M. A. J. Angstrom's Optical Resea7-ches. 339 



which here showed themselves upon the surface of the metal 

 only. It seems, therefore, that in this case the air alone trans- 

 mitted the electricity, but did not impart to the ordinary spec- 

 trum a sufficient strength of light, because the metallic particles 

 did not by their motion cause the particles of air to glow. It 

 would be well, however, to make this the subject of further in- 

 vestigation. To the same portion of the subject belongs pro- 

 bably a fact which I have several times observed, namely, that 

 when one of the poles is formed of a fluid, of sulphuric acid for 

 example, the lines in the corresponding half of the air spectrum 

 are much weaker than in the other. 



19. The experiments which I made with different metallic 

 sulphurets led to no remarkable result. In general they gave 

 the composition of the metal more or less completely. There 

 was no difference observed, whether the sulphuret received from 

 the apparatus positive or negative electricity. As examples, I 

 have given in fig. 2 the spectra for the sulphuret of lead and 

 sulphuret of iron. The latter, however, is distinguished by the 

 fact, that in the vicinity of the line j, and also in violet, bright 

 lines are shown which were not observed in my experiments with 

 iron. These lines could not, however, belong to the sulphur, 

 because they must then exhibit themselves with other metallic 

 sulphurets. 



Sulphur, like carbon, seems to have no peculiar lines. I have 

 made the experiment, not only with charcoal, but also with gra- 

 phite, and have obtained no lines which could be regarded of the 

 same character as those exhibited by the metals. The lines of 

 the air spectrum, indeed, particularly in the violet, appeared 

 stronger than ordinary ; and in several of the bright portions fine 

 lines were plainly visible (e. g. at H, fig. 1), but I have had no 

 reason to conjecture that these lines belong to the carbon, par- 

 ticularly as platinum also shows the violet end of the spectrum 

 in a peculiarly manifest manner. The deportment must be 

 referred to the fact, that the coal particles, by their finer division, 

 set more easily the air particles in a glow. 



20. Summing up the observations on the spectra of gases, it 

 follows that, in the oxygen spectrum, the greatest number of 

 bright lines occur in the blue and violet ; in the nitrogen spec- 

 trum, in green and yellow ; and in the hydrogen spectrum, in 

 red. These phenomena must necessarily be connected with the 

 chemical and thermic properties of the gases. Regarded from 

 the theory of undulation, the chemical activity must exert itself 

 chiefly in the motion of the molecules ; and it may be assumed 

 with some probability, that two bodies, whose vibrations are iso- 

 chronous, may combine more easily than two others. In this 

 way the possibility of a combination is finally referred to the 



Z2 



