2l8 HOW TO WORK 



ON SPECTRUM ANALYSIS. 

 By H. C. SORRY, F.R.S., &c. 



318. The Spectrum Microscope. Spectrum analysis, as applied 

 to the microscope, must not be confounded with that branch 

 of the subject which has yielded such admirable results in the 

 hands of Bunsen, Kirchhoff, and other physicists. In that method 

 of analysis it is the number and position of the narrow 

 bright lines or bands, into which the light of the incandes- 

 cent body is divided by the spectroscope, that enable the experi- 

 menter to identify each different substance. It is, in fact, the 

 analysis of the emitted light, whereas in spectrum analysis applied 

 to the microscope, it is the analysis of light which has been modified 

 by transmission through the substance under examination, and it is 

 the absence, and not the presence, of particular rays which makes the 

 spectra characteristic of different substances. In this respect it is 

 more analogous to spectrum analysis as employed in studying the 

 chemical .nature of the atmosphere of the sun or stars, as illus- 

 trated by the researches of Kirchhoff, Miller, and Huggins, but 

 the principles involved are materially different. The absorption 

 bands in such cases are narrow, sharply defined lines, characteristic 

 of absorption by gases, whereas those which play such an important 

 part in researches with the spectrum microscope are usually broad, 

 gradually shaded off on each side, and only in a few cases so narrow 

 and sharply defined as to vie with some of the broader dark lines in 

 the solar spectrum.* 



Confining then our attention to spectrum analysis as applied to 

 solid and liquid substances, it may be said that the object of our 

 researches is to distinguish substances by their colour, studied in 

 the most accurate and scientific manner. Colour alone is, of course, 

 often made use of as a criterion in qualitative chemical analysis, and 

 is extremely characteristic of particular substances, even when seen 

 in the ordinary manner ; but when more accurately studied by 

 means of the spectroscope it becomes incomparably more charac- 

 teristic. The colour of a body, as seen with the naked eye, is the 

 general impression made by the whole of the transmitted light, when 

 all the rays are mixed together, and this total impression may be 

 the same, though the compound parts may differ in a striking manner; 



* Though I was the first to publish a paper OH spectrum analysts applied to the 

 microscope, after having made use of it in various researches for nearly a year 

 (Quarterly Journal of Science, April, 1865, vol. IT, p. 198), yet it is only fair 

 to state that Mr. Huggins had independently thought of such an application 

 (Trans. Microscopical Soc., May 10, 1865). [H. C. S.] 



