- Email: [email protected]
Stereoscopic and Perspective Vision
S T E R E O S C O P I C AND P E R S P E C T I V E V I S I O N . ISADORE FRANKLIN, M. D. SHEBOYGAN, W I S .
This paper points out the differing significance of these two terms, and indicates the numerous factors that enter into perfect binocular vision. It also suggests the experiment of reversing the position of the two pictures viewed through the stereoscope. This experiment should he tried in order to appreciate the important points brought out in this paper.
Concerning the nature, relationship and modus operandi of these two visual senses, there exists in ophthalmic liter ature, it seems to me, considerable con fusion. Dr. Ball in his "Modern Oph thalmology" expresses the prevalent views on the subject, when he says that "This (stereoscopic vision) is another name for binocular vision or the sense of depth." He asserts a truth, when he states that "An object appears to us solid when each eye views it from a dif ferent point;" but follows it with the conclusion that "It is the unlikeness of the two pictures which gives the idea of depth." And again that "This (stereo scope) shows that the perception of depth is caused by a slight nonidentity of the retinal images." By reaffirming well established truths concerning this subject, and treating it from the point of view of my reflections and observations, I hope to show that stereoscopic vision, and per spective vision or the sense of depth, are essentially different, though they complement each other in a rather com plicated manner; that neither of them is necessarily binocular; that the stereo scope shows, on the contrary, that it isn't the slight nonidentity of the two retinal images that produces the sense of depth; and finally that stereoscopic and perspective vision are relative terms, the elements of both senses ex isting in various degress and combina tions in all forms of image formation. Stereoscopic vision, or sense of solid ity or of "relief," or what is more vul garly, tho very aptly described as "standing out effect," must be sharply differentiated from perspective vision, or sense of depth. Stereoscopic vision may be defined as the sense of seeing an object from two different aspects, which through the association of ster
eoscopic memory gives the impression of solidity. Perspective vision, on the other hand, may be defined as the sense of seeing objects at varying distances, which through the association of mem ory of bodily or limb excursions through space, produces a vivid sense of depth. In case of stereoscopic vision, it suf fices to fix an object with both eyes to get a complete sense of solidity. The muscles of convergence and accommo dation, tho in use, are not called into "play." Whereas to obtain a complete sense of depth, it is necessary to con verge and accommodate for varying distances. The difference is therefore quite clear: One is static and the other principally dynamic in its nature. On analyzing the two senses, even apart from their psychologic aspects, the elements that enter into their make up are surprisingly numerous and com plicated. The following are the factors which separately or in combination produce the effects of stereoscopic and perspective vision. ELEMENTS
OF STEREOSCOPIC VISION S E N S E OF SOLIDITY
OR
1. Sense of seeing an object from two different aspects. 2. Sense of completeness of image thru the mutually complementing ef fect of two different images. 3. Shadow effects. 4. Augmenting effect of perspective vision. ELEMENTS
OF PERSPECTIVE S E N S E OF D E P T H
VISION
OR
1. Sense of seeing objects at varying distances through (a) the play of the muscles of convergence; and (b) the play of the muscle of accommodation. 2. Blurring of part of the field of vision in an antero-posterior direction,
STEREOSCOPIC AND PERSPECTIVE VISION
due to ( a ) doubling in front and be yond point of convergence; (b) being out of focus in front or beyond point of accommodation; (c) diminution of images below visual angle with inincreasing distance; and (d) atmos pheric effects. 3. Relation of size of images to dis tance. 4. The relative position of objects in which we are accustomed to find them. 5. The enhancing effect of stereo scopic vision. It is hardly necessary to state that the sense of seeing an object from two different aspects is the most important element in stereoscopic vision. This factor, however, is in its turn composed of the following separate elements: (a) sense of converging muscles; (b) sense of common source of rays [identical lo cation of image on both retinas with respect to field of vision, same side] ; (c) sense of difference of the two ret inal images. Though the sense of con verging muscles is normally a factor in producing the sense of common source of rays, it is not essential, since it may be replaced by prismatic effects. Furthermore, both factors combined, (a) and ( b ) , are incapable of produc ing a stereoscopic effect without the aid of factor (c), since merely fixing a picture of a solid with both eyes doesn't by itself produce a stereoscopic effect, because of the absence of sense of dif ference of two retinal images. This last element is therefore important; but only in so far as it contributes to the principal factor, that of seeing a solid from two different aspects. Alone it is quite incapable of producing a stereo scopic effect. If we hold the two stereoscopic pic tures in front of the eyes, so that each eye is looking into the center of its pic ture and the visual axes are parallel, no ■stereoscopic effect will be obtained, even though each eye perceives a dif ferent image, the reason being that we are looking at both pictures from the same aspect (same plane). True, the pictures must be made to overlap, but not for the mere purpose of "uniting" them, that is incidental, but to give the
237
