222 CARNEGIE INSTITUTION OF WASHINGTON. 



as variables, exhibit the chemical characteristics of reduced tempera- 

 tures, made familiar by our investigations of the vapors in sun-spots. 

 While stars of this class must be studied with moderate dispersion 

 because of their faintness, it is fortunate that we have a few examples 

 of red stars bright enough, with the 100-inch, to permit their spectra 

 to be critically investigated with high dispersion. For this purpose 

 the gain of a magnitude, especially if realized at the coude focus 

 (equivalent focal length 250 feet), is absolutely essential. Thus, one 

 obvious application of the Hooker telescope will be in an investiga- 

 tion of the red stars of various classes, both with reference to their 

 place in the galactic system and their physical development, inter- 

 preted in the light of a comparative study of sun-spot and laboratory 

 spectra. The proposed high dispersion work on the brightest red 

 stars, and on other stars of the first and second magnitudes, should aid 

 in this interpretation, and may possibly reveal minute line displace- 

 ments indicating progressive changes in pressure or other physical 

 condition characteristic of the transition from class to class. 



We thus begin with a piece of research demanding the full capacity 

 of the Hooker telescope, and intimately related to other phases of our 

 stellar, solar, and laboratory work. Let us now consider another 

 investigation, which can also be attacked to peculiar advantage with 

 the Hooker telescope, especially in the light of recent important theo- 

 retical advances. 



In his fundamental researches on the dynamics of rotating masses, 

 Darwin dealt with incompressible matter, which assumes the well- 

 known pear-shaped figure, and may ultimately separate into two 

 bodies. Roche, on the other hand, discussed the evolution of a highly 

 compressible mass, which finally acquires a lens-shaped form and ejects 

 matter at its periphery. Both of these are extreme cases, hardly 

 realized in practice. Jeans has recently dealt with intermediate cases, 

 such as are actually encountered in stars and nebulae. He finds that 

 when the density is less than about one-quarter that of water a lens- 

 shaped figure will be produced with sharp edges, as depicted by Roche. 

 Matter thrown off at opposite points on the periphery, under the 

 influence of small tidal forces from neighboring masses, may take the 

 form of two symmetrical filaments, though it is not yet entirely clear 

 how these may attain the characteristic configuration of spiral nebulae. 

 Jeans goes on to discuss the evolution of the arms, which will break up 

 into nuclei (of the order of mass of the sun) if they are sufficiently 

 massive, but will diffuse away if their gravitational attraction is small. 

 The mass of our solar system is apparently not great enough, according 

 to Jeans, to account for its formation in this way. 



These investigations, which lead to conclusions very different from 

 those derived by Chamberlin and Moulton from the planetesimal 

 hypothesis, afford with the latter hypothesis valuable suggestions for a 



