22 Turner, 7\nal Solar Eclipses. 



disturbing cause in addition to the sun's rotation. But 

 for considerations of this kind to have any value, the 

 velocities of ejection must be of the same order as that 

 of a particle rotating at the sun's equator, which is about 

 one mile per second. And for this to be effective in 

 carrying it out to several radii from the sun, we must 

 suppose the attractive force diminished by light pressure 

 to the ioo,OOOth part of itself — a supposition which has 

 not much to recommend it. 



To sum up, then, the result of this discussion : — 

 The observed falling off in light of the Corona must 

 be referred to the fact that light pressure nearly, but not 

 quite, balances gravity for the particles forming the 

 corona ; that the difference accordingly varies in magni- 

 tude ; that there are many more particles for which this 

 difference is large than for which it is small ; and that, 

 without any further considerable diversity in conditions, 

 particles ejected vertically from the sun's surface with 

 velocities similar to those observed in the granules might 

 then distribute themselves according to the observed law. 

 It must be remembered that we have not taken account 

 at all of the incandescence of the particles, which may 

 vary according to almost any law of distance. We have 

 started from the fact that the corona shows strong polar- 

 isation — in other words, that scattered light must play, 

 at any rate, a considerable part in the appearance — and 

 limited the enquiry to this part of the received light. 



The discussion is therefore partial only, and reaches 

 no final conclusion. But it is, perhaps, the more character- 

 istic of eclipse work. We are only slowly spelling out, 

 with long intervals between each letter, the lessons to be 

 learnt from the Corona ; and even those we seem to have 

 mastered we must be ready to modify in the light of new 

 facts. 