eyes the sensation of looking at one thing from two different points. Nevertheless, the fact that each eye perceives an image from its own point of view is a factor of some value in itself, since each image complementing the other produces a sense of complete ness to which we are accustomed when looking at real solids with both eyes. Powerful as the principal factor of stereoscopic vision is, it is not capable of producing a complete effect without the aid of shadows. A cube, the three presenting sides of which are equally illuminated, will lose a great part of its solid appearance. On the other hand, a picture of a cube drawn with one eye open, but properly shaded, does pro duce a strong" stereoscopic effect. Suf fice it furthermore to reflect, that solids do not appear flat, even if one does look at them with one eye; though to be sure, they lose much in "relief." Bin ocular vision, therefore, tho important, is not essential for stereoscopic vision. It is in monocular stereoscopic vision particularly, that the contributing ef fect of perspective vision is of value, since a sense of antero-posterior exten sion of an object cannot fail to produce a sense of solidity. PERSPECTIVE VISION
In examining the elements of per spective vision, it is obvious that the sense of play of the muscles of converg ence is a powerful factor in producing a sense of depth, since the distances of objects correspond to the efforts of convergence; tho it is to be remem bered that beyond about twenty feet, it ceases to be of value, (which inci dentally applies to the stereoscopic vision.) In the stereoscope, as well as in real life, it is the principal factor of pro ducing a complete sense of depth. Again, more so than in a case of stereo scopic vision, I must emphasize that it is not the mere dissimilarity of the pic tures that produces a sense of depth, but the actual play of muscles of con vergence that the stereoscope as ivell as conditions in real life require. It is almost unbelievable that "flat" pictures in the stereoscope, where all images arise from one plane, should re-
2J8
ISAUOHK F K A N K L I N
quire the convergence of the visual axes for various distances. It is, however; explained by the following: Because of each picture being taken from a differ ent point of view, the objects in each picture are placed away from its outer field towards the center of binocular fixation (towards inner edge of pic ture). This relationship is strongest in the foreground of the picture, and is gradually reduced towards the back ground, i. e. objects in the foreground are displaced more towards the binoc ular center than those in the back-J ground. The result is that the eyes actually converge more for objects that are supposed to be near than for ob jects that are supposed to be far, giv ing rise to a powerful sensation of depth. Supporting proof is to be had in the following experiment: An ordinary stereoscopic card is cut so as to sepa rate the two pictures. They are then approximated with their outer edges, so that they are reversed in relation to each othei. Thus placed in the stereo scope, they arc brought as near to the eyes as is comfortable. When thus used, a remarkable effect is obtained— the view is turned "inside out," i. e., objects that should appear in the fore ground appear in the background and vice versa. | It may be necessary to look for some time to get the effect, since the other factors of perspective vision militate against such an unnat ural arrangement of things.] The explanation for this phenome non is easily to be had in the fact that because of the reversed positions of the pictures, the displacement of near ob jects is toward the outer field instead of the inner field, with the result that the. eyes converge less for objects in the foreground than for those in the back ground; for which reason, what should appear near appears far, and vice versa. It is impossible, on the other hand, to see how the mere interchange of dis similar pictures, per se, could produce such phenomenon. It appears to me that the theory that "The perception of depth is caused by a slight nonidentity of the two retinal images" is definitely
disproven from the above considera tion. The play of the muscles of accommo dation is next in importance. As in the case of convergence, the distance of ob jects corresponds to the accommodative effort, though again beyond twenty feet it is of no value in judging distance. As already mentioned, it is a great aid to monocular stereoscopic vision. Let us bear in mind that even though we look at a series of solids with only one eye, we still have a strong sense of depth. Thanks to the play of accommodation and several other minor factors, we are able to get a fair sense of depth without binocular vision (convergence play). The next subjective element in our sense of depth is blurring of a part of the field of vision (in an antero-posterior direction). Experience has taught us that when objects are blurred within twenty feet, it is either because they are in front of or beyond our point of convergence (doubling); or because they arc in front or beyond our point of accommodation (out of focus). Or dinarily we are but vaguely aware of the condition, but nevertheless uncon sciously, we are in part guided by it in projecting objects into their proper depth. The blurring of images because of increasing distance and because of atmospheric effects (decrease of air transparency, etc.), is of great value in perspective vision beyond the twenty feet limit. * It is common knowledge that we judge distance of objects by comparing their apparent size with what we con sider their real size. And this is an other useful factor in perspective vision beyond twenty feet. That to a certain extent we judge the relative position of objects as we are accustomed to find them, is perhaps not so generally sus pected, but is nevertheless true. For example, a half mile away, we cannot by our natural optical means really tell whether a tree is in front of a building or in its wall. Nevertheless, knowing from experience that trees grow in front of buildings and not in their walls, we are strongly under the im pression that we really see it in front of the building.
