»h v.] PLANETARY EVOLUTION. 387 



regarded as settled for ever, and as settled in favour of the 

 nebular hypothesis. Henceforth, to the evidence found in 

 the structure of our planetary system, there may be added 

 the weighty argument that masses of matter still exist in 

 space, in the very condition in which our system must have 

 originally existed. 



If the nebular hypothesis was ever to be subjected to a 

 hazardous trial, one would suppose that the discovery of 

 spectrum analysis must have furnished the occasion. Here 

 is a discovery which has suddenly enlarged our knowledge of 

 the stellar universe in a manner utterly beyond the power of 

 the greatest and subtlest mind to have predicted twenty 

 years ago, — a discovery which not only reveals to us the 

 actual motions of the stars, but even penetrates into their 

 molecular structure, and discloses the chemical elements of 

 which their surfaces are composed as well as the physical 

 state of aggregation of those surfaces. Now if ever, one 

 might think, is the time to find out whether our nebular 

 hypothesis, devised in an era of comparatively scanty astro- 

 nomical knowledge, is a sound hypothesis or not. If it 

 survives this immense, unprecedented extension of our know- 

 ledge, what more magnificent triumph could we wish for it ? 

 And here we see that the very first result of the application 

 of spectrum analysis to sidereal phenomena has been the 

 placing of the nebular hypothesis upon a firmer basis than 

 ever before, removing the only serious obstacle which had 

 hitherto deterred many cautious thinkers from committing 

 themselves to it. 



Spectroscopic researches but lately undertaken, and not yet 

 carried out to a decisive result, seem likely not only further 

 to strengthen the noble theory of Kant and Laplace, but to 

 give it a comprehensive significance of which those great 

 thinkers could never have dreamed. Along with further 

 eonfirmation of the process of mechanical and physical 

 evolution, as originally formulated in their hypothesis, evi- 



C c 2 



