582 Meeting of the British Association. [Oct., 



detailed accounts that, except in the immediate neighbourhood of 

 the sun, the hght of the prominences consisted mainly of three 

 bright lines, of which two coincided with C and F, and the inter- 

 mediate one nearly, but, as subsequent researches showed, not 

 exactly, with D. The bright lines coinciding with C and F indicate 

 the presence of glowing hydrogen. Several of the other lines were 

 identified with those which would be produced by the incandescent 

 vapour of certain other elements. 



Valuable as these observations were, it is obvious that we should 

 have had long to wait before we could have become acquainted with 

 the usual behaviour of these objects, and their possible relation to 

 changes which may be going on at the surface of the sun, if we had 

 been dependent on the rare and brief phenomenon of a total solar 

 eclipse for gathering information respecting them. But how, the 

 question might be asked, shall we ever be able so to subdue the over- 

 powermg glare of our great luminary, and the dazzling illumination 

 which it produces in our atmosphere when w^e look nearly in its 

 direction, as to perceive objects which are comparatively so faint ? 

 Here again the science of optics comes in aid of astronomy. 



When a line of Hght, such as a narrow sht held in front of a 

 luminous object, is viewed through a prism, the light is ordinarily 

 spread out into a colom-ed band, the length of which may be 

 Increased at pleasure by substituting two or more j)risms for the 

 single prism. As the total quantity of light is not thereby increased, 

 it is obvious that the intensity of the light of the coloured band 

 will go on decreasing as the length increases. Such is the case 

 with ordinary sources of Hght, like the flame of a candle or the 

 sky, which give a continuous spectrum, or one generaUy continuous, 

 though interrupted by dark bands ; but if the light from the source 

 be homogeneous, consisting, that is, of light of one degree of refran- 

 gibihty only, the image of the slit will be merely deviated by the 

 prisms, not widened out into a band, and not consequently reduced 

 in intensity by the dispersion. And if the source of light emit 

 light of both kinds, it will be easily understood that the images 

 of the slit corresj)onding to light of any definite refrangibilities 

 which the mixture may contain will stand out, by their superior 

 mtensity, on the weaker ground of the continuous spectrum. 



Preparations for observations of the kind had long been in 

 progress in the hands of Mr. Lockycr ; and on the 20 th of October 

 last year, in examuiing the space immediately suri'ounding the edge 

 of the solar disk, he obtained evidence, by the occurrence of a bright 

 line in the spectrum, that his slit was on the image of one of those 

 prominences, the nature of which had so long been an enigma. 

 Notices of this discovery were received from the author by the 

 Koyal Society on October 21st and November 8rd. These were 

 Bhortly afterwards foUowed by a fuller paper on the same subject. 



