230 ANNUAL EEPOET SMITHSONIAN INSTITUTION, 192 8 



ing- radiation is liidden away behind an impenetrable screen of mat- 

 ter — a screen that transforms all its energy into heat before the ray 

 can get out. If the cosmic rays originated within the stars they 

 would of course be similarly screened. 



On the otlier hand, that the atom-building processes responsible 

 for the cosmic rays, as distinct from the atom-destroying process just 

 considered, actually occur, as our experiments definitely show, outside 

 the stars, or at least where the rays produced by them can get to us, 

 and in an energy that is of the same order of magnitude as that of 

 the heat poured out by the star, is an extraordinarily illuminating 

 fact. For it suggests at once, when combined with the MacMillan- 

 Eddington argument, the following incomplete cycle, each element 

 in which now has the experime7ital credentials indicated in the 

 parentheses ; 



(1) Positive and negative electrons exist in great abundance in 

 interstellar space (see the evidence of the spectroscope) ; 



(2) These electrons condense into atoms under the influence of the 

 conditions existing in outer space, viz, absence of temperature and 

 high dispersion (see the evidence of the cosmic rays) ; 



(3) These atoms then aggregate under their gravitational forces 

 into stars (see the evidence of the telescope) ; 



(4) In the interior of stars — under the influence of the enormous 

 pressures, densities, and temperatures existing there — an occasional 

 positive electron, presumably in the nucleus of a heavy atom, trans- 

 forms its entire mas,s into an ether 'pulse the energy of which when 

 frittered away in heat maintains the temperature of the star and 

 furnishes most of the supply of light and heat which it pours out 

 (see the evidence of the lifetimes of the stars, MacMillan-Eddington- 

 Jeans). 



The foregoing is as far as the experimental evidence enables us to 

 go; but the recent discovery of the second element of the above 

 unfinished cycle, namel}^, that the supply of positive and negative 

 electrons is being used up continually in the creation of atoms the 

 signals of whose birth constitutes the cosmic rays, at once raises 

 imperiously the question as to why the process is still going on at all 

 after the eons during which it has apparently been in process — or 

 better why the huilding sto7ies of the atom^ have not all heen used up 

 long ago. And the only possible an^swer seems to be to complete the 

 cycle, and to assume that these building stones are continually being 

 replenished throughout the heavens by the condensation, with the aid 

 of some as yet wholly unknown mechanism, of radiant heat into 

 positive and negative electrons. This has been urged for years by 

 MacMillan. Indeed it is implicit in Einstein's 1905 equation unless 



