800 AN AMERICAN TEXT-BOOK OF PHYSIOLOGY. 



sion. When the moon is high above our heads we have no means of esti- 

 mating its distance from us, since there are no intervening objects with which 

 we can compare it. Hence we judge it to be nearer than when, seen on the 

 horizon, it is obviously farther off than all terrestrial objects. Since the size 

 of the retinal image of the moon is the same in the two cases, we reconcile 

 the sensation with its apparent greater distance when seen on the horizon by 

 attributing to the moon in this position a greater actual size. 



If the retinal image have the form of a familiar object of regular shape 

 e. g. a house or a table we interpret its outlines in the light of experience 

 and distinguish without difficulty between the nearer and more remote parts of 

 the object. Even the projection of the outlines of such an object on to a plane 

 surface (i. e. a perspective drawing) suggests the real relations of the different 

 parts of the picture so strongly that we recognize at once the relative distances 

 of the various portions of the object represented. How powerfully a familiar 

 outline can suggest the form and relief usually associated with it is well illus- 

 trated by the experiment of looking into a mask painted on its interior to 

 resemble a human face. In this case the familiar outlines of a human face 

 are brought into unfamiliar association with a receding instead of a projecting 

 form, but the ordinary association of these outlines is strong enough to force 

 the eye to see the hollow mask as a projecting face. 1 The fact that the pro- 

 jecting portions of an object are usually more brightly illuminated than the 

 receding or depressed portions is of great assistance in determining their rela- 

 tive distance. This use of shadows as an aid to the perception of relief pre- 

 supposes a knowledge of the direction from which the light falls on an object, 

 and if we are deceived on the latter we draw erroneous conclusions with 

 regard to the former point. Thus, if we look at an embossed letter or figure 

 through a lens which makes it appear inverted the accompanying reversal of 

 the shadows will cause the letter to appear depressed. The influence of 

 shadows on our judgment of relief is, however, not so strong as that of the 

 outline of a iamiliar object. In a case of conflicting testimony the latter 

 usually prevails, as, for example, in the above-mentioned experiment with the 

 mask. 



Aided by these peculiarities of the retinal picture, the mind interprets it as 

 corresponding in its different parts to points at different distances from the eye, 

 and it is interesting to notice that painters, whose work, being on a plane sur- 

 face, is necessarily in all its parts at the same distance from the eye, use similar 

 devices in order to give depth to their pictures. Distant hills are painted with 

 indistinct outlines to secure what is called " aerial perspective." Figures of 

 men and animals are introduced in appropriate dimensions to suggest the dis- 

 tance between the foreground and the background of the picture. Landscapes 

 are painted preferably by morning and evening light, since at these hours the 

 marked shadows aid materially in the suggestion of distance. 



1 In the experiment the mask should be placed at a distance of about two meters and one 

 eye closed. Even with both eyes open the illusion often persists if the distance is increased to 

 five or six meters. 



