200 CARNEGIE INSTITUTION OF WASHINGTON. 



stant mean atomic mass which changes but little in giants of increasing tem- 

 perature, even with marked changes of density (p. 235). 



Through a combination of the values of the masses of the dwarf stars with 

 the space-velocities obtained at Mount Wilson, Seares has found that the mean 

 energy (mean mass X mean square space-velocity) for all types of dwarfs from 

 A0 to Ma is practically constant, in spite of a variation in mass of 10 to 1. 

 Assuming that equipartition of energy holds for giants such as the Cepheid 

 variables, the probable dispersion in mass among 28 such stars for a given 

 magnitude and type is only about 20 per cent. 



The spectroscopic method of determining absolute stellar magnitudes has 

 received further important development in the hands of Mr. Adams and Mr. 

 Joy. Through the rigorous classification of stellar spectra and the separate 

 grouping of stars giving diffuse lines and stars giving sharp lines, the absolute 

 magnitudes of the white (A) stars, not measurable by the earlier method, are 

 derived. This permits the spectra of 544 stars, previously photographed with 

 the 60-inch and the 100-inch telescopes, to be utilized for parallax work, thus 

 yielding a large and welcome addition to available data (p. 231). 



The continuation of measurements of radial velocity has led to the accumu- 

 lation of many results which have already been used by Stromberg 

 and others in discussions of stellar motions. A list of over 1,000 stars with 

 radial velocities obtained at Mount Wilson has been prepared for publication 

 by Adams and Joy and should prove of especial value to astronomers because 

 of the large proportion of stars with known distances. In addition to these 

 results, the completion of the determination of the velocities of stars of the 

 Md type of spectrum by Merrill provides the material for an investigation 

 of the motions of this important class of variables (p. 229). 



The stellar spectroscopic work of the year has included a great variety of 

 investigations in addition to those already mentioned. Among these, 

 reference may be made to the studies of the spectra of variable stars by Adams, 

 Joy, and Merrill; Sanford's computations of the orbits of spectroscopic bina- 

 ries; spectral classification of faint stars in the Selected Areas by Humason; 

 and measurements by Sanford tending to confirm the relative displacements of 

 enhanced and arc lines in the spectra of bright stars. The report of the 

 Commission on the Spectral Classification of Stars, prepared for the Inter- 

 national Astronomical Union by Adams and Russell, represents an attempt 

 to add to the existing system of classification symbols which shall describe 

 in a more adequate way numerous important characteristics of stellar spectra 

 (p. 233). 



Of especial significance in its bearing on our knowledge of the nature of 

 the two star-streams discovered by Kapteyn and the character of the motions 

 of the stars within our stellar system is an investigation by Mr. Stromberg of 

 the space-motions of stars of spectral types A to M. With the aid of the 

 parallaxes obtained for the A-type stars by the recently discovered spectro- 

 scopic method of Adams and Joy, and the radial velocities observed at Mount 

 Wilson and other observatories, Stromberg has computed the individual 

 space-velocities for 332 stars. An analysis of these motions at once leads 

 to the important conclusion that Kapteyn's first stream may be identified 

 with the Taurus group and his second stream with the Ursa Major group. 

 The remaining stars belong to a central group of small systematic motion , 



