336 M. A. J. Angstrom's Optical Researches. 



observations began. I made use of the same flint-glass prism 

 as before, so that the results obtained, strictly speaking, are only 

 valid for those kinds of light for which the flint-glass is diapha- 

 nous. This is the more necessary to be mentioned here, as it is 

 easy to forget how much glass in general absorbs the extreme 

 ends of the spectrum. I must also mention that the gases made 

 use of were not perfectly pure, although I do not believe that an 

 erroneous result could be produced by the impurity present; 

 the reader will be able to infer this from the statement of the 

 manner in which the gases were prepared. 



12. Oxygen. — The gas was obtained from chlorate of potash, 

 and was introduced directly into the apparatus. The spectrum 

 obtained in this instance is shown at 0, fig. 3. If this spectrum 

 be compared with that of the atmosphere, it is seen, — 1st, that 

 the strongest lines of the air spectrum, which in fig. 1 are de- 

 noted by 7, 8, D, E, have disappeared ; and 2nd, that the blue 

 and violet field exhibit new lines which do not appear in the 

 other case. 



With regard, further, to the metal lines in the case of copper 

 and zinc, they were almost imperceptible at the end of the expe- 

 riment. This was particularly the case with the zinc in the blue 

 field. From this it seems that we may conclude, that the bright 

 lines corresponding to the metal are not produced by the oxida- 

 tion of the latter, for if this were the case, they must show them- 

 selves strongest in oxygen. Nevertheless the surface of the brass 

 spheres showed that an oxidation had taken place, and it is pos- 

 sible that the cause of the phaenomenon alluded to is to be sought 

 here. 



13. Carbonic Acid. — The gas was prepared in the ordinary 

 way by means of sulphuric acid and chalk, and was carried, for 

 the purpose of drying it, through a bottle filled with pumice- 

 stone which had imbibed sulphuric acid. The spectrum obtained 

 is shown at CO 2 in fig. 3 ; it resembled perfectly the spectrum of 

 oxygen, as far as the strong lines in the blue and violet are con- 

 cerned. Some difference was, however, observed in regard to 

 the weaker lines ; a bright band, corresponding to 13, fig. 1 of 

 the air spectrum, was also observed, which did not appear in the 

 oxygen spectrum. Nevertheless both spectra might be regarded 

 as identical. This is easily explained by the fact, that, according 

 to Berzelius, the electric spark decomposes the carbonic acid 

 into carbonic oxide and oxygen, and the latter gas produces its 

 own peculiar lines in the spectrum. 



As it was possible that the violet and blue lines in the oxygen 

 spectrum, and in that of carbonic acid, might have been produced by 

 the oxidation of the metallic particles during their motion through 

 the gas, and as, if this were the case, they must disappear if an 



