480 
PROF. J. W. NICHOLSON AND DR. T. R. MERTON ON THE 
several components, and this will not be attempted in the present paper in any 
individual case ; but much information can be derived by taking definite cases of the 
Stark effect and plotting the results which the present wedge should give. This is 
a more rapid method than direct calculations for determining the essential features 
which the curve should present under various circumstances. 
In his most recent paper, Stark has given the details of the various components 
of H a as follows, although, in view of the difference in the conditions, we must not 
expect them to appear without serious modification in the present experiments. 
Under a field of 104,000 volts/cm. Stark finds for H a : 
p components 
• -I 
s components 
No. of component. 
Separation 
(A.U.). 
Intensity. 
+ 3 
+ 11-'5 
1-2 
+ 2 
+ 8-8 
1-1 
+ 1 
+ 6-2 
1-0 
-1 
- 6-2 
1-0 
-2 
- 8-8 
1-1 
-3 
-11-5 
1-2 : 
+ 1 
- 2-6 
1 
0 
0 
2-6 
- 1 
- 2-6 
1 
Now if a broadened line has an intensity I at its centre, and follows the simple 
exponential law, the quantity of energy in it is proportional to 
I e qx dx = I/q. 
Jo 
Stark’s lines are not broad, and intensity in such a case is rather a measure of 
contained energy than of central brightness, whatever the mode of measurement. 
The last column of Stark’s table, therefore, may be taken as a measure of I /q where 
I is the central brightness of a line. The distinction is immaterial if q is the same 
for each component. 
Now for the wedge adopted, it was known that if an intensity I x passing through 
emerged as I 2 , and if 
iog M (iA) = x, 
then X = 0'2 + Q‘4 y, where y was the distance from the thin, end, which was not 
indefinitely thin. But 
J _ J^g-pi/tana __ J Q-py tan a. logic e 
and therefore 
p tan a = 0‘4/log ia e — Q'922, 
or practically unity. 
