THE SOLAR SPECTRUM IN 1877-1878. 289 
of those violet-hot furnaces that the more gigantic, as well as the numerous 
thinner lines, and all of them sharp-edged and well-defined, appeared with 
almost a personality, in their most marked physiognomies, about them; and 
were found continually multiplying with every further advance in the direction 
_ of greater refrangibility. This continual increase in the number of the lines 
almost reminded one, beginning at the red-end of the spectrum, of the few and 
far between individuals, you see by rail-road side when travelling through wild 
hills and Caledonian moors; and then as you go on through the spectrum 
colours, the lines increase, just as the human figures do when you approach the 
house-covered suburbs of London. While finally the spectral scenes of closely 
standing ranks behind ranks of lines between little # and great H, is like the 
crowds of passengers in Cheapside itself. And further, when at last all light 
itself fails and there is no more spectrum to be seen by the human eye 
through glass lenses,—you feel that you have not passed beyond, or even 
arrived at, by any means the last, or the least, or the most closely packed 
of those peculiar existencies, but are more probably left in the very thick 
of their legions still. 
Then how important in appreciating, or rather merely approximating to, 
the temperature of the Sun, the study of the mere statistical arrangement of the 
lines in its spectrum must be; if, nmdeed we can depend on the chemical 
elements being similar there, on the whole, to what they are on the earth! 
So at least I thought on that magnificently practical occasion; knowing 
well what is already so often remarked among spectroscopists, that in chemical 
investigations so long as we confine ourselves to the temperature of lamp-flame, 
the observable spectral lines are chiefly in the red, and only 2 or 3 of them 
are identifiable with solar lines; while when we introduce the higher 
temperature of the electric spark, an immediate increase of lines in the green 
and blue is noticed, and a far greater number of solar correspondences 
obtained. 
Having therefore collected about 5000 spectroscopic observations from 
various sources, and reduced them all to one and the same absolute scale 
of Wave-number, I have sub-divided them into the following six steps of 
temperature. 
Step 1. Lowest temperature, much below freezing. This being afforded 
by the telluric lines in the day-light sky spectrum, when a ray of day-light from 
beyond traverses a long stratum of the earth’s absorbing atmosphere in the 
upper regions. The data here are only 116 in number, and are partly taken 
from ANncsTRom’s solar chart, and partly from my own observations. 
Step. 2. Ordinary vital temperature, say 68° F, This has been presented 
by the absorption spectra of glasses and liquids printed by the Royal Society, 
Edinburgh, in my paper of last year ; the data being 719 in number. 
VOL. XXIX. PART I. 45 
