OTHER THEORIES 281 



vigorous effect upon the red substance in both stages. Reasons are 

 given for deciding in favour of the third possibihty. That the red 

 substance is set free in excess in the first stage is shown bv the pheno- 

 menon of simultaneous induction. That red light exerts a greater 

 action on the red substance in the second stage is shown by the pre- 

 dominance of red sensation over green and blue in spite of facts which 

 show that the free green and blue substances become at least equal in 

 quantity to the free red substance. These facts are : (1) the induced 

 light, at first red, on prolonged fixation loses in saturation as it gains 

 in brightness until it becomes white or slightly tinged with green ; 

 (2) the phenomenon of reversal of red light to green-blue or green during 

 continued fixation, either on diminishing its brightness or on adding 

 white light to it, or on simple prolonged fixation ; (3) the predominance, 

 sometimes complete, of the green and blue systems in the after-image 

 of red light of medium brightness. 



This gradual increase in the green and blue substance cannot be 

 attributed to fatigue of the process of setting free the red substance, 

 but if the red light is more selective on the retinal processes in the second 

 stage, that of excitation of the nerve-fibres, it follows as a corollarv. 

 For, if, for example, the red, green, and blue substances are set free in 

 the proportion 3:2:2 and are used up in the proportion 4:2:2, the 

 green and blue substances must accumulate more rapidly. This purely 

 hypothetical assumption explains complementary after-images and the 

 various cases of the reversal of colour. 



Though there is no fatigue in the retinal processes of the first stage, 

 it occurs in the second stage. It accounts for the rapid fall in intensity 

 of the sensation during the first few seconds when a patch of bright 

 white light is fixated, and also for the fact that the after-image of 

 white fight is more vivid the shorter the period of fixation, but declines 

 in brightness more rapidly. The more intense the fight the more the 

 first retinal process exceeds the second in activity. The preponderance 

 of the retinal excitement of the red system over the excitement of the 

 blue and green systems is probably slight, even when a sensation of 

 red of good saturation is experienced. For owing to the mutual anta- 

 gonism between corresponding areas of the cortex any predominance of 

 the excitement of one system in the retina is greatly exaggerated in the 

 cortex. 



McDougall has shown that these principles can be made to explain 

 all the phenomena of after-images without undue straining. He 

 emphasises the importance of the part played by the cortex by the 



