RUMFOEi) SPECTROHELIOGRAPH. 133 



lioiico its width must ho such as exactly to inchicle this line and to 

 exclude all li^ht from other parts of the spectrum. It is evident, 

 tiiat the spectroheliogTaph may he considered simply as a form of 

 monochromatic exposing; shutter, ditfering from the ordinary focal 

 plane shutter only through the use of a nai'rower exposing slit and 

 the inclusion of an optical train which limits the light to a single 

 line in the sj)ectrnm. 



Although this idea suggested itself to me quite independently in 

 1S8D, I subsequently learned that the principle was by no means new. 

 Indeed. Janssen had suggested it as early as 1869, while Braun. of 

 Kalocsa, and Lohse, of Potsdam, had designed instruments involv- 

 ing the same principle in 1872 and 1880, respectively. Indeed, 

 Lohse had constructed and experimented Avith the instrument he de- 

 signed, but his work was not successful. This may have been due 

 in part to the fact that the hydrogen line which he employed is not 

 nearly so well adapted for i:)rominence photography as are the H 

 and K lines of calcium. This was one of the difficulties experienced 

 in my first (unsuccessful) experiments, Avhich, through the kindness 

 of Professor Pickering, were made at the Harvard College Observa- 

 tory in the winter of 1889-1890. 



In April, 1891, after the Kenwood Observatory httd been equipped 

 with a 1'2-inch equatorial i-efractor and a powerful solar spectroscojje, 

 a photographic study of the ultra-violet spectrum of the chromosphere 

 and prominences was undertaken in the hope of finding lines better 

 adapted than those of hydrogen for the photography of the promi- 

 nences. The brilliant H and K lines of calcium, previously observed 

 \'isually in full sunlight by Professor Young and photographically at 

 total eclipses, were found in all cases to be the most consi)icuous lines 

 in the spectrum of the chromosphere and prominences. The remark- 

 able l)rightness of these lines, and more particularly their position at 

 the center of the dark, broad shades, due to the denser calcium vai)or 

 in the lower portion of the solar atmosphere, render them i)eculiarly 

 well adapted for the purposes of prominence photography. Indeed, 

 it was possible with their aid to obtain good photographs of single 

 [)rominences merely by opening the slit of the spectroscope to such an 

 extent as to include a considerable part of the prominence and giving 

 a very short exposure to the inuige formed directly upon a photo- 

 graphic plate. But this method was too limited to be of general 

 application. In order to record photographically the entire surface 

 of the sun, with the chromos])here and prominences, it was necessary 

 to employ the principle of the sjiectnjheliograph, involving the use 

 of narrow slits, moved with reference to the solar image and jihoto- 

 graphic plate. The first successful spectroheliograph was brought 

 •iiUo use at the Kenwood Observatory in January. 1892. After this 

 time it was employed regularly on every clear day until May, 1895, 