S T E R E O S C O P I C AND P E R S P E C T I V E VISION
In this connection, it m a y be m e n tioned t h a t the a p p a r e n t direction of m o v e m e n t of objects and their m o t i o n in general, is of s o m e aid in b o t h per spective and stereoscopic vision, as is particularly the case w i t h the " m o v i n g picture." A s to stereoscopic vision as a factor in perspective vision, it is b u t necessary here to repeat t h a t it is in itself an i m p o r t a n t g u i d e in e s t i m a t i n g the depth of objects, since the solid appearance of objects decreases with distance. I a m n o w a w a r e that it will be ob jected to the idea of r e g a r d i n g m i n o r e l e m e n t s , here e n u m e r a t e d , as g e n u i n e factors of stereoscopic a n d perspective vision, on t h e g r o u n d t h a t they a r e largely mental in their n a t u r e , and not real optico-physiologic p h e n o m e n a (such as sense of c o n v e r g e n c e , accom m o d a t i o n , e t c . ) . T o this I m u s t point out that n e i t h e r the sense of solidity nor the sense of depth are purely ocu lar. Not all the factors of stereoscopic and perspective vision combined could produce either the sense of d e p t h . o r solidity w i t h o u t the association of oilier senses, particularly the tactile and m u s cle senses (of the body in general ). Suf fice it to mention here, the well known observation that those sceing-for the first time must develop a sense of depth and solidity by learning to associate their ocular impression's with their tactile and muscular experiences. In other words. the senses of depth and solidity are neither essentially ocular, nor tactile, nor muscular in their n a t u r e ; but are a psychic complex, outgrowing principally from the above three senses, but em bracing in addition other mental reac tions to bodily experience. It may now be of interest, for the pur pose of e m p h a s i z i n g the relativity <>f the senses of depth and solidity, to briefly examine the combinations of elements
239
which enter into the various forms of image formation. B e g i n n i n g with the p h o t o g r a p h , the "flattest" i m a g e of all, we find t h a t w e are by no m e a n s justified in c o n s i d e r i n g it devoid of all stereoscopic and per spective effects. A n d it is indeed sur p r i s i n g h o w m a n y of the e l e m e n t s are present. T o begin with, in p h o t o g r a p h s in which t h e light is well han dled, the s h a d o w factor does produce a s t r o n g stereoscopic effect. A s to per spective effects, all factors, except those of a c c o m m o d a t i o n and c o n v e r g e n c e play, are p r e s e n t . T h e sense of depth due to the b l u r r i n g of objects which are out of focus is especially m a r k e d in p h o t o g r a p h s , for which reason a good p h o t o - p o r t r a i t rivals a p o r t r a i t paint ing in stereoscopic effects. In p a i n t i n g s this factor is a b s e n t , since the artist a c c o m m o d a t e s , and therefore paints all objects within t w e n t y feet with equal clearness. T h i s is quite a defect, and is one reason w h y p a i n t i n g s lose in perspective the closer they are looked at. It is h o w e v e r partly c o m p e n s a t e d by objects b e i n g r e p r e s e n t e d "complete"-; i. e., as seen with both eyes. In addition, the artis tic h a n d l i n g of a t m o s p h e r i c effects in landscapes ( a n d o t h e r stimuli to i m a g inafon lacking in the p h o t o g r a p h ) , m a k e s the effect of depth in p a i n t i n g s far superior to p h o t o g r a p h s . T h e stereoscope, already discussed, differs from the image formation in ac tual life by lack of a c c o m m o d a t i o n play, and indeed lack of a c c o m m o d a t i o n al together, since it is replaced by the use of convex lenses. As to actual life, it is b u t necessary to repeal that all factors of stereoscopic and perspective vision are p r e s e n t u p to the t w e n t y feet limit. B e y o n d t h a t everything depending upon con vergence and a c c o m m o d a t i o n is, of course, absent.
