214 CARNEGIE INSTITUTION OF WASHINGTON. 



(2) The decline in radiation in passing from the photosphere to the center 

 of the spot is continuous over the penumbra, the rate of change increasing as 

 the umbra is approached. 



(3) The total radiation from the faculae is from 7 to 10 per cent greater than 

 that from the adjoining photosphere. This indicates that the faculse consid- 

 ered as black bodies have temperatures about 130° higher than the photo- 

 sphere. 



Energy curves have also been made of the spectra of sun-spots and the 

 neighboring photosphere. The ordinates of these curves are measured and 

 from a series of photographs the mean values are derived for the ratio between 

 the radiation from the spot and that from the photosphere for any given 

 wave-length. These ratios are, of course, independent of losses by transmis- 

 sion through the earth's atmosphere. A curve constructed from these values 

 shows the following characteristics: 



(1) The radiation in sun-spots decreases from 82 per cent at 1.7 /x to 30 

 per cent at 0.4 fjL. 



(2) Broad absorption bands are indicated in the sun-spot spectrum at 1.37 /x, 

 1.92 fx, and to a less extent at 0.94 ^u. 



From measurements of the slope of this curve in the region of maximum 

 radiation, on the assumption of Planck's law and a temperature of 6500° C. 

 for the photosphere, the radiation temperature of spots is found to be 5530° C. 

 The total radiation gives very nearly the same result. 



THE DISPLACEMENTS OF SOLAR LINES. 

 In the last annual report a brief discussion was given of the state of the 

 observational evidence regarding the existence of displacements between solar 

 and arc lines of the size required by the generalized theory of relativity. 

 According to this theory, the gravitational shift of the solar lines should be 

 -}-0.010 A at X5000. A complete study of the question, therefore, involves: 

 (1) an accurate determination of terrestrial wave-lengths; (2) an accurate 

 determination of solar wave-lengths; (3) an extensive study of the causes 

 giving rise to displacements of lines in the sun, such as general and local con- 

 vection, lateral drifts, pressure and possible effects from density distribution, 

 and irregular refraction and dispersion. Investigations of all three of these 

 problems are in progress by Mr. St. John and Mr. Babcock, and their results at 

 the present time, as well as their methods of studying the complicated phenom- 

 ena involved in solar displacements, may be summarized under corresponding 

 heads. 



Terrestrial Wave-Lengths. 



The report of the Commission on Wave-Lengths, adopted at the Rome 

 meeting of the International Astronomical Union in May 1922, shows the 

 important progress which has been made in the determination of the wave- 

 lengths of the lines in the spectrum of the iron arc. The 300 lines adopted 

 as tertiary standards after a careful investigation of the character of the source 

 and the elimination of those recognized as unstable, show such a degree of 

 accordance for the different observers as to indicate a probable error of less 

 than 0.001 a. On the average, each wave-length has been measured by 

 five or six observers, and so complete and accurate are the results that the 

 spectrum of this important element naturally will serve as the basis for future 

 determinations of wave-length. 



