132 CARNEGIE INSTITUTION 



5. The development of new instruments and methods should be 

 encouraged. The more general provision of instrument shops in 

 connection with observatories is to be recommended for this purpose. 



6. The construction of large reflecting telescopes offers the greatest 

 possibility of improving the instrumental means of astrophysical 

 research. For all classes of stellar spectrOvScopic work, the meas- 

 urement of the heat radiated by the stars, the photography of neb- 

 ulae, and many similar investigations, the reflector has very great 

 advantages over the refractor. It is safe to predict that many future 

 advances will be made with its aid. (See page 141.) 



In closing these introductory remarks, I wish to add a word on 

 the question of cooperation. In seeking to further the interests of 

 astronomical and astrophysical research through cooperative effort, 

 it seems to me that great care must be taken not to defeat what is 

 perhaps the highest aim of the Carnegie Institution — the encourage- 

 ment and development of individual genius. Cases will doubtless 

 arise in which guidance of man}' workers by some well-known au- 

 thority may be desirable and even necessary to the accomplishment 

 of far-reaching results ; but for the most part such cases will involve 

 only the application of routine methods, sufficient for the collection 

 of a mass of data needed in certain investigations. My argument 

 is directed mainly against the centralization of authority and its con- 

 centration in a few individuals, however able they may be. It seems 

 probable that the greatest progress will come rather through the 

 advent of new minds, each competent to select its own point of view 

 and to plan researches unhampered by the regulations of established 

 systems. For this reason I believe that the best returns will be 

 realized from assistance given to individual workers, cooperating, it 

 ma}' well be, with others, but constantly encouraged to advance 

 science through the development of ideas and methods of their own. 



The following sections contain statements of the progress and 

 present state of certain departments of astrophysical research, for 

 reference in connection with the report of the Advisory Committee. 



SteIvLar Spectroscopy. 



Although Fraunhofer discovered in 1823 that the spectra of stars 

 differ among themselves and from the spectrum of the Sun, it was 

 not until 1859 that spectroscopy was established upon a firm basis 

 by Kirchhoff and Bunsen. Their analysis of sunlight by an instru- 

 ment which permits the chemical and physical phenomena of distant 



