180 THE SUN — ITS CHEMICAL ANALYSIS. 



ment, we easily deduce that the air of the apartment held in suspen- 

 sion but one twenty-millionth of its weight of sodium. Admitting 

 that a second suffices for conveniently observing the reaction, and 

 that during this time the flame uses fifty cubic centimetres or 0.0647 

 grams of air, containing only the twenty-millionth of a milligram of 

 salt of soda, we may calculate that the eje very distinctly perceives 

 the presence of less than the three-millionth of a milligram of salt of 

 soda. In view of such a sensibility as this, we recognize that it 

 must be seldom, indeed, that atmospheric air, at a high tempera- 

 ture, does not present the reaction of sodium. The surfece of the 

 earth is more than two-thirds covered with a solution of chloride of 

 sodium, which, by the agitation of the waves, continually produces 

 spray; the minute drops of water thus diffused in the atmosphere 

 abandon in evaporating a highly attenuated dust of chloride of sodium, 

 which constitutes an atmospheric element, variable as regards the pro- 

 portion, but rarely, it would seem, deficient in the air.'' 



Nothing is easier than to produce this apparition of the yellow line 

 of sodium in the spectrum obtained by the comparatively obscure 

 flame of an ordinary gas-lamp. While my eye was applied to the 

 glass with which I was observing the faint spectrum of such a lamp, 

 M. Grandeau, the chemist, who kindly repeated for me at the labora- 

 tory of the normal school the experiments of MM. Bunsen and Kir- 

 choff, struck several times with his hand on the sleeve of his coat, and 

 I saw the yellow line of the sodium display itself in a momentary 

 flash on the dim field of the eye-glass. The stroke of a hand on a 

 garment had sufficed to throw into the illuminating gas some of the 

 molecules of sodium mixed with the dust, and these scanty molecules 

 had instantly exerted their almost magic influence on the properties 

 of the light. M. Grandeau, at the time when he initiated me in the 

 experiments of the two German savants who had themselves given, 

 at Heidelberg, an account of their extraordinary researches, was en- 

 gaged in analyzing the mineral water of Bourbonne-les-Bains, and 

 had just detected therein the two new metals which MM. Bunsen 

 and Kirchoff" had discovered in the water of Diirckheim. He took 

 some drops of the water of Bourbonne, introduced them into the 

 flame, where they were at once converted into vapor, and I had quite 

 enough time to distinguish on the field of the spectrum the lines which 

 characterize the new metals, rubidium and ccesiiim, the red line of the 

 first and the blue of the second. 



It was, in fact, solely through the inspection of different spectrums, 

 which they obtained by introducing divers substances into a flame, that 

 MM. Bunsen and Kirchoff were led to the discovery of these two new 

 simple bodies. Familiar with the bright lines characteristic of all the 

 known metals, they were warranted in attributing to new metals the 

 bright lines which corresponded neither with iron, nor sodium, nor 

 lithium, nor potassium, &c. Guided by this induction, they were en- 

 abled to seek directly for these metals in substances which called forth 

 in the spectrum the appearance of the new lines. Hence it was that 

 they extracted cmsiuni from the mineral water of Diirckheim, and 

 rubidium from a mineral of Roxena, in Moravia, called by mineralo- 



