292 REPORT— 1880. 



could never change the relative intensity of two lines when the least 

 refrangible of them is the strongest. Tet we constantly find that a less 

 refrangible line is longer, but weaker throughout its length, than one which 

 is situated more towards the violet. If an increased thickness of luminous 

 matter was the cause of the appearance of the shorter line near the pole, 

 we should have in this case, for a larger thickness, the more refrangible 

 line the stronger; but in the centre of the spark, where there is little 

 luminous matter and where the least refrangible long line is the only one 

 seen, this would be the strongest. This cannot possibly be due to the effect 

 only of decreased thickness. The fact that the longest lines are those 

 appearing at lower temperature, though in the experiments of Lecoq and 

 Liveing and Dewar, the quantity of luminous matter is really larger than 

 when a high tension spark is employed, also disproves the theory that 

 thickness of the luminous layer has much to do with the explanation of 

 the long and short lines. 



The next explanation which we shall discuss, starts from the fact that 

 the longest lines become stronger when the temperature is reduced. A 

 metal at low temperatures has a certain number of lines ; as the tempera- 

 ture is increased other lines may come out, which may gain in intensity 

 and finally surpass the original lines. These lines coming out at higher 

 temperature would be the short lines, while the long lines would be the 

 low-temperature lines. This explanation accounts very satisfactorily for 

 a part of the phenomena, as for instance the disapjjearance of the long 

 lines when the pressure of the gas in which the spark is taken is reduced ; 

 for the temperature of the spark will be lowered in that case. Also the 

 fact that the longest lines are those which first reverse would flow 

 naturally out of the explanation given. But the explanation is not com- 

 plete. Why should a mixture of different elements only show the longest 

 line of that constituent which is present in small quantities ? In the 

 case of chemical combinations we might assume that the spark having to 

 do the work of decomposition is weakened, and that therefore the low- 

 temperature lines ai'e obtained. But this could no longer be if a chemical 

 compound is replaced by a mechanical mixture. There is no reason why 

 a spark taken from a mechanical mixture of two bodies should be cooler 

 than one taken from each body singly. Nor could the remarkable effects 

 of dilution, which we have already mentioned, be accounted for solely 

 on the supposition that the long lines are low-temperature lines. We 

 require aa additional hypothesis. Speaking of the widening of the 

 sodium lines it has already been suggested that, under the same pressure, 

 at the same temperature, and for the same number of radiating molecules, 

 we might have a difference in the spectrum if the molecules with which it 

 comes into collision are molecules of the same kind only or are chiefly 

 molecules of a different kind. I think we must have recourse to the same 

 explanation in order to account for the facts which are now before us. 

 For if an alloy shows us at the same temperature the longer lines only of 

 each constituent, this may at any rate be due to the fact that the shorter 

 lines are more easily produced by the molecules of the same kind than by 

 those of a different kind. We must, in fact, assume, in order to account 

 for the phenomena, that the spectrum of a molecule, when it is excited by 

 molecules of another kind, consists of those lines chiefly which a molecule 

 of the same kind is capable of bringing out at a lower temperature already. 

 It would follow from this that the effects of dilution are the same as those 

 of a reduction of temperature, which is the case. We shall speak of this 



