THE SENSES 



825 



'When the retina is stimulated by a succession of short flashes of 

 white light, that are not completely fused (as when the image of a 

 flame is looked at in a small revolving mirror, or the flame directly 

 viewed through a slit in a revolving disc), the proportion between 

 the amount of excitation in the three hypothetical groups of fibres is 

 not constant, and the resultant sensation is not that of white light. 

 For any given intensity of light, violet preponderates with a certain 

 duration of each stimulus ; with a shorter duration, green ; with a still 

 shorter duration, red.' These phenomena are especially seen at the 

 edges of the image, which is surrounded by coloured fringes. The 

 explanation is that the sensation does not reach its maximum at the 

 same time for different colours, the excitation in the red fibres in- 

 creasing at first more rapidly than in the green, and in the green more 

 rapidly than in the violet. When the flashes are completely fused, 

 the colour phenomena disappear, and the resultant impression is 

 white, because now the maximum excitation for the given intensity 

 of light and duration of each stimulus is steadily maintained. 



The colour triangle is a graphic method of representing 

 various facts in colour-mixture, the most important of which 

 are mentioned in the description of Fig. 317. 



The chief points to 

 be noted are the fol- 

 lowing: (i) On the curve 

 the spectral colours are 

 arranged at such dis- 

 tances that the angle con- 

 tained between straight 

 lines drawn from the point 

 W and intersecting the 

 curve at the positions cor- 

 responding to any two 

 colours is proportional to 

 their difference in tone. 

 (2) The distance of any 

 point of the curve from 

 W is proportional to the 

 stimulation intensity of the 

 colour corresponding to it. (If the stimulation intensities of all the colours be represented 

 by proportional weights lying at the corresponding points on the curve, W will be the 

 centre of gravity of the system.) (3) The position of a colour produced by the mixture 

 of any pair of spectral colours is found by joining the corresponding points by a 

 straight line. The mixed colour lies on this line at distances from the two points 

 inversely proportional to the stimulation intensity of the two colours, i.e., it lies in the 

 centre of gravity of the weights representing the two colours. (4) It is a particular 

 case of (3) that the complementary colours are situated at the points where straight 

 lines drawn through W intersect the curve, since W is the centre of gravity correspond- 

 ing to a pair of colours only when it lies on the straight line joining them. The non- 

 spectral purple is represented by a broken line. 



It is a point of great theoretical interest that on the Young-Helm- 

 holtz theory the pure spectral colours, although physically saturated 

 (i.e., due to ethereal vibrations of a definite wave-length for each 

 colour), ought not to be physiologically saturated, since they all 

 excite the three fibre groups, although in different degrees. In other 

 words, the red, let us say, of the spectrum ought not to be the purest 

 or fullest red which it is possible to perceive. Now, it is found that 



FIG. 317. COLOUR TRIANGLE. 

 (In the description the point marked ' White ' is 

 referred to as W. ) 



