THE ILLUMINAXrON OV THE ECLIPSED MOON. 36!) 



01- iriiigijiliocl by the roducod illumination, jnst as tho jm-o- 

 gross of tlio ponumbra in a lunar oclipso may bo rendered 

 strikingly moro obvious by the interposition of a thin cloud 

 — a oircunistanco which I have actually observed. 



But apart from inequalities of light which are referable 

 to tho moon itself, there are other inequalities which are 

 unquestionably proper to the earth's shadow ; or, as I have 

 expressed it, there are variations in the amount of light in 

 different parts of a transverse section of tlie shadow. I 

 believe that most of these latter variations may be com- 

 prehended in the single statement that the amount of liwht 

 increases from tho centre of tho shadow outwards to the 

 edge. In the eclipse of January last, at the time when tho 

 moon was near the centre of the shadow, careful observation 

 showed that tho marginal parts of the moon were brighter 

 all round than the more central parts. Much moro obvious 

 was tho gradually increasing glow of light, towards the 

 close of totality, on that part of the moon's edge which was 

 about to emerge. This, indeed, is always an interesting 

 feature in a total eclipse. I believe it is an invariable rule 

 that whenever the moon is at all near the edge of tho 

 shadow, a marked increase of light may be observed on the 

 parts which are nearest tho edge ; and even when tho moou 

 is not very far from tho centre of the shadow, it is generally 

 possible to judge, by the distribution of the light, on which 

 side the centre lies. 



I have never seen a satisfactory explanation of this. In 

 attempting to explain it, I would first call attention to the 

 fact, obvious on inspection of the diagram, that the solar 

 rays which strike the edge of any given transverse section 

 of the shadow are the same rays which strike the centre of 

 a transverse section further down the shadow. They are, 

 therefore, in accordance with a law already explained rays 



