Vol.. 6, 1920 PHYSICS: A. A. MICHELSON 473 
THE VERTICAL INTERFEROMETER 
By a. a. MichejIvSOn 
Ryerson PhysicaIv Laboratory, University of Chicago 
Read before the Academy, April 26, 1920 
The investigation of the distribution of intensity in spectral Hnes by 
interference methods while somewhat laborious as compared with the direct 
methods, visual or photographic, has the important advantage of freedom 
from instrumental uncertainties such as arise from optical imperfections. 
1.2 4 8 
FIG. 1. 
finite slit width and diffraction, which are especially objectionable in the 
case of lines or groups so narrow as to be barely within the resolving power 
of the spectroscopic appliances employed. 
Besides the theoretical value of such investigations as affecting the 
interpretation of such distribution as based on various assumptions of 
kinetic theory, effect of temperature, pressure, electric and magnetic 
fields, a practical application may be mentioned, namely the accurate 
description of the appearance of spectral lines, which at present are char- 
acterized as broad, narrow, diffuse, etc. Thus, for symmetrical lines 
this distribution may be expressed hy y = 2 -^^/^)", in which A is the half 
width of the line and n determines the "square shouldered" character, 
as illustrated in figure 1. 
One of the difficulties in the experimental determination of the visibility 
of the interference fringes as a function of the difference in path, from which 
the distribution of intensity may be inferred, is the maintenance of parallel- 
ism of the moving mirror, and the vertical interferometer is designed to 
obviate this objection. 
For this the moving mirror is the level surface of mercury in a shallow 
dish (at least 7 cm. in diameter to avoid curvature due to capillarity). 
