A THEORY OF COLOUR VISION 



38i 



explanation of the selective action of the eye. Let us suppose 

 that the energy of the compound pendulum represents the 

 energy of the light wave, that in the retina there are a great 

 number of oscillators or vibrators to perform the function of 

 the simple pendulum, and that the energy of the simple pen- 

 dulum represents the part of the energy of the light wave 

 taken up by the eye. 



Fig. 3 is taken from a paper by H. E. Ives, 1 and represents 

 the sensitiveness of the eye to light of different wave-lengths. 

 We learn from it, for example, that if the same quantities of 

 energy are in succession changed into light of the wave-lengths 

 6400 A.U. and 5600 A.U., the luminosities of the sensation 

 produced will be in the ratio of the numbers \"j and 9*8. The 

 curve gives the mean of the results for eighteen observers, and 

 the field of observation had the brightness of a magnesium 



HP 



Fig. 3. 



oxide surface undergoing an illumination of 25 metre-candles 

 and viewed by an eye in which the pupil is at its normal aper- 

 ture. If we compare Ives' curve with the two curves shown 

 in fig. 2, we find it is broader than they are. They can be 

 made broad enough by increasing the frictional resistance to 

 which the simple pendulum is subject, but the shape does 

 not then agree very well ; the maximum of Ives' curve is 

 smoother and not so pointed as- they are. If, however, we 

 assume that the vibrators in the eye have not all the same 

 period, but that their periods vary about a mean value, which 

 is itself in the yellow, we can represent Ives' curve perfectly. 

 The question arises as to the nature of the vibrators. It 

 is possible, though not essential to the explanation, that they 

 are electrons, and that, when they vibrate too far and break 



1 Phil. Mag. (6), 24, p. 860, 1912. The curve has recently been verified and 

 extended by P. G. Nutting, Phil. Mag. (6), 29, p. 301, 1915. 



