542 Sir DAVID BREWSTER on the Colours of Natural Bodies. 



of an olive-green, which grows more and more of a brownish- 

 yellow, till it becomes almost colourless. During these various 

 changes, the specific action of the fluid upon the spectrum 

 changes also ; but neither the change of colour nor the change 

 of action have any relation whatever to the effects of an increase 

 or decrease of thickness in the tinging corpuscles, by which Sir 

 ISAAC NEWTON explains the changes which take place in the 

 colour of leaves. When the fluid has become almost colourless 

 like water, it still exercises a powerful action upon the middle 

 of the red space, and a faint, but still perceptible action, at two 

 points of the green band. This curious fact may lead us to 

 expect that transparent media may yet be discovered, which 

 shall absorb different parts of the spectrum, while they them- 

 selves are perfectly colourless. This effect of course cannot take 

 place unless the rays absorbed compose white light. 



In the course of these experiments, I observed a very re- 

 markable phenomenon, which at first sight appeared to be some- 

 what favourable to the Newtonian theory. In making a strong 

 beam of the sun's light pass through the green fluid, I was sur- 

 prised to observe that its colour was a brilliant red, complemen- 

 tary to the green. By making the ray pass through greater 

 thicknesses in succession, it became first orange and then yellow 

 and yellowish-green, and it would undoubtedly have become blue, 

 if it had been transmitted through a greater thickness of fluid. 

 This mode of producing a spectrum by reflexion from the par- 

 ticles of a fluid, exhibits the phenomenon of opalescence in a very 

 interesting form. Had the green fluid shewn the same colour 

 at all thicknesses, or had it absorbed only the red rays, the 

 opalescent beam would have been red throughout the whole of 

 its path : but as the different colours are absorbed in different 

 proportions, and, in the present case, in the order of their refran- 

 gibility, excepting the blue and violet, the colour of the intro- 

 mitted beam must vary from red to greenish-yellow, as these 

 colours are successively taken out of it. 



