244 



Profs. G. D. Liveing and J. Dewar. 



[May 31, 



its maintenance in the metallic state. As regards temperature, it 

 should be observed that while substances merely heated by the flame, 

 and not undergoing chemical change, are not likely to rise to a tempe- 

 rature above the average temperature of the flame, it will be 

 otherwise with the materials of the flame itself and other substances 

 in it which are nndergoing chemical change, and have at the instant 

 of such change the kinetic energy due to the change. 



In a recent communication to the Society, "Researches on the 

 Spectra of Meteorites," Mr. Lockyer has directly connected the 

 appearance in nebulae of these bands, namely, " the magnesium fluting 

 at 500" with the temperature of the Bunsen burner ('Roy. Soc. 

 Proc.,' vol. 43, p. 133). That the bands are persistent through a large 

 range of temperature there is no doubt, bnt we cannot help thinking 

 that Mr. Lockyer is mistaken in supposing them to be produced at 

 the temperature of a Bunsen burner. It does not follow because the 

 bands are seen when magnesium is burnt in a Bunsen burner that the 

 molecules which emit them are at the temperature of the flame. In 

 the combustion of the magnesium the formation of each molecule of 

 magnesia is attended with a development of kinetic energy which, if 

 it all took the form of heat and were all concentrated in the molecule, 

 must raise its temperature to very nearly the point at which magnesia 

 is completely dissociated. The persistence of the molecule of magnesia 

 when formed will depend upon the dissipation of some of this energy, 

 and one of the forms in which this dissipation occurs is the very 

 radiation which produces the bands. The character of the vibration 

 depends on the motions of the molecules, which in the case in question 

 are not derived from the heat of the flame, but from the stored energy 

 of the separated elements, which becomes kinetic when they combine. 

 The temperature of complete dissociation of magnesia is very far 

 higher than any temperature which can reasonably be assigned to the 

 Bunsen burner. 



Nor do the observations we have made on magnesia in the oxy- 

 hydrogen flame appear to us to be inconsistent with the conclusion 

 that the spectrum of the oxide is produced only at a high temperature, 

 as we have a decomposition of magnesia by the hydrogen at the 

 highest temperature of the blowpipe flame, and when hydrogen is in 

 excess little but the metallic lines is visible, because the re-formation 

 of magnesia is, for the most part, the reversal of the former action, and 

 occurs in the cooler part of the flame by the interchange of oxygen 

 between steam and magnesium with scarcely any rise of temperature. 

 On the other hand, when the oxygen is in excess the reduced magne- 

 sium carried up into the flame combines for the most part directly 

 with oxygen, and individual molecules thereby acquire a motion of 

 far greater intensity than they could derive from the average heat of 

 the flame. 



