THE FUEL OF THE SUX. 91 



limits of the gaseous prominences. As the luminosity of these 

 glowing particles must be very smalll compared with that of 

 the photosphere, they will be invisible in the glare of ordinary 

 sunshine ; but if our eyes be protected from this, they may 

 then be rendered visible, both by their own glow and the solar 

 light they are capable of reflecting. They should be seen 

 during a total eclipse, and should exhibit radiant streams pro- 

 ceeding irregularly from different parts of the sun, but most 

 abundantly from the neighborhood of the spot regions. As 

 these spot regions occupy the intermediate latitudes between 

 the poles and the equator of the sun, the greatest extensions of 

 the outstreamings should be N. E. and S. W. and S. E. and 

 N. W., while to the N., S., E., and W. that is, opposite the 

 poles and equator of the sun there should be a lesser exten- 

 sion. The result of this must be an approximation to a 

 quadrilateral figure, the diagonals of which should extend in a 

 N. E. and S. W., and a S. E. aad N. W. direction, or there- 

 abouts. I say * ' thereabouts, " because the zone of greatest 

 activity is not exactly intermediate between the poles and the 

 equator, but lies nearer to the solar equator. 



Examined with the polariscope, these radiant streams should 

 display a mixture of reflected light and self-luminosity. Ex- 

 amined with the spectroscope, a faint continuous spectrum 

 due to such luminosity of solid particles should be exhibited, 

 with possibly a few lines due to the small amount of vapor 

 which, in their glowing condition, they might still give off. 

 Besides this, there should appear the spectroscope indications 

 of violent electrical discharges, which must occur as a neces- 

 sary concomitant of the furious ejections of aqueous vapor and 

 solid particles. All these metallic hailstones must be highly 

 charged, like the particles of vesicular vapor ejected from the 

 hydro-electric machine, or the vapors and projectiles of a 

 terrestrial volcanic eruption. 



I need scarcely add that this exactly describes the actually 

 observed results of the recent observations on the corona, and 

 that all the phenomena of this great solar mystery are but 

 necessary and predicable results of the constitution I ascribe to 

 the sun. 



There is a method of manufacturing hypotheses which has 

 become rather prevalent of late, especially among mathemati- 

 cians, who take observed phenomena, and then arbitrarily and 

 purely from the raw material of their own imagination con- 

 struct explanatory atoms, media, and actions, which are shaved 



