428 



Profs. Gr. D. Liveing and J. Dewar. [Mar. 1 6, 



posures and with more sensitive plates, to obtain information on this 

 and other points. It is, perhaps, not too much to hope that the 

 further knowledge of the spectrum of the nebulse afforded us by photo- 

 graphy, may lead by the help of terrestrial experiments to more 

 definite information as to the state of things existing in those bodies. 



III. " On the Disappearance of some Spectral Lines and the 

 Variations of Metallic Spectra due to Mixed Vapours.'* By 

 G. D. Liveing, M.A., F.R.S., Professor of Chemistry, and 

 J. Dewar, M.A., F.R.S., Jacksonian Professor, University 

 of Cambridge. Received March 11, 1882. 



The theory 1 of spectral lines most commonly received is that the 

 otions of the luminif erous ether producing them are not directly due 

 to any motion of translation of the molecules of the emitting substance, 

 but to relative motions of the parts of the same molecule, or in other 

 words, to vibrations occurring within the molecules ; and that the 

 mutual action of the molecules, while it may give rise to irregular 

 vibrations of the ether, affects the regular vibrations producing the 

 lines only in an indirect manner, by converting part of the motions of 

 translation into internal vibrations. On this theory the spectral lines 

 which any given substance can readily take up will in general be 

 limited to a certain number of fundamental lines and a number of 

 others harmonically related to them, though not necessarily simple 

 harmonics of the fundamental lines. And variations of temperature, 

 by altering the rapidity and the violence of the action of one molecule 

 on another, will alter the intensities of the several vibrations, but not 

 their periods, unless the violence should extend to the disintegration 

 of the molecules, which would be equivalent to the formation of new 

 molecules with new fundamental periods of vibration. In view of 

 this theory, the observations on the spectrum of magnesium have a 

 special interest, because from the close analogy of magnesium to zinc 

 and cadmium, it is inferred that the molecules of magnesium vapour 

 are chemical atoms of that substance, that is to say, they pass appa- 

 rently undivided through all the chemical changes to which magnesium 

 may be subjected ; and it seems reasonable to suppose that any sub- 

 division of the chemical atoms could not fail in this case to be attended 

 with a change of chemical qualities, which, in the presence of other 

 elements, would give rise to new compounds. No such new com- 

 pounds have in fact been detected. We have already described in 

 detail the differences between the spectra of magnesium as seen in the 

 flame of the burning metal, the electric arc, and the spark discharge, 

 and we have now some further observations upon them to place before 

 the Society, which are confirmatory of the received theory. 



