534 MEMOIR OF KIRCHHOFF. 



describe; ouly when tbe exiisting" words seemed uot precise euougb, he 

 uses oircumlocutioiis and definitions, and that mostly iu matliematical 

 language. He held the liigbest rank among those who strove to re- 

 move from exact sciences all want of clearness, all subjective judg- 

 ments, all phrases. The influence of such an endeavor will transcend 

 the limits of his particular science. 



The most popular of Kirchhoff's works is his spectrum analysis. It 

 had in fact most extraordinary consequences of the most palpable kind, 

 and has become of the highest importance for all branches of natural 

 science. It has excited the admiration and stimulated the fancy ol' 

 men as hardly any other discovery has done, because it has permitted 

 an insight into worlds that seemed forever veiled for us. It is accord- 

 ingly the most celebrated of Kirchhoff's discoveries. 



But however wonderful the results, what seems to us more admirable 

 still is the truly masterly work itself, the unusually keen aud at the same 

 time ingenious and diligent way in which Kirchhoff deduced from the 

 outset, from an accidental observation, a general theoretical law and all 

 the surprising inferences, and demonstrated them with full strictness 

 aud certainty. Great men had already held in their hand before him 

 the threads of his discovery without being able to unravel them. The 

 French as well as English brought forward and still produce claims 

 of priority. Kirchhoff repelled them quietly but firmly. All had seen 

 something, made guesses, considered as possible or probable (without 

 Kirchhoff" haviug been aware of it at the time, however). A solid basis, 

 a rigorous demonstration had been given by nobody ; it was reserved 

 to the acuteness, thoroughness, and perseverance of a German searcher 

 to elevate the lucky guess to the rank of a sure knowledge. 



Spectrum analysis in the narrowest sense, i. e., the " analysis of the 

 chemical elements by means of spectral observations," is due, if we 

 wish to make a distinction, to an idea and a suggestion of Bunseu's. 

 Among the most ingenious performances of Bunsen may be reckoned 

 certain very simple physical methods of qualitative chemical analysis, 

 i. e., the detection and the discrimination of chemical elements. A 

 characteristic re-action of this kind he found to be the coloring of non- 

 luminous flames. Each chemical element vaporized or burned in anon- 

 luminous flame, for instance a blue-burning gas-flame, imparts to it a 

 definite characteristic coloring. We should be able accordingly to 

 recognize each substance by the light its incandescent vapor emits if 

 our eyes had the power to distinguish as uumy differences of color as 

 there are substances in nature. Kirchhoff' and Bunsen helped the eyes 

 however by decomposing the light of flames into its separate colors by 

 means of a prism. This giv^es rise to the f-pectrum of tbe flame. The 

 rainbow is a natural spectrum of tbe solar light made by the raindrops. 

 But this spectrum, as well as tbe spectra of all glowing solid or liquid 

 bodies, offers quite another aspect from the spectra of flames, *. e., in- 

 candescent gases. The first consist of known colors varying iu a con- 



