SOLAR OBSERVATORY. 135 



precisely what he required. The principal advantage of his visit, however, 

 was to familiarize us with the phenomena of anomalous dispersion, and to 

 bring out, in a series of discussions, many suggestions required in the formu- 

 lation of a general attack on the question, which must be made here, in imme- 

 diate connection with the solar work. 



In preparing for this attack I have borne in mind the requirements of 

 several closely correlated lines of research. Our interpretation of sun-spot 

 spectra has advanced rapidly during the year. The hypothesis that the 

 strengthening and weakening of solar lines in spots is due to reduced tem- 

 perature has received support from several sources. Chief among these are 

 the results of laboratory studies of spectra corresponding to various temper- 

 atures, the discovery in spot spectra of bands due to compounds which may 

 not exist at the higher temperature of the photosphere, and the strengthening 

 of spot lines in the red stars. It is nevertheless true that certain phenomena 

 can not be accounted for by this hypothesis, and our minds must be open to 

 the possibility that anomalous dispersion plays some part in producing the 

 widened lines. In the hope that it would assist in the interpretation of spot 

 spectra, we have made a comparative photographic study of the spectra of 

 various parts of the sun's disk. We found, to our surprise, that near the 

 limb the spectrum exhibits very marked changes, not only in the relative 

 intensities of the lines, but also in their positions. While the results so far 

 obtained suggest that a difference in effective pressure may be the principal 

 cause of the displacements, it is by no means impossible that anomalous dis- 

 persion may play a principal or a minor part. When we also consider the 

 diverse and complex problems encountered in the interpretation of the spec- 

 troheliograph results, we perceive that nothing short of an extensive series 

 of laboratory investigations, involving the comparative study of the temper- 

 ature, pressure, and anomalous dispersion changes of a selected list of solar 

 lines, must be made in conjunction with a parallel study of the same lines in 

 the sun and stars. 



In such work we are concerned with purely physical questions, which 

 underlie the interpretation of solar and stellar phenomena. Let us now con- 

 sider a very different phase of the problem of stellar evolution. Thanks to 

 the labors of such men as Gill, Boss, Campbell, and Kapteyn, rapid progress 

 is being made in our knowledge of the structure of the sidereal universe.* 

 The closeness of the bond that unites the astronomy of position with astro- 

 physics is well illustrated here, for one of the greatest of present needs is 

 the determination of the velocities of motion in the line of sight of a very 

 large number of stars. These spectrographic results would settle the ques- 

 tion of the sun's motion in space, add greatly to our knowledge of the two 



* Ably summarized by Sir David Gill in his recent address as president of the British 

 Association. 



