ASTRONOMY: C E. ST. JOHN 
21 
5. Such a change of radial velocity is a necessary consequence of a 
velocity distribution (for the peculiar motions) different from that given 
by Maxwell's law. 
6. A first approximation to the velocity distribution has been derived 
for the K stars. It explains the change of velocity with proper motion 
in a satisfactory manner. 
7. Some positive indications have been found of a change of radial 
velocity with absolute magnitude, the brighter stars moving more slowly 
than the fainter stars. 
^Lick Obs. Bun.,7, 113 (1913). 
2 See Annals Harvard Coll. Obs., vol. 50. 
3 Adams and Kohlschiitter, ''Some special criteria for the determination of absolute stellar 
magnitudes," Contrib. Mt. Wilson Solar Obs., No. 89; Astrophys. J. 40 (1914). 
* For example, Asirophysical Journal, 31, 266 (1910), 
^Kapteyn, Monthly Notices, 72, 743 (1912). 
^Eddington, Observatory, 34, 355 (1911). 
^ Proc. Amsterdam Acad. Sciences, 14, 524 (1911). 
8 Schwarzschild, Astron. Nachr., No. 4557 (1912). 
^ See Publ. Astron. Lab. Groningen, No. 11, p. 12. 
CRITIQUE OF THE HYPOTHESIS OF ANOMALOUS DISPERSION 
IN CERTAIN SOLAR PHENOMENA 
By Charles E. St. John 
MOUNT WILSON SOLAR OBSERVATORY. CARNEGIE INSTITUTION 
OF WASHINGTON 
Presented to the Academy, November 18. 1914 
According to the theory of anomalous dispersion, the interpretations 
of solar observations in terms of absorption, motion, pressure, and level 
are generally misleading since the observed phenomena, such as prom- 
inences, flash spectra, flocculi, and displacements of the Fraunhofer 
lines are from this point of view mainly the effects of anomalous refraction 
in the solar atmosphere, so that in their study we are facing optical 
illusions. The establishment of such a point of view would revolutionize 
or render futile many of the present Hues of solar and stellar observation, 
and would make practically impossible the solution of many problems 
which confront the investigator. It is therefore of the highest impor- 
tance to determine the form and extent of the influence of anomalous 
refraction in the solar atmosphere, if it obtains to a detectible amount. 
There is a degree of vagueness in the deductions from the theory, due 
to its extreme flexibility, that makes a quantitative examination of its 
claims difficult. During the present year, however, W. H. JuKus, its 
