264 CAPTAIN W. DE W. ABNEY ON THE COLOUR 



be placed. A small ray of light, Z, was allowed to pass beyond P 2 , and fell on a sum II 

 mirror, G IV , which reflected it on to the back of D, casting a shadow of a needle, N, 

 fixed to B, the camera, on S, a scale at the back of D. 



LO is a lens of short focus which could be moved into a fixed position behind L 4 to 

 throw an enlarged image of the slit on a scale placed below dc. 



In order to form colour mixtures on ad three slits had to be placed along the 

 surface D. The slits were arranged in an open brass frame which slid along the 

 plane D in grooves cut in B. At the bottom and top of the frame or slit holder two 

 pairs of grooves were cut. In the front pair the slits could slide and be clamped in 

 any desired position as determined by a scale engraved along the lower groove, 

 whilst the back pair of grooves was used to hold blackened cards which filled up the 

 intervals between the slits. The position of the slit holder was determined by the 

 shadow cast by the needle N on the scale engraved on its back. 



(VII.) Ascertaining the Position of the Slits in the Spectrum. 



By placing one slit at some fixed number on the front scale and then causing the 

 slit holder to move along the spectrum till known lines (due to metals vaporised in 

 the arc) passed through the centre of the aperture, and there noting the scale number 

 at the back, the position of the slits however placed was known. The position of the 

 principal Fraunhofer lines in regard to the front scale was thus determined when 

 the slit holder was placed in a fixed position as indicated by the needle shadow. 



(VIII.) Method of Determining the Colour Sensations in an Orange Ray of the 



Spectrum. 



It has already been stated that from a preliminary survey it was found that in 

 the orange no blue sensation was excited, and that only red and green had to be 

 determined. Further, it was stated that A, fig. 1, was a colour where the green 

 sensation existed mixed only with white. If these two slits were placed in the 

 spectrum, one in the pure red which only excited the red sensation, and the other 

 at A, it should be possible to make such a mixture of the two colours that they 

 should match the orange colour to which white in known quantity was added. A 

 cell containing bichromate of potash in solution was placed in the beam X, and one 

 slit in D (fig. 2) was caused to traverse the spectrum till the colours appeared to 

 match. It was found, however, that the bichromate was a little paler than the 

 orange of the spectrum, and the beam Y was diverted so that it fell only on ad. 

 The white was diminished till a match was secured and the luminosities of the two 

 were measured, when it was found that the bichromate colour contained 4 '995 per 

 cent, of white as compared with the orange that matched it. 



The bichromate solution could now be used to give a colour to be matched. The 



