ABSORPTION OF LIGHT, ETC. 43 



spectrum are not equally bright. But that is a point which 

 we need not for the moment take into consideration. A 

 horizontal line, then, parallel to the axis of abscissae, may be 

 supposed to represent for each particular colour, the place 

 of 'which is defined by the abscissa, the original intensity of 

 that colour. Now after passing through a certain stratum of 

 the medium of a certain thickness, that intensity will be re- 

 duced differently for the different colours, and consequently 

 the locus which defines to the eye the composition of the 

 light which is passed through that stratum will be a certain 

 curve, but it will depend on the nature of the medium 

 what the nature of the curve will be. Now I have drawn 

 here [referring to figure] what represents a curve for a green 

 colour in a certain medium. To find how the light will be 

 composed after passing through a second stratum of equal 

 thickness, we have nothing to do but for a sufficient number 

 of abscissae to take an ordinate which bears to the ordinate 

 of this curve the same ratio that the latter bears to the 

 original ordinate or unity, and so on for additional strata 

 of the same thickness. Thus we get a succession of. curves 

 representing to the eye the composition of the light which 

 has passed through successive thicknesses of the medium. 

 You may notice that the quantity of light altogether goes 

 on decreasing ; but that is not all, the proportion of the 

 different parts goes on changing as well. In this case, if 

 the opacity of the medium is such as is represented in this 

 curve, the blue or bluish-green light which predominated a 

 little at first will predominate more and more, and the 

 colour of the medium will become a purer and purer green 

 as you look through greater and greater thicknesses of it. 

 Here is another curious curve representing in the same 

 manner the type of light which is transmitted through one 

 of the ordinary blue glasses coloured by oxide of cobalt. I 

 do not pretend that it is an exact representation, but it is 

 an approximate one, and you will see the curious alternations 

 which there are in this case, of comparative opacity and 

 transparency. In this way we can readily understand how 

 it is that the colour of a coloured fluid is continually 

 changing according to the thickness looked through. This 

 phenomenon in its more striking examples is sometimes 

 called dichroism, but as that word has been employed to 

 designate so many phenomena totally different from one 



