Transparency of liquids and colour of the sea

Abstract

In an earlier note in Nature (Nov. 24, 1921, vol. 108, p.402) I pointed out that the scattering of light in its passage through a liquid resulting from the local fluctuations of density, the magnitude of which is given by the Einstein-Smoluchowski relation, should enable its transparency to be determined for the parts of the spectrum in which it does not exercise selective absorption. It should be mentioned that in making an experimental test of this point, account has also to be taken of the scattering resulting from the anisotropy of the molecules and that there is an important difference between this and the scattering due to density fluctuations. The orientation-scattering is almost completely unpolarised and is therefore distributed symmetrically in all directions. The density-scattering is polarised and is twice as intense longitudinally as in a transverse direction.