228 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 2 8 



conditions all continue unchanged in direction. So that from the 

 foregoing we may not only conclude quite definitely that the stars 

 are not the sources of the observed cosmic rays, but also that the 

 main atom-huilding processes prohdbly do not take flace inside of 

 stars at all. 



Second. The foregoing conclusions may also be arrived at from an 

 entirely different mode of approach, namely, from our measure- 

 ments upon the absorption coefficients and the total energy content 

 of the cosmic rays. 



The hardest rays which we have observed are completely absorbed 

 (reduced to say 2 per cent of their initial intensity) in going through 

 70 meters of water. This means that even if the atom-building pro- 

 cesses went on inside a star, the resulting cosmic radiations could not 

 possibly get out, but would all be frittered away in heat ^^ before 

 emergence, save in the case of those rays that originated in the star's 

 very outermost skin — a skin equivalent in absorbing power to a 

 hundred or so meters of water.-" 



But we have also found that the total energy coining into the 

 earth's atmosphere in the form of cosmic rays is about one-tenth 

 the total heat and light energy coming to the earth from the stars 

 exclusive of the sun. This last fact means that if the cosmic rays 

 have their origins within the stars they can not even at the points 

 of their oHgin have an intensity more than 10 tiines tJmt which 

 they have when they reach the eartKs atmosphere.^ for if they liad 

 then the cosmic-ray energy transformed into heat hy absorption on 

 the way out woidd yield a total heat outfoio from the stars larger 

 than the observed 10 to 1 ratio. In other words, if the stars are 

 the sources of the observed cosmic rays, it follows from our measure- 

 ments on absorption coefficients and on total energy content that the 

 total heat output of the stars must be furnished by the atom-building 

 processes going on in their merest outer skins of a thickness equiva- 

 lent in absorbing power to about a hundred meters of water, and 

 that therefore no atom-building processes, nor any other activities 

 capable of furnishing heat, can then be going on in their interiors. 



It is, however, so altogether absurd to suppose that atom-building 

 processes are going on actively at the surface of a star, down to a 

 depth of a hundred meters, and then suddenly stop there that we 

 are forced back by this present mode of approach to the same con- 

 clusion arrived at from the direct determination of the lack of 



"^ It is important to remember that, as v.'e have already shown, rays of this liind become 

 frittered away into heat in tiiis passage tlirough matter without any change in the 

 ijualitj/ (i. e., frequency or absorption coefHcient) of the residual beam. This can only 

 nioan that the conditions eHsting in and about the sun, and presumably also in and about 

 tther stars a^ well, are unfavorable to the atom-buildino processes ichich (jive rise to these 

 fays. 



-' Millikan, R. A., and Cameron, G. H., Phys. Rev. 28, 8GG, 1920. 



