( 302 ) 



which is nearly identical with the value above. Here, however, the 

 brightness is diminished to 0.257 of the original, hence by 1.48 

 magnitude. 



Therefore it appears here that these two different causes produ-ce 

 similar colours, but that they correspond to an entirely different decrease 

 of brightness. When comparing the two we may say that atmospheric 

 absorption is more apt to redden, a decrease of temperature more 

 apt to fade the light. Therefore it - is impossible to derive the 

 radiating power fi'om the colour only, as we do not know to what 

 degree each of the two influences, temperature and absorption, is at 

 work in the different spectral classes. Perhaps that one day accurate 

 spectral-photometric measurements will enable us to separate the 

 two influences, for they give a different distribution of intensity over 

 the spectrum. For the log. of the brightness of different A with regard 

 to ;i500 we find 



A = 650 600 550 500 450 400 



with abs. 1.05 atm. +0.114 +0.083 + 0.051 0.000—0 084—0.231 

 with glowing— Vj + 0.154 +0.111 + 0.061 0.000 -0.074 —.0166 



For the latter the decrease in intensity from the red to the violet 

 is more regular, for the former the decrease is slower for the greater 

 and more rapid for the smaller wave-lengths. 



These calculations show that it is not strictly true that, as has 

 been said in the preceding paper, a redder colour must necessarily 

 involve a smaller radiating power. Where we have two influences 

 which in different ways bear on the colour and the brightness, the 

 possibility exists that a redder colour may be accompanied by a 

 greater radiating power, namely when one source of light has a 

 much higher temperatui'e and at the same time a greater atmospheric 

 absorption than the other. An increase of the degree of glowing of 

 + \/j combined with an absorption of 2 atmospheres gives such a 

 case according to the figures given above. 



Herein we have therefore a new possibility to account for the 

 peculiarities found in the K stars, namely by assuming that, as 

 compared with the G stars, they have a much higher temperature, 

 which causes a stronger radiation, and which by very strong atmos- 

 pheric absorption, is only little faded but greatly reddened. We 

 must add, however, that this explanation seems little probable to us 

 as the band-absorption, which begins at the /iT-stars and which is 

 characteristic for the M stars (the 3'<^ fype) indicates a lower tem- 

 perature. 



