INTERSTELLAR SPACE ABBOT 213 



to show spectral lines), he deduced that "the sharp calcium lines 

 must have their ori<j;in in an absorbing medium outside the com- 

 ponents of the binary systems." This conclusion was, of course, 

 facilitated by the doubling of all the spectral lines belonging to these 

 stars because of their mutual revolution. But his analysis did not 

 end there, for by combining his results wdth the then recent studies 

 of Campbell on the rapid motion of our sun with reference to 280 

 of the brighter stars Slipher was able to prove " that the origin of 

 the calcium absorption lines remains fixed with reference to the 

 frame fixed by Campbell's 280 stars, or, at most, has a very low 

 velocity, and therefore that it is outside the solar system, apparently 

 in stellar space."' He advances as an hypothesis, " It might then, 

 for the present, be assumed that the calcium absorption has its origin 

 in an interposing cloud covering at least certain extensive regions of 

 the sky." When, in addition, he suggested that the sodium D lines 

 be investigated for their stationary character, it will be evident from 

 what foUoAvs how far he w^as in advance of tlie time, and how un- 

 fortunate for the progress of this subject that this contribution 

 remained comparatively unnoticed. 



In 1920, in an investigation by Plaskett,* on the massive eclipsing 

 system Y Cygni, it was found that the relative shifting of the cal- 

 cium spectral lines indicated a difference of 40 kilometers per second 

 in velocity for calcium away from the center of mass of the two stars, 

 as measured by the lines of other chemical elements. This large 

 motion is quite beyond the possibility of explanation as errors of 

 measurement or wave length, and definitely rules out the hypothesis 

 of any limited calcium cloud surrounding the observed stars. For 

 out of such a cloud with such rapid relative motion the eclipsing 

 system would soon escape. Similar velocity differences were soon 

 found, and a new phase of the problem developed in a further re- 

 search on the O-type stars.^ It was conclusively shown that these 

 stationary calcium lines were present, not only in spectroscopic bina- 

 ries but in all stars of the class ; and that the velocity given by these 

 lines generally differed, frequently to a marked degree, and up to 

 50 and 60 kilometers per second, from the velocity given by the lines 

 of hydrogen, helium, etc., arising in the stellar atmosphere itself. 

 This seemed to leave no doubt of the presence in space of diffuse gas- 

 eous matter containing ionized calcium, the source of the so-called 

 " H " and " K " lines, tlirough whicli these high temperature O-type 

 stars were rushing rapidly aljout in all directions. 



Further, when it was shown quite independently of Slipher's ear- 

 lier work, which was then unknown, that the velocities given by these 



* I'ubl. Dominion Astrophys. Obs., vol. 1, p. 213, 1920. 



" Publ. Dominion Astropliys. Obs., vol. 2, p. 335, 1924 ; Monthly Not., vol. 84, p. 80, 

 1923. 



