210 Dr Schuster. Some results of the [Nov. 18, 



American observers successfully studied the spectrum of the corona. 

 Further facts were established in 1871 ; we now know that the 

 spectrum of the corona consists : 



1. Of a continuous spectrum in which the dark Fraunhofer 

 lines have been faintly seen. 



2. Of the spectrum of Hydrogen gas. 



S. Of the spectrum of an unknown substance giving a bright 

 green line (A, = 5316). 



Other lines have sometimes been suspected to exist, but 

 nothing definite is known about them. 



It will be the object of future eclipse observations to settle the 

 relative intensity of these various spectra as well as the distance 

 to which they reach away from the sun. That the relative bright- 

 ness of the continuous spectrum to the bright line spectrum varies 

 has been proved during the late eclipse, and the importance of 

 obtaining some numerical data will appear when we come to speak 

 of the polariscopic observations. 



The presence of a continuous spectrum indicates the presence 

 of liquid or solid particles ; for although we know of many gases 

 and vapours which give continuous spectra at comparatively low 

 temperatures, the presence of polarised light indicates the presence 

 of solid or liquid particles in a finely divided state. It is, as I 

 shall shew, most likely due to matter falling into the sun and being 

 gradually broken up by the heat of the sun. 



The existence of the hydrogen lines is not astonishing. The 

 presence of the green line indicates the existence of an unknown 

 gas most likely lighter than hydrogen. 



During the last eclipse the first attempt was made by Prof. 

 Eastman, assisted by Mr Pritchett, to gain precise ideas as to height 

 to which the various spectra reach all round the sun. Four direc- 

 tions were taken and the distance in fractions of a solar diameter 

 were estimated at which the spectrum disappeared. The result 

 was rather remarkable, for although the corona was not equal in 

 intensity in the four directions, the spectrum disappeared nearly 

 at the same distance all round the sun. The value of these 

 measurements will appear when compared to similar measurements 

 which no doubt will be made during future eclipses. 



It is clear that the spectroscopic observations of the corona 

 would be greatly facilitated if we could succeed in photographing 

 the spectrum of the corona. Attempts in this direction, which to 

 a great extent were successful, have been made during the two last 

 eclipses. The ordinary way of condensing the image of the corona 

 on the slit of a spectroscope, the telescope of which has been 

 replaced by a camera, has hitherto failed. The impossibility of 

 ever succeeding has been asserted ; the possibility will be shewn 

 as soon as the experiment has had a fair trial with instruments 



