426 MR WILLIAM SWAN ON THE PRISMATIC SPECTRA OF THE 



From the fact just stated, that most of the lines in the carbohydrogen spectrum 

 occupy positions where, in the solar spectrum, no conspicuous dark lines occur, 

 direct comparison of the spectra, by simultaneous observation, seems almost 

 impossible ; for, before the fainter lines of the carbohydrogen spectrum become 

 visible, the solar spectrum must be rendered so faint, that its finer lines have 

 disappeared. On the other hand, to make a complete comparison of the spectra by 

 actually measuring with the theodolite, the positions of the finest lines of the solar 

 spectrum, would be a most formidable task. For when we consider that Fraun- 

 hofer has represented on his map of the solar spectrum, 350 lines, while Sir 

 David Brewster, by the aid of very excellent optical means, has observed the 

 spectrum to be " divided into more than 2000 visible and easily recognized por- 

 tions, separated from each other by lines more or less marked,"* it follows, allow- 

 ing 5° for the angular dispersion of the extreme rays of the spectrum, — that the 

 average interval between the lines observed by him is only 9". Extremely de- 

 licate theodolite measurements would therefore be required, in order to determine, 

 whether or not any bright line of a flame spectrum was or was not coincident 

 with one or other of the numerous small lines of the solar spectrum ; and even 

 where a coincidence was ascertained, it might be fairly attributed to chance, just 

 as a binary star may be only optically, and not necessarily physically double. 



In cases, however, where there is a remarkable analogous configuration of 

 two groups of lines, accompanied by exact coincidence, as between the double lines 

 a and D ; and more especially where we actually view the striking phenomenon 

 of the lines in the spectra optically superimposed, the impression of some phy- 

 sical connection between the two groups becomes irresistible. 



The coincidence of 7, the most brilliant line of the carbohydrogen spectrum, 

 with the clear space immediately beyond 6 2 , — the most refracted line of a group, 

 which, whether we regard the singular configuration or the strength of the lines 

 which compose it, is perhaps the most notable in the solar spectrum, — is a phe- 

 nomenon which seems deserving of attention as probably indicating also some 

 physical relation. 



In conclusion, I may observe, that from the facility with which, by means of 

 the Bunsen lamp, the carbohydrogen spectrum may be obtained, and from the 

 definite and readily identifiable character of the lines which compose it, these 

 lines may be useful in optical researches, where, from any cause, sun light can- 

 not be employed. It will be seen, from Table IV., that, for most practical purposes, 

 the lines a, 7, d 3 , and £, may be assumed as identical with D, b 2 , F v and G of the 

 solar spectrum ; any error in the index of refraction calculated on that assump- 

 tion, affecting only the fourth or fifth place of decimals. — 7th June. 



* Edinburgh Trans., vol. xii., p. 528. 



