forming the Low-Temperature Spectrum of Oxygen. 333 



that it fell in pieces of its own accord, until lie had a heap of 

 tunnelled fragments large enough to make a cautionary photo- 

 graph of. And when he had at last succeeded by his 

 unconquerable perseverance in securing two blocks that stood, 

 and the anti-prisms were to be fastened on their faces, he 

 tried almost every patented and unpatented cement before he 

 found one, or rather a particular method of using it, which 

 could withstand the action of the bisulphide for more than a 

 day or two. 



At last, however, though long after the contracted time, 

 he brought the two prisms here complete; and they passed 

 successfully through the severest trial I had prepared for 

 them, viz. that when a hydrogen-line was at its brightest, 

 shining like a ray of sunlight in a dark field, its light must 

 be entirely and sharply confined to the width of opening of 

 the slit for the time being. Xot, however, until some months 

 after, when the telescope power was also improved, and a new 

 class of difficulties with the prisms had been overcome, could 

 the desired trial on oxygen be made. low, whereas I had, 

 as mentioned already, on previous occasions with the best 

 compound glass prisms I could procure, seen only an uncertain 

 idea that some of the oxygen-lines might be double, I now 

 saw the real triplicity of four of the lines, and measured them 

 micrometrically with a degree of certainty and satisfaction 

 that I had never dreamt of with the older apparatus; and 

 this triplicity of these lines never came out more remarkably 

 than when the singularity of really single lines, such as those 

 of hydrogen impurities in the same gas-tube, was rendered 

 most distinctly. I will therefore now only seek to conclude 

 with a few words on the bearing of this tripleness of the 

 oxygen-lines, first, on the disputed question of the existence 

 of oxygen in the atmosphere of the sun, and, secondly, on the 

 absence of hitherto recognized oxygen-manifestations (though 

 oxygen is so well known to exist as a large part of the earth's 

 atmosphere) in the telluric rays that become visible in the 

 solar spectrum at sunrise or sunset. 



Oxygen in the Sun. 

 For many years it was a sort of crying wonder that the 

 spectroscope showed no traces in the sun of so necessary a 

 gas to combustion as oxygen — the expected test being, that 

 we should see there as dark lines all the lines which are seen 

 bright when oxygen gas, on being rendered incandescent in 

 a high electric temperature, then shows what is called its 

 " elemental-line spectrum." But not one of those lines could 

 be detected by its dark counterpart in the sun. At length 



