MOUNT WILSON OBSERVATORY. 217 



period-luminositj^ law for Cepheids, and concludes that the average 

 heat content per unit mass is nearly identical for all of these stars hav- 

 ing periods longer than three days. 



The continuation of the systematic investigations of stellar spectra 

 by Messrs. Adams, Joy, Merrill, Sanford, and Stromberg, assisted dur- 

 ing the past year by Messrs. Duncan and Hoge, has yielded 135 new 

 determinations of (constant) radial velocity and the orbits of five new 

 spectroscopic binaries. Both the 60-inch and 100-inch telescopes have 

 been employed, the former chiefly for the absolute magnitudes of stars 

 brighter than the eighth magnitude, the latter for the study of fainter 

 stars, including those with very large proper motions, Md stars, faint 

 Cepheids, Algol variables, Novse, and miscellaneous objects (p. 245). 



The reductions of Mr. Adams's determinations of stellar luminosity 

 and parallax have been finished during the year, and the complete 

 tables will be published within a few weeks. The spectroscopic method 

 of measuring parallaxes has proved increasingly satisfactory, and the 

 detailed comparison of the results with the best trigonometric paral- 

 laxes indicates a very high order of precision. The tables provide data 

 for many interesting discussions, some of which are also in preparation 

 for publication. 



Much attention has been devoted by Mr. Adams and his associates to 

 the spectra of temporary stars, and many new and interesting phenom- 

 ena have been observed. In Nova Lyrae, for example, the displace- 

 ments of the two components of the hydrogen lines on February 5 were 

 in the ratio of 2 to 1, while the oxygen and nitrogen lines were shifted 

 by the larger of these values. Displacement ratios of 3 to 2 and 2 to 1 

 were noted in the last annual report for the components of the hydro- 

 gen Unes in other Novae, and the discovery of a similar phenomenon in 

 the spectrum of the variable T Pyxidis may prove to be of significance 

 in the development of the theory of temporary stars. As the spectrum 

 of T Pyxidis in April also showed other close resemblances to that of a 

 Nova at an intermediate stage, the question of possible relationship will 

 be closely pursued (p. 247). 



The interesting phenomena observed in the spectrum of o Ceti (p. 

 249), and the discovery by Mr. Merrill of the characteristic nebular 

 lines in the spectrum of R Aquarii, emphasize the necessity of direct- 

 ing renewed attention to the spectra of variable stars in all stages of 

 brightness. Nebulium has not been founa previously in any stellar 

 spectra except those of Novse, and it is very important that many other 

 Md stars be observed near minimum, when the nebular lines, which in 

 R Aquarii do not appear to change in brightness, should be most easily 

 detected. The Hooker telescope affords the necessary light-gathering 

 power for such observations, which will be included in Mr. Merrill's 

 general investigation of the Md stars. The radial velocities of 46 of 

 these stars have now been determined, making a total of 90 thus far 



