196 CARNEGIE INSTITUTION OF WASHINGTON. 



2. The proper-motions of the Class N stars are systematically small; their 

 radial motions so far as known are small; they must as a class, therefore, be 

 numbered among the more slowly moving members of the stellar system. 



3. The direction of the solar motion derived from the red stars agrees well 

 with that derived from the stars of other spectral classes. 



4. The red stars follow the general tendency of the other stars in prefer- 

 ential motions towards Kapteyn's vertex. 



Since the publication of this investigation a great deal more material has 

 become available, especially in Gyllenberg's observations of 391 of these 

 stars (Meddelanden fran Lunds Astronomiska Observatorium, Series ii, vol. 

 32, No. 11), and a complete reinvestigation of the material has been carried 

 through. From this study the proper-motions of 312 red stars have been 

 derived. An analysis of these motions in now under way. 



STELLAR WAVE-LENGTHS. 



The study of wave-lengths in stellar spectra was continued by Dr. Albrecht. 

 Wave-lengths were determined for the Class B8 star c Carinae. The publica- 

 tion of these wave-lengths is being delayed temporarily in the expectation 

 that it may shortly be possible to more nearly complete their identification 

 with laboratory results. 



Detailed study of Class B stars gives promise of contributing very important 

 results toward the problem of the constitution of matter. It is well known 

 that the lines present or measurable in these types change very markedly with 

 minor changes in type. Thus, the published measures of the Yerkes and 

 Lick Observatories indicate that Class B8 is differentiated from Class B5 by 

 the rather abrupt appearance of a considerable number of metallic lines and 

 the practical disappearance of the oxygen and nitrogen lines, and from Class A 

 by a pronounced falling off in the number of metallic lines. The silicon lines 

 increase from BO to Bl or B2 and then gradually fade out toward B8, while 

 the magnesium line 4481.4 is very faint in the early B-types and becomes 

 very prominent in B8. The metallic lines involved are the so-called "en- 

 hanced lines." The importance of Class B8, which is, as it were, a transition 

 stage for the enhanced lines, has been materially increased by the recent 

 advances in the theory of ionization. 



The K-term, which has been attributed to various causes, including the 

 Einstein effect, has been shown to be due, at least in part, to the use of in- 

 accurate normal wave-lengths in the radial velocity determinations. In the 

 June number of the Astrophysical Journal it was shown that recent laboratory 

 wave-lengths for twenty oxygen and nitrogen lines occurring in B-type stars 

 are systematically 0.063 a longer than the adopted normals. This corre- 

 sponds to a change of —4.2 km. in the radial velocities derived from them. 

 A discussion of the available portions of the data on which the K-term is 

 based indicates that the new wave-lengths for this one group of lines alone 

 would reduce the K-term, which is about 4 km., by about 0.3 km. for the 

 entire B class and by about 0.8 km. for classes BO to B2. In a second article, 

 which is practically ready for publication, it will be shown that new labora- 

 tory wave-lengths by Crookes for silicon lines and a new wave-length for the 

 helium line 4713 by Merrill still further reduce the K-term by about 1 km. 

 for the entire B class and by about 2 km. for classes BO to B2. Moreover, it 

 is pointed out that the Unes upon which necessarily most emphasis has been 



