io 7 2 VISION. 



it depends on movement of the eyes, and therefore becomes much less 

 marked if the disc is observed through a slit (Schenck). 1 Griinbaum 2 

 has recently worked out a definite relation between the breadth of the 

 sector and the diameter of the aperture through which the disc is 

 observed, and there is little doubt that the phenomenon is due to 

 the influence of contrast. The larger the breadth of the aperture in 

 relation to that of the individual sector, the larger will be the fraction 

 of the total revolution of the disc during which black and white will 

 be exposed simultaneously, and will, by reciprocal intensification of each 

 other, increase the rapidity of stimulation necessary to abolish flicker. 



Two other factors found to be of importance are the size and 

 position of the retinal area stimulated. Exner 3 found that the dura- 

 tion of the sensory process varied with the size of the retinal image ; 

 the duration decreasing in arithmetical, when the size increased in 

 geometrical, progression ; and Charpentier 4 similarly found that increase 

 in the size of the retinal image raised the point of fusion. These observers 

 have also found that flicker disappears less readily on the peripheral 

 retina, especially in the dark-adapted eye, while Exner found that 

 fusion took place more readily for the fovea than for an area 1-33 mm. 

 from it. Bellarminoff 5 has further found a higher frequency necessary 

 for fusion on the nasal than on the temporal half of the retina for 

 certain kinds of light. Exner 6 found the point of fusion to be lowered 

 by pressure on the eyeball (p. 1098). 



Increase in the size of the retinal image and removal of the image to 

 the peripheral retina act in the same way as increase of intensity ; and 

 the influence of these factors, and of adaptation, induction, and duration 

 of the stimulus, all seem to be instances of a general law, that the point 

 of fusion of intermittent stimuli, so as to produce a continuous sensation, 

 depends, not on the physical intensities of the stimuli, but on their 

 physiological intensities, as determined by the condition and nature of 

 the stimulated retinal area. 



The previous considerations have applied to colourless sensations. 

 The point of fusion in the case of colour sensations seems to depend wholly 

 upon the brightness of the colour, and to be independent of its colour tone. 

 If coloured light stimulates the retina intermittently, it is found that the 

 brighter the colour, the greater is the rate of stimulation necessary to cause 

 fusion. This has led to the adoption of flicker as the basis of a method of 

 heterochromatic photometry; and the subject will be more fully considered 

 later, when we come to the means of determining the brightness of a colour. 



Recurrent images. It has so far been assumed that the course of 

 the sensory curve is a simple rise and fall. Facts must now be con- 

 sidered which show that the curve is more complex, and that in nearly all 

 conditions of stimulation there is a secondary process, which takes place 

 later than that already considered, and gives rise to what may best be 

 known as a recurrent image, to distinguish it from the after-images 

 described on p. 1057. The recurrent image was first noticed by Purkinje, 7 

 as an elliptical strip following a moving light. It has usually been 



1 Arch.f. d. ges. Physiol., Bonn, 1896, Bd. Ixiv. S. 165. 



2 Journ. Physiol., Cambridge and London, 1897, vol. xxi. p. 396. 



3 Sitzungsb. d. Tc. Akad. d. Wissensch., Wien, 1868, Bd. Iviii. Abth. 2, S. 601. 



4 Arch, d'opht., Paris, 1890, tome x. p. 340. 



5 Arch.f. Ophth., 1889, Bd. xxxv. Abth. 1, S. 25. 



6 Arch.f. d. ges. Physiol., Bonn, 1878, Bd. xvi. S. 407. 



7 " Beobachtungen u. Versuche z. Phys. d. Sinn.," S. 111. 



