THE ILLUMTNATrON OF THE ECLIPSED MOON. 



359 



when the sun is seen like a red ball tlirougli a fog. In 

 certain states of the atmosphere the sun, even at setting, 

 shows no redness, and then it is too bright to look at, even 

 when the greater part of its orb has dipped beneath the 

 horizon. Indeed I believe the law to be universal, that 

 whatever state of the atmosphere reddens the sun also 

 reduces its brightness. 



Sir John Horschol's idea that the unusual obscuration 

 of the moon in certain eclipses was due to an unusual 

 prevalence of cloud in the ring of atmosphere through which 

 the sun's i-ays would have to pass, is not O])on to the 

 objection that it is at variance with observed facts, for we 

 know that clouds will intercept the sun's rays. But it is 

 open to an objection that applies to any theory which seeks 

 to explain the variation in the moon's visibility by varia- 

 tions in the conditions of the earth's atmosphere. Con- 

 sider the circumstances of the case. The atmospheric ring 

 through which the sun's rays must pass is a ring 25,000 

 miles in circuit ; and it is something more than a ring, for 

 owing to the earth's rotation during the progress of the 

 eclipse, a perpetual succession of rings will come into 

 position. An eclipse of the moon may be total for an hour 

 and a half or more, and in that interval a breadth of 

 atmosphere will have boon brought successively into the 

 position of the ring, extending in the equatorial regions to 

 about 1,500 miles. To me it seems in the highest degree 

 improbable that a tract of atmosphere 25,000 miles in 

 length, and many hundreds of miles in average breadth, 

 embracing, moreover, every description of terrestrial climate, 

 .should vary materially in its aggregate or average con- 

 dition at any two epochs. 



The same line of argument may, I think, bo carried a 

 step further. It is important to note that, as far as our 



