1882.] producing the reversal of spectral lines of metals. 265 



as if the fundamental vibrations, i.e. those most easily assumed by 

 the molecule of the metal in a free state were in the higher region 

 of the spectrum. As if in fact they might be vibrations of which 

 we have as yet perceived only lower harmonics. The great line of 

 magnesium at wave-length 2852 certainly seems a fundamental 

 one, if we may judge by its persistence, strength, and facility of 

 reversal. 



It is, however, by no means always the strongest lines which 

 are most easily reversed, but rather those which are both per- 

 sistent and strong. The ultra-violet lines of iron, which are so 

 expanded and reversed in our photographs, are present, but not 

 at all specially conspicuous, amongst the lines in the spectrum 

 of the spark, but they are strong lines always in the arc. It 

 seems as if in the spark, especially when a Leyden jar is used, 

 violence were done to the metals, different in some way from that 

 due to mere elevation of temperature, which gave them violent 

 vibrations but not exactly those which they most readily take 

 up when under less constraint. It is remarkable how easily the 

 well-known pair of aluminium lines between H and K are reversed 

 in the arc as well as the other pairs near S and above, and yet the 

 metal in the arc does not shew at all either the triplet near N 

 (w. 1. about 3585, 3598, 3605) or the lower lines which are strong 

 in the spark spectrum. The quantity of metallic vapour is of 

 course much less in the spark than in the arc, but it is much more 

 violently, probably much more irregularly, agitated. 



A very important factor in chemical action is the pressure to 

 which the materials are subject, and we have not overlooked its 

 importance in regard to solar chemistry. The difficulties of 

 operating at pressures differing much from that of the atmosphere 

 have hitherto delayed the extension of our observations in that 

 direction, but we hope that those difficulties will not prove in- 

 superable and that we shall be able on a future occasion to lay 

 before the Society some observations on the effects of pressure on 

 the phenomena of reversals. 



