64 THE CHEMISTRY OF THE SUN. [CHAP. 



of which synchronize with their own ; the light will reach us 

 deficient in those rays and will present a dark line for each ray 

 so absorbed. 



At the same time that the Fraunhofer lines in the solar 

 spectrum were shown to be in the main due to the presence of 

 vapours in the sun, the chemical nature of which vapours we 

 could recognise by this means ; the idea of Brewster that some 

 \vere produced by an absorption due to the atmosphere of the 

 earth, was fully confirmed. This fact, as we have seen, was first 

 suspected because certain lines were found to vary with the 

 height of the sun. They were very faint when the sun was high, 

 and got thicker and darker as the sun approached the horizon. 



In continuation of Brewster' s work, Professor Piazzi Smyth l 

 experimented at the Peak of Teneriffe in 1856. At one station, 

 Guajara, 8,903 feet above the sea, he found that the spectrum 

 when the sun was near the zenith began between a and B, and 

 with the exception of c and D, contained no marked lines 

 between B and E. When the sun was near the horizon the 

 spectrum commenced outside A, and contained many powerful 

 groups of lines between A and B ; the thickness of B was quad- 

 rupled, while c remained unaffected ; a number of fine lines 

 between c and D were greatly thickened, and many new lines 

 appeared beyond D. So striking was the change that at sun- 

 rise he could see the spectrum lengthen out and the new lines 

 grow visibly under his eye. 



At the sea-level the zenith spectrum approximated to the 

 horizon spectrum as observed on the mountain, but there were 

 some minor points of difference. 



At the violet end, Professor Smyth found that with a high sun 

 the spectrum extended beyond H, but when the sun was low it 

 terminated between H anda. At the sea-level it stopped imme- 

 diately beyond H, the two bars of which were nebulous ; while 



1 Phil Trans. 1858, part ii. pp. 503-507. 