This paper points out the differing significance of these two terms, and indicates the numerous factors that enter into perfect binocular vision. It also suggests the experiment of reversing the position of the two pictures viewed through the stereoscope. This experiment should he tried in order to appreciate the important points brought out in this paper.
Concerning the nature, relationship and modus operandi of these two visual senses, there exists in ophthalmic liter ature, it seems to me, considerable con fusion. Dr. Ball in his "Modern Oph thalmology" expresses the prevalent views on the subject, when he says that "This (stereoscopic vision) is another name for binocular vision or the sense of depth." He asserts a truth, when he states that "An object appears to us solid when each eye views it from a dif ferent point;" but follows it with the conclusion that "It is the unlikeness of the two pictures which gives the idea of depth." And again that "This (stereo scope) shows that the perception of depth is caused by a slight nonidentity of the retinal images." By reaffirming well established truths concerning this subject, and treating it from the point of view of my reflections and observations, I hope to show that stereoscopic vision, and per spective vision or the sense of depth, are essentially different, though they complement each other in a rather com plicated manner; that neither of them is necessarily binocular; that the stereo scope shows, on the contrary, that it isn't the slight nonidentity of the two retinal images that produces the sense of depth; and finally that stereoscopic and perspective vision are relative terms, the elements of both senses ex isting in various degress and combina tions in all forms of image formation. Stereoscopic vision, or sense of solid ity or of "relief," or what is more vul garly, tho very aptly described as "standing out effect," must be sharply differentiated from perspective vision, or sense of depth. Stereoscopic vision may be defined as the sense of seeing an object from two different aspects, which through the association of ster
eoscopic memory gives the impression of solidity. Perspective vision, on the other hand, may be defined as the sense of seeing objects at varying distances, which through the association of mem ory of bodily or limb excursions through space, produces a vivid sense of depth. In case of stereoscopic vision, it suf fices to fix an object with both eyes to get a complete sense of solidity. The muscles of convergence and accommo dation, tho in use, are not called into "play." Whereas to obtain a complete sense of depth, it is necessary to con verge and accommodate for varying distances. The difference is therefore quite clear: One is static and the other principally dynamic in its nature. On analyzing the two senses, even apart from their psychologic aspects, the elements that enter into their make up are surprisingly numerous and com plicated. The following are the factors which separately or in combination produce the effects of stereoscopic and perspective vision. ELEMENTS
OF STEREOSCOPIC VISION S E N S E OF SOLIDITY
OR
1. Sense of seeing an object from two different aspects. 2. Sense of completeness of image thru the mutually complementing ef fect of two different images. 3. Shadow effects. 4. Augmenting effect of perspective vision. ELEMENTS
OF PERSPECTIVE S E N S E OF D E P T H
VISION
OR
1. Sense of seeing objects at varying distances through (a) the play of the muscles of convergence; and (b) the play of the muscle of accommodation. 2. Blurring of part of the field of vision in an antero-posterior direction,
STEREOSCOPIC AND PERSPECTIVE VISION
due to ( a ) doubling in front and be yond point of convergence; (b) being out of focus in front or beyond point of accommodation; (c) diminution of images below visual angle with inincreasing distance; and (d) atmos pheric effects. 3. Relation of size of images to dis tance. 4. The relative position of objects in which we are accustomed to find them. 5. The enhancing effect of stereo scopic vision. It is hardly necessary to state that the sense of seeing an object from two different aspects is the most important element in stereoscopic vision. This factor, however, is in its turn composed of the following separate elements: (a) sense of converging muscles; (b) sense of common source of rays [identical lo cation of image on both retinas with respect to field of vision, same side] ; (c) sense of difference of the two ret inal images. Though the sense of con verging muscles is normally a factor in producing the sense of common source of rays, it is not essential, since it may be replaced by prismatic effects. Furthermore, both factors combined, (a) and ( b ) , are incapable of produc ing a stereoscopic effect without the aid of factor (c), since merely fixing a picture of a solid with both eyes doesn't by itself produce a stereoscopic effect, because of the absence of sense of dif ference of two retinal images. This last element is therefore important; but only in so far as it contributes to the principal factor, that of seeing a solid from two different aspects. Alone it is quite incapable of producing a stereo scopic effect. If we hold the two stereoscopic pic tures in front of the eyes, so that each eye is looking into the center of its pic ture and the visual axes are parallel, no ■stereoscopic effect will be obtained, even though each eye perceives a dif ferent image, the reason being that we are looking at both pictures from the same aspect (same plane). True, the pictures must be made to overlap, but not for the mere purpose of "uniting" them, that is incidental, but to give the
