520 WRIGHT— RECENT SPECTROSCOPIC OBSERVATIONS. 



lege Observatory. The nature of the arrangement in that system is 

 indicated by the typical spectra of the illustration (Fig. i). The 

 top spectrum is that of a Welf-Rayet star, or Class O star in the 

 Draper system; following it is one of Class B, in which the dark 

 lines of helium are strong. In the next the rhythmic hydrogen series 

 predominates, while, as we go down, that in turn fades, and metallic 

 lines, notably the strong calcium doublet, become pronounced. It 

 will now be observed that while the sequence on the screen is based 

 entirely on the occurrence of lines in the spectrum, another charac- 

 teristic is shown as we pass from the top to the bottom of the pic- 

 ture, that is, the cutting ofif of the spectrum from the left side. This 

 means the impoverishment of the violet end of the spectrum, as com- 

 pared with the red. In other words we pass from stars that are 

 bluish white, through those that are yellow to those that are red. It 

 is one of the facts of our ordinary experience, fortified by elaborate 

 theory and experiment, that this succession of color phenomena 

 marks the cooling of an incandescent body ; and the view has been 

 generally adopted that exhibits such as that on the screen represent 

 the spectra of stars in an order of continually decreasing tempera- 

 ture. Evidence confirmatory of this opinion is afiforded by the fact 

 that the spectral lines in the upper spectra are found in the labora- 

 tory to be characteristic of high temperature, while some of those in 

 the lower are due to chemical compounds which can exist only in a 

 comparatively cool environment. 



Astronomers have, for a long time, thought that the spectral 

 sequence here indicated oflfers the basis for inference with respect 

 to stellar evolution, though there is not unanimity of opinion as to 

 just how the evidence should be interpreted. It is manifestly im- 

 possible in the available time to indicate the many points of view 

 from which the evidence has been considered. Probably a majority 

 of those who are interested in such matters are inclined to interpret 

 directly the sequence in the illustration, that is, to assume that the 

 upper spectrum is one of the newly formed star, and that the 

 following spectra are of stars in successive stages of development. 

 This is equivalent to the hypothesis that a star originates at a 

 high temperature and cools continuously throughout its period of 

 visibility, fading out as a dim red object. This view is opposed 



