STELLAR KVOLUTTON. 151 



obsemation. In seekino" foi' oviciciu-c of stellar evolution, on a plan 

 conipreheu.sive enoiiyb to'include a place for every star in the heavens, 

 we may begin with visual and' |)hotographii' observations with the tele- 

 scope. Such reniarkal)le photoo-raphs as that of the Andromeda nebula 

 seem to bring us into the very presence of a greater system, perhaps 

 more nearly compai'able in size with the Milky Wa}' than with the solar 

 system, in the actual process of formation. But on ticcount of the long 

 periods of time which must elapsi^ before changes in this distant mass 

 may become sufficiently great to be appreciable, and for many other 

 reasons, we could not hope to l)ase a complete scheme of stellar evolu- 

 tion on such photographs alone. Our observational methods must also 

 include the means of solving physical, chemical, and gravitational 

 problems as they present themselves, not close at hand in the labora- 

 tory, ])ut in inconceivably distant regions of space. For this reason 

 it would have been impossible prior to the invention of the spectro- 

 scope to arrange the stars according' to any clearly defined system of 

 development. The principal advances which have been made in the 

 study of stellar evolution are therefore confined to the period which 

 has elapsed since the middle of the nineteenth century. 



Thus the investigation of stellar evolution has been contemporaneous 

 with the investig'ation of organic evolution. Indeed, the epoch-making- 

 discovery of the chemical comp()sitit)n of the sun 1)v Kirchhofl and 

 Bunsen was made in the year of the pul>lication of the Origin of Spe- 

 cies. Before this discovi^w the meaning of spectral lines had been as 

 obscure as the meaning of Egyptian hieroglyphs ])rior to the discovery 

 of tlu> Kosetta stone. After it the chemical analysis of a star l)ecame 

 hardly less difficult than the analysis of an uidvnown su))stance in the 

 la1)oratorv. Furthermore, it soon ])ecame apparent that the light of a 

 star, as decomposed by a prism, was competent to define the star's posi- 

 tion in a general scheme of development, in which exery ad\ance from 

 the unformed neladous (doud on through the highest degree of stellar 

 brilliancy to such a final stage as is typified by the moon can be ilefined 

 with but littl(> danger of (M-ror. Before we proceed to considin* some 

 of the evidences of stellar evolution, let us examine some of the instru- 

 ments and methods without which the discoveries to ])o subse((uently 

 described would have been impossible. 



1 shall confine my remarks on modiM'n asti-ophysical instrinnents to 

 those at present (Muployc^l at the Yerkes ()t)ser\atory, j)artly becaus(> 

 nearly all the celestial ph()tograi)hs reproduced in the figures were 

 taken with these instruments and })artly because of the cotivenience of 

 illustrating tliem. Hut befoi'e describing the great t(descop(\ which 

 forms tlie principal a])paratus of the obser\atory, I wish to [)oint out 

 that many of the most important results of astronomy, results which 

 could not be obtained wi(h a i)owei'f"ul telesc()])e foi- the \-erv rt^ason of 

 its great power, have Ikhmi dei'i\'ed from the use of an ordinary c:un(Ma 



