356 THE ILLUMINATION OP TFIE ECLIPSKD MOON. 



stances cannot bo without inflncnco, and sliould bo borno 

 in mind when comparisons are made ; but no one who 

 watched, for example, the two last total oclipses visible 

 in this country, could doubt for a moment that thoro are 

 very great differences in the light of the eclipsed moon, 

 apart from difforonces in the stato of the atmosy)here, or 

 in the moon's position in the sky. 



In a note on the eclipse of October, 1884, which is 

 printed in our Proceedings, I stated what seemed to me 

 very strong objections to the commonly received theory 

 regarding the variation of the light in different eclipses, 

 and I ventured on a suggestion that the solar corona might 

 have something to do with it. TIio eclipse of last January 

 led mo to restate my views, and a correspondence ensued, 

 which stimulated mo to a more careful examination of the 

 question than I had previously made. In a matter which 

 like this has engaged the attention of eminent astronomers, 

 there seems small chance of discovering anything that is 

 both new and true ; and yet I hope to bo able to show that 

 some important features of the question have been hitherto 

 overlooked. 



At first sight thoro seems no reason why the moon, when 

 immersed in the shadow of the earth, should receive or 

 exhibit any light at all, and hence an attempt to account 

 for the variation in the amount of light (which is tho 

 special object of this paper) involves of necessity a.n in- 

 quiry into the possible sources whence tho light may be 

 derived. 



Kepler, some throe centuries ago, was the first to suggest 

 an explanation of the phenomenon, and it is substantially 

 his theory which is accepted by astronomers to-day. It 

 is believed, first, that tho illumination of tho eclipsed moon 

 is brought about by tho refraction of tho solar rays through 



