326 CAPTAIN L. DARWIN, DR. A. SCHUSTER, AND MR. E. W. MAUNDER 
The part ijnmediately above the prominence towards the north, which must be close 
to the body of the Sun, is comparatively light; this is partly due to a speck of dust 
on the slit, but the continuous spectrum of the corona here must have been weaker. 
Generally speaking, the spectrum all over the northern side is fainter than that 
towards the south. It will be noticed that the bands are nearer together in the 
violet than in the blue. This is due to the fact that the back of the camera was 
tilted in order to have the whole range of the spectrum as nearly in focus as possible. 
The magnifying power was, therefore, smaller in the violet than in the blue, and 
hence the tapering of the spectrum towairds the more refrangible side. 
Eig. .5. 
The spectrum can be traced from a wave-length 4950 in the bluish-green to about 
3700 in the ultra-violet. The maximum of actinic intensity of the coronal light was 
decidedly more towards the red end of the spectrum than that of Sun light. This 
shows that the continuous spectrum is principally due to incandescent matter of a 
lower temjierature than that of the Sun. Although the polarisation of the corona 
and the appearance of Fraunhofer lines show that part of the light is due to 
