406 Professors Liveing and Dewar on Sun-spots 
regions of the sun's atmosphere where the temperature is high 
enough to make the hydrogen luminous, but where the pressure, 
if we may judge by the width of the lines, is certainly very low. 
But besides the disappearance of some lines in some spots, 
new lines and bands frequently make their appearance. Now 
of the Fraunhofer lines only a fraction have as yet been iden- 
tified as corresponding to lines of terrestrial elements; but we 
are inclined to the opinion that this is mainly due to the very 
imperfect examination which has as yet been made of the 
spectra of terrestrial substances. At any rate, we have found 
that it is only necessary to examine with high dispersive 
power any small section of the spectrum of the arc while 
different chemicals are dropped into it, in order to see a vast 
number of new lines develop themselves which have been 
hitherto unrecorded. The lowest horizon of the diagram 
shows the results of a somewhat hasty examination of the lines 
developed in this way by titanium and cerium, and in a few 
places by other substances. The titanium lines are permanent, 
but many of the cerium lines are evanescent : they come out 
strongly when fresh cerium is introduced, sometimes as broad 
bands, and quickly vanish as the cerium is dissipated, a few 
lines only remaining. The lines which remain are the lines 
of vapour of low tension, or long lines, while the others we 
suppose to be lines of vapour of high tension. It seems very 
probable, then, that if a careful examination were made of all 
the lines developed in the arc by all the known substances, 
most, if not all, of the Fraunhofer lines now ascribed to un- 
known substances would be accounted for. It is not a little 
remarkable that several of the most striking developments of 
lines observed at Greenwich in sun-spots closely correspond 
to new lines which we have observed to be given in the arc by 
cerium or titanium. In the case of cerium, the correspondence 
is so marked that it can hardly be accidental. Of course it is 
always easy to account for lines in the sun by the supposition 
of unknown elements brought up from the interior, or of 
unknown combinations or decompositions occurring under cir- 
cumstances which wo have not learned to imitate ; but that 
is to cut the knot, not to untie it. When we consider the 
influence which circumstances besides temperature have on 
the vibrations which a given substance can assume, the com- 
paratively partial way in which the spectra of our chemicals 
have been examined, the tendency of recent observations such 
as the artificial production of auroras with the characteristic 
auroral line at no very great elevation in our atmosphere, and 
the identification by Egeroff of the groups A and B with the 
absorption of oxygen, we shall hesitate to say that there is 
