66 The Great Solar Eclipse [Jan.^ 



ment, formed of three very dispersing prisms, was arranged of short 

 length, and to give much hght. 



The sht of the spectroscope having been arranged so as to cnt 

 at right angles the image of the narrow luminous arc which would 

 remain some seconds before total obscurity, the light from the 

 extremity of the horns was first examined. On the gi-ound of a 

 spectrum with very sharp black lines formed by the diffused atmo- 

 spherical light was seen a much more luminous band, which was the 

 spectrum of the light emitted by the extremity of the horn. What- 

 ever was the width of this part, nothing particular could be noticed 

 in it ; the rays had an appearance in respect to width and intensity 

 identical mth those of the ordinary solar spectrum. 



The observation of the horns was interrupted some seconds before 

 totality, in order to remove the diaphragms from the telescope, to 

 shghtly open the slit of the spectroscope, and thus be prepared to 

 examine the protuberances. 



At the instant of total obscurity, the slit of the spectroscope 

 having been brought on to the image of the long protuberance, 

 which became visible on the eastern edge of the sun, M. Eayet im- 

 mediately saw a series of nine brilliant lines, which, by their arrange- 

 ment in the field, their relative distance, their colour, and their 

 general efi'ect as a whole, appeared to be; related to the principal 

 lines of the solar spectrum — b, d, e, h, an unknown line, f, and two 

 lines of the group, g. These hues possessed great brilliancy, and 

 appeared strongly relieved from the ashy grey very pale ground. 



During these observations, the slit of the spectroscope was 

 parallel to the principal length of the protuberance. Thus the 

 luminous lines were seen in the apparatus of a length in proportion 

 to the height of the protuberance ; the slit having been turned 90^^ 

 round, the rays appeared reduced to the appearance of brilliant 

 points corresponding to the slight wdth of the luminous horn ; no 

 error of observation is therefore possible, and the brilliant lines 

 actually represent the spectrum of the Hght of the protuber- 

 ances. 



The spectroscope being in the first position (the slit parallel to 

 the length of the protuberance) the very brilliant lines correspond- 

 ing to D, E, and F, were seen to be prolonged beyond the mean 

 length, by a very feeble luminous tract, the spectrum presenting 

 the appearance given in the coloured diagram. A certain portion 

 of the incandescent gaseous light which forms the protuberances 

 therefore spreads into the solar atmosphere beyond the limits which 

 the eye in general assigns to these expansions. 



The examination of this first i^rotubcrance being finished, the 

 slit was du'ected to the large luminous region at the west of the 

 sun. This time, also, the spectrum was seen to consist of brilliant 

 lines arranged as in the first case, with the exception of there being 



