148 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1924 



have finally witnessed the start of these detached masses on their 

 voyages into space, all precisely as pictured by Laplace. 



The one essential difference is that of size. The evolutionary 

 process we have been watching occurs on a scale such as Laplace 

 never dreamed of. His primeval nebula was supposed to be of 

 about the size of Neptune's orbit, a size represented on the scale I 

 used at the beginning of this lecture by a threepenny-bit. On this 

 same scale the nucleus alone of a good-sized spiral nebula, such as 

 those shown in Plate 3, Figure 2, and Plate 4, Figure 1, would be 

 about the size of the Albert Hall, while the arms would sprawl over 

 the whole of Hyde Park and Kensington. The pictures of these 

 nebulae tlmt you have before you would have to be enlarged to the 

 size of a whole country, or even possibly of a whole continent, before 

 a body the size of our earth became visible in them at all. 



Although the parent nebulae we have been considering are all in- 

 comparably greater than Laplace's imaginary nebula, yet each tiny 

 condensation, as it starts off into space, is a gaseous nebula the mass 

 of which is just about equal to that imagined by Laplace and the 

 size of which is not perhaps very greatly different. If, then, this 

 younger generation of nebulae meet with the same experiences in life 

 as their giant parents before them, we should not have to look far 

 for an explanation of the origin of the planets, and if the third gen- 

 eration again repeated the experience of their ancestors, the satellites 

 of the planets are also accounted for. But mathematical research 

 and observation agree in disposing of so simple an explanation of 

 the origin of the solar system. As we have seen, it is only because 

 the filaments in the spiral nebulae are of such huge size that gravita- 

 tion is able to cause condensation in opposition to the expansive 

 tendency of gas pressure. A nebula of mass comparable to our sun 

 might go through the same life history as the bigger nebula until 

 matter began to be thrown off from its equator, but after this the dif- 

 ference of scale would begin to tell, and the subsequent course of 

 events would be widely different. The ejected matter could not 

 condense into filaments, still less into detached globules; it would 

 merely constitute a diffuse atmosphere surrounding the parent 

 nebula. As such a system shrank by the emission of radiation, the 

 constancy of angular momentum would, at first, merely demand that 

 more and more gas should be transferred from the center to the 

 atmosphere. 



But mathematical investigation shows that in time, after the cen- 

 tral star had shrunk to a certain critical density, perhaps somewhere 

 about one-tenth of that of water, a cataclysmic period would ensue, 

 from which the mass would emerge as a binary star — two stars of 

 comparable masses revolving about one another nearly in contact 

 and in approximately circular orbits. This is a formation with 



