and on those produced by the Earth's Atmosphere. 527 



gation were two very fine rock-salt prisms, executed by myself; a 

 large hollow prism made of plates of parallel glass for holding 

 fluids ; a fine plate glass prism, executed by FRAUNHOFER, and 

 which I owe to the kindness of Mr TALBOT ; a copious supply 

 of oil of cassia and oil of cinnamon, which Mr GEORGE SWINTON 

 transmitted to me from Bengal with his usual liberality ; a good 

 achromatic telescope, by BERGE ; and an excellent wire microme- 

 ter by TROUGH-TON. To this apparatus Mr ROBISON made two 

 important additions, which he executed with his own hands, the 

 one a brass stand with a variable aperture for admitting the inci- 

 dent light, and the other a stage for holding and adjusting the 

 prisms in front of the object glass ; and I have recently been fa- 

 voured, by Sir JAMES SOUTH, with the use of his fine five-feet 

 achromatic telescope, executed by DOLLOND. 



After a little practice in the observation of the solar spectrum, 

 I discovered most of the lines, which I had in vain sought for in 

 FRAUNHOFER'S map, as the counterpart of those in the gaseous 

 spectrum. I saw well marked groups, of which he had only 

 given one of the lines, and shaded bands, and well defined lines, 

 which his methods of observation had not permitted him to dis- 

 cover. After I had laid down all the principal features on the 

 spectrum, I was able to examine the two classes of lines pari 

 passu. The action of the gas upon invisible lines in the spec- 

 trum rendered them visible by slightly enlarging them, and this 

 enlargement of a solar line indicated the existence of a corres- 

 ponding line in the gaseous spectrum. 



By this double process, and by methods of observation which I 

 believe have never before been used in optical researches, I have 

 been able to execute three different maps of the spectrum ; first, 

 a map of the lines in the solar spectrum ; secondly, a map of the 

 same spectrum, exhibiting at the same time the action of nitrous 

 acid gas upon solar light, previously deprived of a number of its 

 definite rays ; and, thirdly, a map shewing the action of the gas 



