142 ANNUAL UP:I'0RT SMITHSONIAN INSTITUTION, 1937 



Another result of the examination of the stability of a star is 

 important. The rate of Uberation of subatomic energy must increase 

 with temperature but not too fast; if it increases more steeply than a 

 certain limit the star will be thrown into pulsation. Some stars do 

 pulsate, namely, the Cepheid variables, but the majority do not. 

 Perhaps we may infer that the actual law of increase is pretty near 

 the limit, so that the conditions of most of the stars are on the one 

 side and those of the Cepheids just beyond it. But there is a way 

 by which the star can escape this pulsatory instability. We have 

 been supposing that the response of the subatomic energy to the 

 stimulus of temperature is immediate; if there is a lag — if the rising 

 temperature stimulates the formation of active material which emits 

 the energy later on in its own good time, or if it starts a chain of 

 processes of which the actual energy hberation is the last, then there 

 w^ll be no pulsation. A lag of some days at least is required. Pro- 

 vided there is this lag, the stars wdll be stable, even though the energy 

 liberation increases very rapidly with the temi)orature — as our 

 observational results for the main series stars indicate and as is also 

 indicated by the recent laboratory experiments. 



This is the main information about subatomic energy that we have 

 learned from astronomy. I suppose that, taken altogether, it seems 

 a meager amount. But its importance is considerably enhanced, 

 when we recall that on almost every point it was completely at 

 variance with the views then held by physicists. The only form of 

 liberation of subatomic energy with which physicists were then 

 acquainted was radioactivity — a process independent of density and 

 unaffected by temperature unless the temperature were far higher 

 than 15 million degrees; and they were inclined to be intolerantly 

 disposed toward considering any other process, no matter how strong 

 the astronomical evidence might be. I cannot but think that this 

 is an instance of the harm done by the writers who give the impression 

 that stellar investigation is a field of loose speculation. Physics and 

 astrophysics are one subject, following the same rules of progress, 

 recognizing the same standards of rigorous deduction, and utilizing 

 the same corpus of accepted knowledge; and liable to the same 

 failures through our human Imiitations. 



Various attempts were made to find a loophole for admitting much 

 higher temperatures in the stai-s so as to satisfy the physicist's objec- 

 tion to admitting energy liberation controlled by low temperature; 

 for example, Jeans's theory of elements of very higli atomic weight, 

 and Milne's theory of the existence of a core of white dwarf density 

 in ordinary stars like the sun. We can scarcely say that such sug- 

 gestions are impossible without attributing to our existing Icnowledge 

 of the laws of physics greater completeness than we care to claim. 

 But I think it can be said firstly that these theories were found on 



