210 
POPULAR SCIENCE REVIEW. 
The image produced by decomposing a white sunbeam 
consists of certain brilliantly-coloured rays, but those rays 
are crossed by spaces giving no Light. The dark lines are 
always found in the same places in the solar spectrum, but they 
vary in number under different aspects of the Sun and varying 
conditions of the Earth’s atmosphere. When the Sim shines in 
its meridian splendour from a clear sky, the number of dark 
lines is slightly different from those observed when the Sun, 
being near the horizon, has to penetrate a greater depth of 
atmosphere. “ It is,” says Dr. Gladstone, “ a most beautiful 
and striking sight to observe the gradual appearance of these 
characteristic lines as the Sun descends towards the horizon,” 
proving that some of these non-luminous spaces are due to 
terrestrial atmospheric absorptions. To quote again the same 
authority, “ That the Earth’s atmosphere has much to do with 
the manifestations of those hues, is beyond all question, and the 
analogy” (alluding to some very striking experiments made by 
Dr. Miller) “of such gases as nitrous acid or bromine vapour, 
suggests the idea that they may originate wholly in the air that 
encircles our globe.” 
This suggestion is italicised for the purpose of giving our 
readers the full force of the evidence in favour of the views that 
the majority of those lines are of solar and not of terrestrial origin. 
The spectra, obtained from some artificial sources of Light, 
exhibit the coloured rays shading one into the other; while 
those produced by some others, consist of a series of luminous 
bands, separated by dark spaces ; and these luminous bands are 
frequently found to coincide with the darlc lines of the solar 
spectrum. 
Dr. W. A. Miller observed, that an intense yellow ray obser- 
vable in the spectra, obtained from the flames coloured with 
soda, lime, strontia, baryta, zinc, iron, and platinum, — and, 
according to Angstrom, in the electric light of every metal 
burnt by him, — had the same refrangibility as the hue D in 
the solar spectrum. 
“ But the most remarkable case occurs when carbon or 
sulphur is burnt in nitre. The brilliant Light, when analysed 
by a prism, exhibits a spectrum about as long as that of the 
sun at noon-day, but marked by bright lines, among which 
three are particularly prominent, respectively violet, yellow, 
and red in colour. The violet ray is not quite so refrangible 
as the solar H ; but the yellow is coincident with D, and the 
red with A ; while between the red and yellow appear at times 
fainter lines, one of which coincides with B, and a bundle some- 
times appears in about the position of A.” — (Brewster and 
Gladstone.) 
Pyrotechnic displays will have made the least scientific of our 
