2-45 
- 246 
-0-71 
1 +0-68 
+ 2-07 
-244 
-0-70 
+ 0-70 
+ 245 
Mg. 5173 
Zn. 4722 - 
sextuplets 
- 1-90 
— 0'66 
+ 0-60 
+ 1-90 
l + 2-86 
+ 2-79 
+ 2-51 
/ - 2-82 
- 2-82 
Mg. 5168 
t - 2-51 
\ - 0-02 
-o-oo 
Zn. 4680 
| - o-oi 
( + 2-84 
+ 2-82 
triplets 
' + 2-52 
, — 2'65 
-2-68 
-2-06 
-204 
) -0-67 
-0'67 
' +0-67 
+ 069 
+ 2-04 
+ 2-04 
4- 2-66 
+ 2-66 
f - 2-72 
- 2-71 
\ o-oo 
000 
[ +2-72 
+ -2-71 
In the author’s experiments with these three metals, the spec- 
trum of each metal was produced quite separately from the others 
by having both electrodes of the same metal, and the magnetic 
field was always of constant strength. The outside constituents 
of the sextuplets were not very clear of each other, and only by 
introducing a self-induction coil and using a very narrow slit was 
it possible to sharpen the lines so as to measure them with any 
degree of accuracy. They were more diffuse and weaker than the 
two middle constituents. In the sextuplets, the distances were 
measured between corresponding constituents ; and in the triplets, 
from the centre of the middle one. The following table gives the 
values of dX/X” calculated from these measurements and in the 
same units as used by Runge and Paschen. 
Zn. 4722 
Mg. 5173 
Cd. 4800 
Mean 
+ 2-77 
+ 2-78 
+ 2-88 
+ 2-81 s 
+ 2-02 
+ 2-06 
+ 2-03 
+ 2-04 s 
+ 0-67 
+ 0-68 
+ 0-70 
+ 0‘68 p 
0 
0 
0 
0 
-0-67 
-0-68 
-070 
- 0-68 p 
- 2'02 
- 2-06 
- 2-03 
- 2 04 s 
-2-77 
-2-78 
-2-88 
- 2-81 s 
Zn. 4680-5 
Mg. 5167-6 
Cd. 4678-4 
Mean 
+ 2-70 
+ 2-75 
+ 2-82 
+ 2'75 S 
0 
0 
0 
0 p 
-2-72 
- 2-74 
- 2-82 
- 2-76 5 
