PHYSICS: A, S. KING 461 
tematically in error; the violet and red components being assigned 
values, respectively too small and too large. 
4. The systematic deviations for lines with violet and red companions 
found by comparing the Rowland and International wave-lengths go 
pari passu with and are referable to the erros in Rowland wave-lengths. 
The coefficient of correlation is -1-0.55=1=0.05. 
5. These systematic deviations, therefore, do not furnish evidence 
that the relative positions of the Fraunhofer lines are systematically 
displaced by mutual influence. On the other hand, the sun-arc dis- 
placem_ents and the relative separation of the components of close pairs 
in solar and arc spectra indicate that, within the limits of error, evidence 
of mutual influence is absent from the solar spectrum, and, in so far 
as mutual influence is a necessary corollary of anomalous dispersion 
in the sun, evidence for it also is absent. 
1 A. Einstein, Leipzig, Ann. Phys, 35, 898 (1911). 
2W. H. Julius, Astrophys. 40, 1 (1914). 
' Trans. Internal. Union Co-op. Solar Res., 4, 74. Charles E. St. John and Harold D. 
Babcock, 'A Study of the Pole Effect in the Iron Arc,' Mt. Wilson Contr., 106; Astrophys. 
J.,46, (1915). 
* Sir Joseph Larmor, Observatory, 497, 103 (1916). 
5 Sebastian Albrecht, Astrophys. J., 41, 333 (1915). 
6 W. H. Julius, Astrophys. J., 43, 53 (1916). 
7 Charles E. St. John and L. W. Ware, Mt. Wilson Contr., — , Astrophys. J., — (1916). 
AN ATTEMPT TO DETECT THE MUTUAL INFLUENCE OF 
NEIGHBORING LINES IN ELECTRIC FURNACE SPECTRA 
SHOWING ANOMALOUS DISPERSION 
By Arthur S. King 
MOUNT WILSON SOLAR OBSERVATORY. CARNEGIE INSTITUTION OF WASHINGTON 
Received by the Academy, July 6, 1916 
When light traverses a transparent medium, the variation of the index 
of refraction with the wave-length of the incident light may undergo 
decided changes if the medium shows selective absorption. If a beam 
of white light is passed through a mass of vapor which is itself emitting 
a spectrum, the sudden change of refractive index close to the absorp- 
tion lines can be shown by suitable optical arrangements. One method 
of producing this effect is based on the principle of crossed prisms used 
by Kundt, and consists in making the absorbing medium form a prism 
whose refracting edge is at right angles to the sKt of the analysing spec- 
troscope. An absorption line showing anomalous dispersion will then 
cause the continuous spectrum to bend sharply in opposite directions 
