Light in the Ultra-violet part of the Spectrum. 135 



the average quantity in a vertical column of the atmosphere, caused 

 an absorption similar to that observed in the solar spectrum, namely, 

 terminating about X 2950. 



Quite recently Liveing and Dewar have made some important 

 experiments on the absorption-spectrum of large masses of oxygen 

 under pressure.* They state that with a tube 165 cm. long and a 

 pressure of 85 atmos., oxygen appeared to be quite transparent for 

 violet and ultra-violet rays up to about X 2745. From that point the 

 light gradually diminished, and beyond X 2664 appeared to be wholly 

 absorbed. 



In some later experiments with a steel tube 18 metres long and a 

 pressure of 90 atmos., oxygen produced complete absorption above P, 

 i.e., X 3359-2. 



M. Janssen, from his observations on the Alps, concludes that both 

 the bands which follow the law of the square of the density, and the 

 dark lines obeying a different law of formation, which are due to 

 oxygen in the solar spectrum, are produced exclusively by the earth's 

 atmosphere — " L'atmosphere solaire n'intervient pas dans le pheno- 

 mene."f 



, .... ,. v 



SPECTRA OF VEOA 



skewing titt limit" oj stellar licjki-m.4liJt ulfra viokbrecjion. 



* ' Chemical News,' vol. 58, p. 163, and < Phil. Mag.,' September, 1888, pp. 286— 

 290. 



f ' Comptes Rendus,' vol. 107, p. 677. 



