PAPERS PRESENTED AT GENERAL SESSIONS 55 



STELLAE EVOLUTION 



Professor W. D. MacMilla^:. University of Chicago 



It is becoming more generally and more clearly recog- 

 nized that the source of the sun's heat does not lie in the 

 gravitational contraction of the sun. It is tnie that if 

 the sun had contracted from infinity to its present size 

 the heat thus generated would be sufficient to supply the 

 sun's present rate of radiation for about twenty millions 

 of years. Xot only is there no evidence that the sun was 

 ever larger than it is at present, but even though we 

 grant that it was larger, twenty millions of years is alto- 

 gether inadequate to satisfy the geological requirements 

 as to the age of the earth. The methods of determining 

 the age of the earth from the thickness of the sedimen- 

 tary rocks and from the amount of salt in the oceans 

 agree in giving a minimum age of at least three hundred 

 millions of years, while the radioactive method deter- 

 mines the age of certain of the Archean rocks to be as 

 high as sixteen hundred millions of years. It seems 

 unlikely that even this figTire represents the full age of 

 the earth. Estimates recently made by Chamberlin and 

 by Lord Eayleigh range from three to ten billions of 

 years. Evidently the contraction hyiDothesis furnishes 

 less than one per cent of the amount of heat that is neces- 

 sary for the life history of the earth. 



Turning from a study of the earth to a consideration 

 of the d^^lamics of stellar systems it is seen that the 

 difficulty is multiplied many fold. There are over eighty 

 groups of stars which are known as globular star clusters. 

 They contain many thousands of stars in what seems like 

 a very minute region of the sky. Actually this size is very 

 great, for the stars of which they are composed are 

 doubtless, on the whole, comparable to our sun. Each 

 star moves to and fro in the cluster under the gravita- 

 tional attraction of all of the other stars of the cluster. 

 This process of oscillating back and forth has gone on 

 for so long a time that the clusters have arrived at a 

 steady state of motion in which the motions have become 

 so uniformly distributed that the cluster has a spherical 



