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



47 



Reynolds' observations, yet in themselves very valuable, do not in this case form 

 any definitive proof against the cluster-theory. 



The difference in colour, resp. spectral type, found by Seares and Lundmark 

 & Lindblad, between the spiral arms and the nucleus in some spiral nebulse, will 

 perhaps be regarded as speaking against Reynolds' opinion, as a reflection 

 on the particles of the nebulous atmosphere would result in making the atmosphere 

 redder than the nucleus, while the observations above-mentioned show, that it is 

 just the contrary that is the case. 



Are the spectral and photometrical researches concerning the spiral nebulse not 

 able conclusively to confirm the theory of their great distances, we must nevertheless 



o 

 260 



1 

 400 



800" 

 Fig. 7. 



r 



1600 



5- 6 



3200 Light-years 



Upper curve: The light curve of Andromeda nebula. 

 Lower curve: The star-density curve in our stellar system derived from Schwaezschild's treatise A. N. 190:361. 



say that, if Reynolds' not very plausible theory is correct, nothing else is to be 

 supposed than that the spirals must be at distances at least equal to those of the 

 globular dusters. If not, implausible suppositions will have to be made concerning 

 the physical state of the stars in the spiral nebulse. 



5. The Motions oj Spiral Nebulce. 



It is obviously by no means out of the question, that the great radial veloci- 

 ties found for the spirals are not real, but that the measured displacements of spectral 

 lines represent some other phenomenon than a DoppLER-effect. We want, however, to 

 call in mind that some stars have a radial velocity amounting to the same order of 



