422 THE CHEMISTRY OF THE SUN. [CHAP. 



motion from any but the spot-zones ; the fact that the faculae 

 always follow the formation of a spot and never precede it ; that 

 the faculous matter lags behind the spot as a rule ; the existence 

 of veiled spots and minor prominences in regions outside the 

 spot-zones; the general injection of unknown substances into 

 the lower levels of the chromosphere which I first observed in 

 1871, and which have been regularly recorded by the Italian 

 observers since that time ; all these phenomena, and many 

 others which might be referred to, are demanded by the 

 hypothesis, and are simply and sufficiently explained by it. 



If then the most vividly incandescent portion of the metallic 

 prominences follow indirectly from the fall of material, as they 

 have been shown to do, they must be produced by violent 

 explosions due to sudden expansions among this cooler material 

 brought down to form the spots, when they reach the higher 

 temperature at and below the photosphere level. 



7. By the hypothesis the effects produced must be on a scale 

 transcending those we should expect from the action of heat on 

 the chemical elements we are familiar with here. 



If we assume that metallic iron can exist in any part of the 

 sun's atmosphere, and that it falls to the photosphere to produce 

 a spot, the vapour produced by the fall of one million tons 1 will 

 give us the following volumes : 



Volume in 



Temperature. Pressure. cubic miles. 



2,000 C 380mm 0-8 



10,000 .... 760 .... 1-8 



20,000 .... 5atmos 0'7 



50,000 .... 760 mm. . . . . 80 



50,000 .... 190 .... 35-2 



If we assume the molecule of iron to be dissociated ten times 

 1 The volume of thess = '00003128 cubic miles. 



