CHEMICAL SCIENCE. 



CHEMICAL ANALYSIS BY OBERVATIOXS OF SPECTRA, 



IT is well known that many substances when introduced into a flame pos- 

 sess the property of causing in the spectrum certain bright lines. Bunsen 

 and Kirch hoff have based upon these lines a method of qualitative analysis 

 which materially extends the domain of chemical reactions, and leads to the 

 solution of many difficult problems. In a memoir recently published (and an. 

 abstract of which may be found in Silliman's Journal for November, 18GO), 

 the authors develop in detail the method to be pursued in the examinations 

 of the metals, of the alkalies, and alkaline earths. They show, in the first 

 place, that the different states of combination of the metals examined, as 

 well as very great differences of temperature in the flames produced, exert 

 no influence on the position of the spectral lines corresponding to the par- 

 ticular metals. The same metallic compound gives a spectrum which is the 

 more intense the higher the temperature of the flame; moreover, the most 

 volatile compound of any particular metal always gives the greatest intensity 

 of light. 



When small pieces of potassium, sodium, lithium and calcium are attached 

 to the extremities of fine platinum wires enclosed in glass tubes, and the spark 

 of aRuhmkorff's induction -apparatus is allowed to pass from one pole to 

 the other, the spectra are found to contain the same bright lines as the 

 flames. From this, it appears that these bright lines may be looked upon as 

 certain indications of the presence of the metals in question. They serve as 

 reactions by which these substances may be recognized more sharply, more 

 quickly, and in smaller quantities, than by any other analytical process. 



This analysis of the chemical composition of substances promises to furnish 

 also a method of investigating the chemical nature of the atmospheres of the 

 sun and of the brighter fixed stars. KirchhofF has shown, from theoretical 

 considerations, that the spectrum of an ignited gas is inverted when a source 

 of light of sufficient intensity, and giving a continuous spectrum, is placed 

 behind it. In other words, the bright lines are under these circumstances 

 converted into dark ones. From this it appears that the solar spectrum, with 

 its dark lines, is nothing else than the inversion of the spectrum, which the 

 atmosphere of the sun would show by itself. The chemical analysis of the 

 sun's atmosphere requires us, therefore, only to determine what substances 

 introduced into a flame will produce bright lines, corresponding to the dark 

 lines in the sun's light. The authors have verified by direct experiment the 



