Celestial Chemistry. 389 



But here a difficulty arose, some of the metals had not had 

 their spectra examined with the necessary precision, while the 

 very minute and accurate maps given by Kirchhoff were not 

 equally complete for all parts of the spectrum ; besides, in 

 these maps the comparisons being with the lines of the solar 

 spectrum, they were not so available for night work. Mr. 

 Huggins then determined on constructing his own maps of the 

 metallic spectra, and struck out the idea of adopting the bright 

 lines of the air spectrum, always obtained when the induction 

 spark is taken in air, as his standard of measurement and of 

 reference for the metallic lines. He therefore entered on a 

 long and elaborate set of experiments, rendered more trouble- 

 some by the extreme difficulty of obtaining the chemical ele- 

 ments in a state of absolute purity, which made it necessary, in 

 many cases, for him to procure the metals by electro deposition 

 from their solutions. For the purpose of this investigation, 

 Mr. Browning- of the Minories, so famous for his spectroscopes, 

 supplied six prisms of very dense glass, which were arranged 

 in an apparatus, the form of which will be understood from the 

 account of such instruments given in the November number. 

 The metals were volatilized by being made the electrodes of a 

 very powerful induction coil. 



By means of a small prism, covering one-half the slit of the 

 collimator, the air spectrum could, when necessary, be seen 

 simultaneously with that of the metal under examination. For 

 many months did Mr. Huggins carry on these difficult and 

 delicate experiments, until, in December, 1863, he was able to 

 communicate the results of his labours in a paper to the B-oyal 

 Society. This paper contains tables of measures, and diagrams of 

 the lines in the spectra of common air, and of twenty-four metals, 

 including sodium, potassium, calcium, barium, strontium, man- 

 ganese, tellurium, thallium, gold, tin, iron, silver, cadmium, 

 antimony, bismuth, mercury, cobalt, lead, zinc, arsenic, chro- 

 mium, platinum, osmium, and palladium. 



Independent of the value of these observations for future 

 researches on the stars, they were also productive of the dis- 

 covery of new lines in some of the spectra observed, among 

 which may be mentioned several lines in the sodium spectrum, 

 in addition to the well known double one corresponding to D 

 of the solar spectrum. Armed with these new materials, Mr. 

 Huggins prepared to resume his operations on the heavens, 

 being again assisted in the stellar observations by Dr. Miller, but 

 making those on the nebulae, which have resulted in the im- 

 portant discovery we shall presently mention, exclusively him- 

 self; and of these stellar observations we shall now present the 

 leading features. 



Before giving in detail the important results which have 



