256 Me. Hunt on the Cinephmdic Colour Top. 



mistake as to the way in which the shading arose, it was apprehended that 

 if two apertvxres were used the shading would become less, or disappear 

 altogether. Five, four, and three apertm'es were consequently first 

 successively tried, and gave four, three, and two sets of colours respec- 

 tively — the openings and blackened intervals being equal in all cases. 

 Contrary to expectation, the shading and general brilliancy were 

 improved at each experiment. At first three sectors of blue, yellow, 

 and crimson were fixed upon the top, but it was afterwai'ds found that 

 green, scarlet, and violet gave much better results. Thus, with these 

 colours on the top, and a loose disc having two apertures, a most beautiful 

 spectrum was obtained, constituting an imitation rainbow in a circular 

 form. The green was delicately graduated through yellow and orange 

 into scarlet, and scarlet through crimson and purple into violet, and the 

 violet through indigo and blue into green. Each radius was of a 

 uniform tint from centre to cu-cumference, but the gradation from radius 

 to radius cireulai-ly was insensible. By tapering to points the black 

 sectors of the loose disc the colours were gradually shaded off radially 

 into the neutral tint seen without the loose disc. In this way, if certain 

 natural colom-s were fixed upon as standards, almost all the possible 

 combinations of colours could be exhibited by this mechanical method 

 of mixing or shading them by insensible gradation, and names referring 

 to their relative positions on the spectrum obtained could be satisfac- 

 torily given to each. 



The spectrum showing one set of colours was afterwards obtained 

 with two, three, or more sets of colours on the top, by using loose discs 

 having as many apertures, plus one, as there were sets of colours ; and it 

 was found that the brilliancy increased with the number of sets of colours, 

 as far as was tried. There would, of course, be a limit to this increase of 

 brilliancy. In these experiments the entii-e disc appeared occupied with 

 colour, the black portions being quite lost to the eye, and in addition the 

 colours appeared more vivid than colours lying on the table, and dupli- 

 cates of those actually fixed on the top. This is the more remarkable 

 when it is remembered that in the experiments there was at no tim e more 

 than /i«(/'the coloured surface exposed. It is probably attributable to 

 the alternate flashing of the coloured and black sectors on the eye. The 

 pupil is well known to accommodate itself to the amount of light reach- 

 ing the eye, so as to reduce the variation of the amount actually entering 

 it. It has suggested itself to the writer that many complex pheno- 

 mena of vision would be accounted for by a second accommodatuig 

 power residing in the retina (or other surface at which the visual image 

 acts on the nerves), each point of the surface possessing this power 

 independently. The supposition of fatigue of the nerves does not 



