Sir David Brewster on the Colours of Natural Bodies. 473 
tints having any such relation as that which subsists between 
adjacent colours of the same order. 
From facts like these, which it is impossible to misinter- 
pret, we are entitled to conclude, that the green colour of 
plants, whether we examine it in its original verdure, or in its 
decaying tints, has no relation to the colours of thin plates. 
I have submitted to the same mode of examination nearly 
one hundred and fifty coloured media, consisting of fluids ex- 
tracted from the petals, the leaves, the seeds, and the rind of 
plants,—the different substances used in dyeing,—coloured 
glasses and minerals,—-coloured artificial salts,—and different 
coloured gases; and in all these cases I have obtained results 
which lead to the same conclusion. I have analysed, too, the 
blue colour of the sky, to which the Newtonian theory has 
been thought peculiarly applicable; but instead of finding it 
a blue of the first order, in which the extreme red and the ex- 
treme violet rays are deficient, while the rest of the spectrum 
was untouched, I found that it was defective in rays, adjacent 
to some of the fixed lines of Fraunhofer, and that the absorp- 
tive action of our atmosphere widened, as it were, these lines. 
Hence it is obvious, that there are elements in our atmosphere 
which exercise a specific action upon rays of definite refrangi- 
bility, and that this, in some of these rays, is identical with 
that which is exercised over them by the atmosphere of the 
sun. I have obtained analogous results in analysing the yellow, 
orange, red, and purple light, which is reflected from the clouds 
at sunset; but it is impossible to convey any correct idea of 
the composition of these colours, without a reference to the 
fixed lines of the spectrum, of which we at present possess no 
distinct nomenclature. 
I may mention, however, this general fact, that in the va- 
rious specific actions exercised upon light by solids, fluids, 
vapours, and gases, the points at which the spectrum is at- 
tacked are generally coincident with the deficient lines of 
Fraunhofer; and particularly with those which are common 
to the light of the sun, and that of some of the fixed stars. 
Hence it appears, that these rays or lines are weak parts of 
the spectrum, or the parts of white light which have the 
greatest affinity for those elements of matter, which, while they 
enter into the composition of sublunary bodies, exist also in 
the atmospheres of the central luminaries of other systems. 
From the preceding experiments, it is impossible to resist 
the conclusion, that the second and leading proposition of 
Newton’s theory of colours is incompatible with the actual 
phznomena; and we may demonstrate the incorrectness of 
the first proposition by simply stating the fact, that there are 
Third Series. Vol. 8. No. 49. June 1836. 3B 
