CHAP. HI.] SIGHT. 889 



fore giving a distance of more than a millimeter in the retinal 

 image, are still seen as one point. 



557. In accordance with the above, we may suppose the 

 retina to be divided into areas, stimulation of the retina within 

 which gives rise to a single sensation ; we might speak of these 

 as visual areas, and of the stimulation of a visual area as a sensa- 

 tional unit. The areas are very small, and the sensational units 

 very numerous in the fovea centralis and yellow spot ; the areas 

 are larger, and the sensational units fewer, over the rest of the 

 retina, increasingly so towards the periphery. The smaller or 

 larger the areas, the more numerous or fewer the sensational 

 units in any retina or in any part of a retina, the more or less 

 distinct will be the vision. 



Now in the human eye 50 cones may be counted along a line 

 of 200 /j, in length drawn through the centre of the yellow spot ; 

 this would give 4 /* for the distance between the centres of two 

 adjoining cones in the yellow spot, the average diameter of a 

 cone at its widest part being here about 3 /* and there being 

 slight intervals between neighbouring cones. Hence if we take 

 the centre of a cone as the centre of an anatomical retinal area, 

 these anatomical areas correspond very fairly in the region of 

 distinct vision to the physiological visual areas just spoken of. 

 If two points of the retinal image are less than 4 /* apart, they 

 may both lie within the area of a single cone ; and it is just 

 when they are less than about 4 p, apart that they cease to give 

 rise to two distinct sensations. It must be remembered, how- 

 ever, that the fusion or distinction of the sensations is ultimately 

 determined by the brain. The retinal area must be carefully 

 distinguished from the sensational unit, for the sensation is a 

 process whose arena stretches from the retina to certain parts 

 of the brain, and the circumscription of the sensational unit, 

 though it must begin as a retinal area, must also be continued 

 as a cerebral area, the latter corresponding to, and being as it 

 were the projection of, the former. Two points of the retinal 

 image less than 4 /x apart might lie both within the area of a 

 single cone ; but the reason why, under such circumstances, they 

 give rise to one sensation only is not because one cone-fibre only 

 is stimulated. For, two points of a retinal image might lie, one 

 on the area of one cone and another on the area of an adjoining 

 cone, and still be less than 4 //, apart ; in such a case two cone- 

 fibres would be stimulated; and yet only one sensation would 

 be produced. 



In the case where the two points lie entirely within the area 

 of a single cone, it is exceedingly probable that, even if tlu- 

 adjacent cones or cone-fibres in tin- retina are not at tlu- same 

 time stimulated, impulses radiate from the cerebral ending of 

 the excited cone into the neighbouring cerebral endings of the 

 neighbouring cones ; in other words, the sensation-area in the 



