( 1266 ) 
lines and the chromospheric lines are asymmetrical with regard 
to the exact positions of the emission lines, so as to have their 
“centres of gravity’ somewhat displaced toward the red. One can 
easily show that this effect must increase from the centre toward 
the limb of the solar disk. Its character corresponds exactly to that 
of the systematic line-displacements, described in recent years by 
Hate and Apbams, Fasry and Buisson, and others, and considered 
by those investigators as consequences of pressure in the reversing 
layer. Objections to their interpretation, and arguments in favour of 
our explauation which is based on the shape of the dispersion curve, 
are to be found in former communications’). The part of the 
dispersion curve, lying between the minimum and the maximum, 
had not yet been taken inte consideration then, that region being too 
narrow, with most Fraunhofer lines, to be observable by means of 
the spectral apparatus at present in use. But we are now extending 
our discussion to that middle part of the curve, which may perhaps 
reveal itself within a few very wide lines. 
The lower part of fig. 1 (derived from the dispersion theory) shows, 
for the environment of a single absorption line, the intensity R, of 
the light transmitted by the solar atmosphere, if, for all waves con- 
sidered, the intensity of the incident light is supposed to be S. The 
influence of scattering appears from the course of the (partly dotted) 
curve d,d,d,d,d,; that of absorption from the shape of the additional 
part lying between d, and d,, and showing a steep drop at O’. 
Only a few gases, strongly represented in the solar atmosphere, seem 
to exhibit so strong an absorbing power, that the minimum and the 
maximum of the dispersion curve are sufficiently distant from each 
ether to make the phenomena, characteristic of the interval, perceptible. 
Where this happens to be the case, we may expect to find, 
according to the dispersion theory *), that the “centre of gravity” 
of the central dark line is displaced toward the red, and that the 
apparent emission line, which is due to the scattering effect passing 
through a minimum at d,, shows a displacement toward the violet. 
This simple deduction is in perfect agreement with the general 
phenomenon, observed by DrsLANDRES, JEWELL, and Sr. JoHn with 
the lines H, and A,, H, and A, in the spectrum of the central 
parts of the solar disk. 
§ 4. The influence of anomalous refraction on the distribution of light. 
In order to find out how the effect will change as we approach 
1) Proc. Roy. Acad. Amst. XIII, p. 2, (1910); “Le Radium’, VII, Oet. i910. 
2) Proc. Roy. Acad. Amst. XIII, p. 895, (1911). 
ee i 
