262 CAPTAIN W. DE W. ABNEY ON THE COLOUR 



however, we can express the colours in all parts of the spectrum in percentages of 

 luminosity of the three sensations, we can readily convert the equation derived for 

 one quality into that for any other. 



Bearing in mind the effect of the retinal area, it will be seen that it is necessary 

 always to compare the patches of mixed colours when of the same size and viewed 

 from the same distance, and with the centre of the retina. The light from the crater 

 of the positive pole of the arc light, being always of the same quality and of great 

 " whiteness," and giving a spectrum which is rich in blue rays, may be conveniently 

 adopted as a standard. Moreover, this white seems to be of the same quality as the 

 white light seen outside the colour fields, thus approaching to the fundamental white 

 sensation. 



(V.) Method of Finding the Value of the Red and Green Sensations. 



It need only be stated that no blue sensation was found from D (sodium line) to the 

 extreme limit of the spectrum in the red, and that in the yellow the amount found is 

 very small compared with that of the red and green sensations. So below D to the 

 red lithium line we have a mixture, in varying proportions, of pure red and green 

 sensations, and from D to the yellow-green the same two sensations, but not absolutely 

 free from the third sensation. 



A colour in the spectrum which matches a solution of bichromate of potash will 

 thus only excite the red and green sensations, and if we can find out in what propor- 

 tions the two exist as luminosities in a given luminosity of the colour, we can readily 

 determine the sensation luminosity composition of any other colours by means of 

 ordinary colour equations expressed as comparative luminosities. To ascertain the 

 composition of such an orange colour was the object of the first part of the 

 investigation. 



Turning to II, in fig. 1, we see that we have only to add to the colour it represents 

 such a quantity of properly chosen violet to form white, the red and green sensations 

 being present in the proper proportions. If, therefore, we ascertain the comple- 

 mentary colour to the violet, we shall find the colour which is equivalent to II, and 

 this we find to be in the yellow. Having ascertained the luminosity of the red and 

 green sensations in the orange, and from them the relative luminosities of the same 

 two sensations in the complementary colour to the violet, we at once get in terms of 

 the red, the green, and the (provisional) violet sensations the equation to the white 

 light of the quality we may be using. 



(VI.) Apparatus Employed. 



The colour patch apparatus employed in ' Colour Photometry ' was again used 

 (fig. 2). The rays R, R, coming from the crater of the positive pole of the electric 

 light, were collected by a lens, L,, and an image of the crater thrown on the slit Sj. 





