108 COLOUR VISION. [BOOK in. 



object, (and it is through this that we chiefly become aware of the 

 form of an object,) is more powerful than the particular rays which 

 give the object colour ; since even a saturated colour behaves in 

 the same way. If with a feeble illumination we allow a very 

 small part of the spectrum to fall on the retina, we are much 

 more distinctly conscious of a sensation of light than of any 

 particular colour sensation ; indeed the minimum sensation thus 

 felt has been called a ' grey ' for all parts of the spectrum. 

 Moreover the colour which is first recognized upon gradually 

 increasing the illumination, appears less saturated, that is to 

 say apparently more mixed with white than when a large amount t 

 of light of the same refrangibility falls on the retina ; and such 

 distinct colour sensation as may be felt at the first moment of 

 looking at such a light soon diminishes, giving way to a mere 

 sensation of light. These results are perhaps on the whole more 

 intelligible on Bering's theory than on the other. 



When we attempt to compare one colour sensation with another 

 in reference to their behaviour towards variations in the intensity 

 of the stimulus we find the results to a certain extent conflicting. 

 When we diminish the intensity of the stimulus by diminishing 

 general illumination, when we look for instance at objects in 

 nature under light of varying intensity, we find that the colours 

 change unequally as the light diminishes ; as is well known the 

 colours of flowers look very different when night is falling from 

 what they do under bright daylight. In particular we find that 

 as the light diminishes red sensations and also yellow sensations 

 disappear earlier than blue sensations. Hence in dim lights, as 

 those of evening and moonlight, blues preponderate, reds and 

 yellows being less obvious, whereas in bright lights yellows and 

 reds become prominent. 



On the other hand, if we test our sensitiveness to different 

 colours in a different way we get results which are opposed to 

 the above. If for instance we determine the distance at which 

 we cease to recognize the colour of a piece of coloured paper, say 

 1 cm. square, we find that the blue goes first, then green and 

 next yellow, red being recognizable at the longest distance, 

 though the difference between red and yellow is not very great. 

 It will be understood of course that in this experiment we are 

 dealing not only with diminished energy, with diminished ampli- 

 tude of the luminous waves, but also with a diminished area of 

 retinal stimulation. 



Or again, if we take the heating effects of rays of different 

 wave-lengths as a measure of their energy, we may determine 

 the amount of energy needed, in the case of the several colours, 

 to produce a given visual effect. When this is done it is found 

 that the rays in the green, about wave-length X 530 are the 

 most effective; from this part of the spectrum the efficiency 

 declines both towards the violet and the red. 



