

KUNGL. SV. VET. AKADEMIENS HANDLINGAR. BAND 60. N:0 8. 45 



The measurements shown in Table X give an interesting contribution to the 

 question of the selective absorption in space. The spectral types having been de- 

 termined from the appearance of the lines, and, on the contrary, X e ff. from the centre 

 of gravity of the continuous spectrum, we may expect that, in case there is a selec- 

 tive absorption prevailing in space, the X e f f :s ought to be displaced towards the red 

 end of the spectrum, and, consequently, the inferred spectral-types redder than the 

 observed ones. As the inferred spectra cannot very well be displaced relatively to 

 the observed ones with more than V2 spectral interval (or 0'".2 in colour-index) at 

 the utmost, and as the parallax of the tabulated spiral nebulse must be considered 

 <0".ooooo5, we conclude that the change in colour index per parsec, d, is 



< 0"',000(K)01 . 



Shapley, in a somewhat other way, finds d <0" ! .ooooo2. 



Thus it is evident that, if there exists a selective absorption in space, it is so 

 small that it is not necessary to consider it, as long as greater distances are not 

 concerned than those here derived for globular dusters and spiral nebulse. 



Fath's investigation 65 concerning the total spectrum of the Milky Way has 

 given the result that our stellar system in its entirety seems to belong to the 

 spectral type G. This circumstance both confirms the above discussion about space 

 absorption and also lends some probability to the supposition that spiral nebulse are 

 remote galaxies. 



The discovery lately made by Evershed 6 - of the occurrence of stationary calcium 

 clouds in the Milky Way will perhaps throw new light npon the appearance of the 

 H and K lines in the spectra of spiral nebulse. If it should prove to be the case 

 that the occurrence of such clouds is something characteristic of the Milky Way, it 

 is near at hand to see in the appearance of the said lines in the spectra of spirals 

 a support for the opinion that we have here to do with galactic systems or large 

 stellar systems. On the other hand, the fact that in certain spiral nebulse there 

 appear bright lines characteristic of the radiation of the Wolf-Rayet stars and the 

 gaseous nebulse, seems to be an argument against this opinion, as it does not seem 

 probable that the objects, which in our stellar systems give emission-lines in their 

 spectra would have such a total magnitude, that they would come in sight in a total 

 spectrum of the Milky Way examined from a point far outside of the same. 



To this, however, the objection might be made, that the emission-lines have 

 not been observed in all the spirals examined, and there is nothing preventing that 

 some of them contain considerably more nebulous matter, planetaries and Wolf- 

 Rayet stars than our Galax v. 



If we do not adopt the opinion that spiral nebulse are stellar sj^stems, there 

 only remains the possibility of a glowing (står-) nucleus, globular in form, and sur- 

 rounded by a gaseous atmosphere that reflects the light from the nucleus. 

 Reynolds 162-163 , who maintains this opinion, has photographically derived the light- 

 curve for the Andromeda nebula. From his intensities, derived by taking the mean 



