46 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 52 



The considerations (3), (5), and (7), taken together, are hard 

 to satisfy; for if the inner corona were hot enough to give out a 

 spectrum of incandescence satisfying (5), the matter composing it 

 must be gaseous, if it is Hke any matter we know of.^ Accordingly 

 we should expect a bright line spectrum like that of the chromo- 

 sphere if the inner corona shines chiefly by incandescence,^ and, fur- 

 thermore, we should expect its rays to increase in transmissibility to 

 the screen and grow red to the eye with increasing distance from the 

 sun. 



If we may suppose that the temperature of the corona is every- 

 where low enough to allow solid or liquid particles to be formed, 

 then all the specifications excepting (3) are easily satisfied by the 

 hypothesis of a corona of reflection.^ Our knowledge is not sufficient 

 to enable us to prove that the particles even of the inner corona 

 would be too hot to be mainly liquid (that is to say, above 3,000° to 

 3,500°). If the particles were all gaseous, the rays reflected would 

 probably be richer than sun rays in visible light, and this would be 

 contrary to (5). May it not be that while a large proportion of the 

 particles of the inner corona is gaseous, a considerable proportion 

 is liquid or solid? Then may not the light of the inner corona be 

 mainly reflected, like that of the outer corona, but with the bright 

 line spectrum of incandescent gases present in sufficient strength to 

 nearly obliterate the dark Fraunhofer lines of the reflected sun rays? 

 The continuous spectrum of the incandescent solid and liquid par- 

 ticles present would tend to increase the transmissibility of the coro- 

 nal brightness to the asphaltum screen ; so that the opposite tendency 

 of the diffuse reflection of the gaseous particles present would be 

 counteracted. At increasing distances from the limb we may suppose 

 the particles would be cooler, and mainly solid or liquid, so that in- 

 candescence would wane and a dark line spectrum would gradually 

 appear. Still, the transmissibility and color would remain nearly 

 unchanged, because the light would be still mainly reflected sunlight, 

 and the particles now so large as not to enrich the proportion of blue 

 light, but rather slightlv to decrease it- 



' Arrhenius computes a possible temperature of 4,620° at 0.7' from the limb, 

 and then suggests that the matter there may be liquid drops. How is this 

 possible? 



*The gaseous material of the sun itself is under enormous pressure, so 

 that its spectrum is thereby made continuous. Not so the corona. 



' Specification (4) is no obstacle, because the particles near the sun receive 

 light from a solid angle of nearly a whole hemisphere, and would therefore 

 show no polarization in any particular direction, because partially polarized 

 in all. 



