70 Mr. J. S. Dow on the Physiological 



forward above, seems to be supplied by the valuable work of 

 Dr. T. 0. Porter*. This observer found that, for illumi- 

 nations varying from 0'25 lux up to 12,800 times 

 this value, the connexion between the critical speed of a 

 sector-disk (i. e. the speed at which the flicker caused by its 

 rotation could just be made to vanish), and the illumination 

 was of the form 



n = k . log I + p, 



where n = the number of revolutions per second when flicker 

 vanishes, k andp are constants, and I is the illumination of 

 the disk. 



Now for illuminations below 0*25 lux a similar relation 

 was found to hold, but the constant, Jc, suddenly diminished 

 to half its former value and the curve connecting n 

 and log I took an abrupt turn. This means that, at illumi- 

 nations below this value, the duration of time during which a 

 luminous image was retained appreciably unaltered in in- 

 tensity was much greater, and the critical speed propor- 

 tionately less. 



"Now it will be recalled that the above order of illumi- 

 nation, 0*25 lux, is just about the value at which, as we 

 have seen earlier, the cones seem to cease action abruptly 

 and cone-vision is replaced by rod-vision. It therefore seems 

 highly probable that the abrupt change in the value of k 

 corresponds to the substitution of rod-vision for cone-vision, 

 and that the sensation of flicker tends to disappear at a lower 

 speed in the case of the rods than in the case of the cones. 

 This, indeed, seems in accordance with the observations of 

 von Kries and other physiologists that the visual purple, 

 with which rod-vision is believed to be connected, possesses 

 a certain visual inertia in virtue of which it reacts appre- 

 ciably later to an impulse than do the organs whieh are 

 believed to be predominant at higher illuminations. 



While mentioning this point, it may be of interest to refer 

 to some other physiological results which seem to bear out 

 this suggestion. The manner in which a luminous impression 

 received by the eye disappears when the stimulus is with- 

 drawn is, of course, of considerable interest as regards the 

 theory of the flicker photometer. The matter has been made 

 the subject of much study among physiologists, and the 

 process of dying away of an impression seems to be of a very 

 complicated character. Even bearing in mind the different 

 behaviour of the rods and cones, this might be anticipated, 

 but there are other contributory causes of complication. 



* Proc. Roy. Sec. London, vol. lxx. p. 315 (1902). 



