MOUNT WILSON OBSERVATORY. 211 



2. Measurements of the wave-lengths of these lines show that they 

 are displaced toward the violet end of the spectrum by amounts vary- 

 ing from 20 A at X 4000 to 24 A at H^, the displacement being pro- 

 portional to wave-length. The shift of the less refrangible component 

 of the hydrogen and helium lines (when both components are present, 

 agrees with these values. Interpreted in terms of radial velocity, 

 this displacement would correspond to a motion of approach of 1,500 

 km. in the gas producing these lines. Although velocities of this order 

 should be accepted with reserve, it is most difficult to conceive of any 

 physical cause other than motion of the source which could produce 

 displacements of this magnitude proportional to wave-length and at 

 the same time leave the lines comparatively narrow and well defined. 

 It now seems probable, from an examination of the photographs of 

 the spectrum of Nova Geminorum No. 2 of 1912, that a similar dis- 

 placement was present, but of about half the amount. 



3. The radial velocity of Nova Aquilse No. 3, as determined from 

 the narrow absorption fines K, H, Di, and D2, is — 15 km. 



4. The first evidence of the presence of the chief nebular fine at 

 X 5007 was found on the photograph of June 20, only 12 days after 

 the discovery of the star. 



Some Spectral Characteristics of Cepheid Variables. 



A study of the spectra of a number of the more prominent Cepheid 

 variables shows certain well-defined pecufiarities which appear to be 

 characteristic of these stars as a class. Among them we may refer to 

 the following: 



1. The enhanced lines are exceptionally strong, being more promi- 

 nent at maximum than at nfinimum light. 



2. The hydrogen lines also are exceptionally intense. This pecu- 

 liarity is so marked that it results in an estimate of spectral type, 

 when the hydrogen lines are used, very different from that obtained 

 from the more general characteristics of the spectrum, such as the 

 intensities of the metallic arc lines. 



3. The variation in spectral type between maximum and minimum 

 of light for the Cepheid variables is mainly a variation in the intensity 

 of the hydrogen and the enhanced lines, and not in the spectral lines 

 as a whole. 



4. At maximum of light, the maximum of intensity of the lines in 

 the hydrogen series shifts toward the violet in the same way as that 

 of the continuous spectrum. 



Minor Investigations. 



Among the results of other investigations we may refer briefly to 

 three: 



1. The spectrum of Ceti has been observed throughout a consid- 

 erable portion of its period. Recently we have been able to identify 



