THE FUNCTION OP LARGE TELESCOPES. 135 



appears, until finally stars of a deep red color are found, wliicli seem to 

 mark the last stage of development before complete extinction of light. 

 Through a part of this line of evolution it is easy to trace the changes 

 in stellar spectra, the solar lines still continuing to be present, and 

 superposed upon them a remarkable series of flutiugs which are charac- 

 teristic of these reddish stars of the third class. But between such 

 stars and those of the class which Vogel has designated as Mb there 

 seems to be a break in the evolutionary chain. 



Stars of Class Mb are of an orange or red color, and with the tele- 

 scope alone some of them can not be distinguished in appearance from 

 the more fully developed stars of Class Ma. But in the spectroscope 

 they are entirely different. All of these objects are extremely faint, 

 the two brightest of them being hardly visible to the naked eye. For 

 this reason but little has been learned of their spectra, although the 

 spectra of stars like Vega and Arcturus, which are some scores of times 

 more brilliant, have been carefully investigated by both visual and 

 photographic means. According to Duncr and others, the spectrum of 

 the star known as 152 Schjellerup consists of certain heavy, dark bands, 

 which coincide closely in position with bands given by compounds of 

 carbon, and, in addition to these, a luminous zone in the orange portion 

 of the spectrum. Three or four of the most intense solar lines have 

 also been detected in these objects. But beyond this it is impossible to 

 go with the appliances used in the earlier investigations, although it 

 may well be that photographic methods would have greatly changed 

 the character of the results obtained. 



During the past winter a photographic study of the red stars has 

 been rendered possible by the 40-inch Yerkes telescope. Photographs 

 of the spectra of many objects of this class have now been obtained, 

 and many lines which were not previously recognized on account of the 

 faintness of the spectrum in small telescopes have been recorded. In 

 the case of two stars of Class Mb, 132 and 152 Schjellerup, the spectra 

 have been photographed with a powerful spectrograph containing three 

 prisms, giving high dispersion and considerable x>recision to the meas- 

 ures. It has been found that among the most characteristic features of 

 these spectra are numerous bright lines, some of which seem to have 

 been glimpsed by Secchi in his pioneer work at the Collegio Bomano, 

 though his drawings do not correctly represent their appearance or posi- 

 tion. In fact, he recorded bright lines where none exist, and failed to 

 record others, among which are the brightest in the spectra. Both 

 Duner and Vogel, who are certainly to be regarded as the best authori- 

 ties on the subject, altogether deny the presence of bright lines. And 

 had my own observations been confined to an examination of the spectra 

 with the instruments used by these observers I would unhesitatingly 

 subscribe to their opinion. But the great light-collecting power of the 

 40-inch telescope renders the detection of the bright lines a compara- 

 tively easy matter. Even with this instrument, visual observations with 



