PHYSICS: A. S. KING 
463 
The density of the absorbing vapor is necessarily a function of the 
temperature when plenty of the material is present. At a given temper- 
ature, the amount of anomalous dispersion is proportional to the vapor 
density, which alters the width of the absorption line, so that for special 
purposes the effect can be regulated by changing the amount of vapor. 
The reproduction shows the anomalous dispersion for the chromium 
lines X X 4254, 4275, 4290, the temperature being approximately 2500°C. 
in each case, with varying vapor densities. 
ANOMALOUS DISPERSION OF CHROMIUM LINES WITH VARYING VAPOR DENSITY. 
The tests on the mutual influence of spectrum lines were carried out 
by mixing titanium and calcium vapors with that of chromium. By 
this means, a titanium line was obtained separated by 0.21 A from 
X4275 of chromium, while a calcium and a titanium line occur 0.36 A 
and 0.65 A respectively from X 4290 of chromium. The titanium and 
calcium lines were thus well within the curved spectrum given by the 
strong anomalous dispersion of the chromium lines. The wave-lengths 
of the former, given as absorption lines under these conditions, were 
compared with their values when produced by the furnace as emission 
lines. 
Twenty-seven spectograms showing varying degrees of anomalous 
dispersion were measured. The titanium line separated by only 0.21 A 
from the chromium line X 4275 gave consistent measurements indicat- 
ing an approach to the chromium, line in the anomalous dispersion 
plates, the mean difference between the two sets of measures being 0.006 
A. It is questionable, however, whether this displacement is genuine, as 
the spreading of the strong chromium line weakened the continuous 
spectrum on one side of the titanium absorption line, which would prob- 
ably affect the micrometer settings. The calcium Hne separated by 
0.36 A from X 4290 of chromium showed an apparent approach amount- 
ing to 0.002 A, to which the same source of error may apply. The line 
