794 AN AMERICAN 1EXT-BOOK OF PHYSIOLOGY. 



will appear enlarged and at a distance corresponding to that of the surface 

 looked at. Hence one and the same retinal image may, under different cir- 

 cumstances, give rise to the impression of objects at different distances. We 

 must therefore regard the perception of distance not as a direct datum of vision, 

 but, as will be later explained, a matter of visual judgment. 



When objects are of such a shape that their images may be thrown suc- 

 cessively upon the same part of the retina, it is possible to judge of their rela- 

 tive size with considerable accuracy, the retinal surface serving as a scale to 

 which the images are successively applied. When this is not the case, the 

 error of judgment is much greater. We can compare, for instance, the relative 

 length of two vertical or of two horizontal lines with a good deal of precision, 

 but in comparing a vertical with a horizontal line we are liable to make a con- 

 siderable error. Thus it is difficult to realize that the vertical and the hori- 

 zontal lines in Figure 246 are of the same length. The error consists in an 



over-estimation of the length of the vertical 

 lines relatively to horizontal ones, and appears to 

 depend, in part at any rate, upon the small size 

 of the superior rectus muscle relatively to the 

 other muscles of the eye. The difference amounts 

 to 30-45 per cent, in weight and 40-53 per cent, 

 in area of cross section. It is evident, therefore, 

 that a given motion of the eye in the upward 

 direction will require a more powerful contraction 

 ., of the weaker muscle concerned in the movement 



FIG. 246.-To illustrate the over-esti- than will be demanded of the stronger muscles 



mation of vertical lines. . in i 



moving the eye laterally to an equal amount. 



Hence we judge the upward motion of the eye to be greater because to accom- 

 plish it we make a greater effort than is required 

 for a horizontal movement of equal extent. 



The position of the vertical line bisecting the 

 horizontal one (in Fig. 246) aids the illusion, as 

 may be seen by turning the page through 90, so 

 as to bring the bisected line into a vertical posi- 

 tion, or by looking at the lines in Figure 247, in 

 which the illusion is much less marked than in 

 Figure 246. 



The tendency to over-estimate the length of 

 vertical lines is also illustrated by the error 

 commonly made in supposing the height of the 



crown of an ordinary silk hat to be greater 



,i ., i T.I FIG. 247. To illustrate the over-estima- 



than its breadth. tion of vertical lines . 



Irradiation. Many other circumstances 



affect the accuracy of the spatial perception of the retina. One of the most 

 important of these is the intensity of the illumination. All brilliantly illumi- 

 nated objects appear larger than feebly illuminated ones of the same size, as is 



