1880.] 



on the Photographic Spectra of the Stars. 



287 



Tbo apparatus is represented in this diagram (Fig. 1). It is 

 mounted on a base plate a with bevelled edges, whicli enables it to bo 

 accurately adjusted at the end of the telescope. The prism is at c. 

 The image of the star is brought upon the slit b. The light is rendered 

 parallel by lens d ; it passes through the prism, and is then, by a 

 second lens of quartz, made to converge and form an image on the 

 photographic plate /, which is inclined so as to bring a considerable 

 part of the spectrum to focus upon the plate. 



Fig. 1. 



This apparatus was found to meet very satisfactorily the one 

 primary condition of diminishing the star's light to the least possible 

 extent compatible with obtaining a spectrum full of fine details and 

 well defined. The photographs taken with this instrument measure 

 not more than half an inch from G to 0, and yet under suitable 

 magnifying power seven lines can be counted between H and K. 



The second important difficulty was to find a ready means of 

 bringing the luminous point, into which the star's light is gathered up 

 by the mirror, accurately upon any part of the very narrow chink, the 

 ■^Ijj part of an inch, through which the light has to enter the spectrum 

 apparatus, and further to maintain the star's image precisely within 

 the same part of this chink during the whole time of exposure of the 

 photographic plate, which might be as long as one hour or even 

 more. 



The telescope was, of course, mounted upon an equatorial stand — 

 that is, one in which the axis of motion is placed parallel to the earth's 

 axis of rotation, — so that the telescope when kept in suitable motion 