237
eyes the sensation of looking at one thing from two different points. Nevertheless, the fact that each eye perceives an image from its own point of view is a factor of some value in itself, since each image complementing the other produces a sense of complete ness to which we are accustomed when looking at real solids with both eyes. Powerful as the principal factor of stereoscopic vision is, it is not capable of producing a complete effect without the aid of shadows. A cube, the three presenting sides of which are equally illuminated, will lose a great part of its solid appearance. On the other hand, a picture of a cube drawn with one eye open, but properly shaded, does pro duce a strong" stereoscopic effect. Suf fice it furthermore to reflect, that solids do not appear flat, even if one does look at them with one eye; though to be sure, they lose much in "relief." Bin ocular vision, therefore, tho important, is not essential for stereoscopic vision. It is in monocular stereoscopic vision particularly, that the contributing ef fect of perspective vision is of value, since a sense of antero-posterior exten sion of an object cannot fail to produce a sense of solidity. PERSPECTIVE VISION
In examining the elements of per spective vision, it is obvious that the sense of play of the muscles of converg ence is a powerful factor in producing a sense of depth, since the distances of objects correspond to the efforts of convergence; tho it is to be remem bered that beyond about twenty feet, it ceases to be of value, (which inci dentally applies to the stereoscopic vision.) In the stereoscope, as well as in real life, it is the principal factor of pro ducing a complete sense of depth. Again, more so than in a case of stereo scopic vision, I must emphasize that it is not the mere dissimilarity of the pic tures that produces a sense of depth, but the actual play of muscles of con vergence that the stereoscope as ivell as conditions in real life require. It is almost unbelievable that "flat" pictures in the stereoscope, where all images arise from one plane, should re-
2J8
ISAUOHK F K A N K L I N
quire the convergence of the visual axes for various distances. It is, however; explained by the following: Because of each picture being taken from a differ ent point of view, the objects in each picture are placed away from its outer field towards the center of binocular fixation (towards inner edge of pic ture). This relationship is strongest in the foreground of the picture, and is gradually reduced towards the back ground, i. e. objects in the foreground are displaced more towards the binoc ular center than those in the back-J ground. The result is that the eyes actually converge more for objects that are supposed to be near than for ob jects that are supposed to be far, giv ing rise to a powerful sensation of depth. Supporting proof is to be had in the following experiment: An ordinary stereoscopic card is cut so as to sepa rate the two pictures. They are then approximated with their outer edges, so that they are reversed in relation to each othei. Thus placed in the stereo scope, they arc brought as near to the eyes as is comfortable. When thus used, a remarkable effect is obtained— the view is turned "inside out," i. e., objects that should appear in the fore ground appear in the background and vice versa. | It may be necessary to look for some time to get the effect, since the other factors of perspective vision militate against such an unnat ural arrangement of things.] The explanation for this phenome non is easily to be had in the fact that because of the reversed positions of the pictures, the displacement of near ob jects is toward the outer field instead of the inner field, with the result that the. eyes converge less for objects in the foreground than for those in the back ground; for which reason, what should appear near appears far, and vice versa. It is impossible, on the other hand, to see how the mere interchange of dis similar pictures, per se, could produce such phenomenon. It appears to me that the theory that "The perception of depth is caused by a slight nonidentity of the two retinal images" is definitely
disproven from the above considera tion. The play of the muscles of accommo dation is next in importance. As in the case of convergence, the distance of ob jects corresponds to the accommodative effort, though again beyond twenty feet it is of no value in judging distance. As already mentioned, it is a great aid to monocular stereoscopic vision. Let us bear in mind that even though we look at a series of solids with only one eye, we still have a strong sense of depth. Thanks to the play of accommodation and several other minor factors, we are able to get a fair sense of depth without binocular vision (convergence play). The next subjective element in our sense of depth is blurring of a part of the field of vision (in an antero-posterior direction). Experience has taught us that when objects are blurred within twenty feet, it is either because they are in front of or beyond our point of convergence (doubling); or because they arc in front or beyond our point of accommodation (out of focus). Or dinarily we are but vaguely aware of the condition, but nevertheless uncon sciously, we are in part guided by it in projecting objects into their proper depth. The blurring of images because of increasing distance and because of atmospheric effects (decrease of air transparency, etc.), is of great value in perspective vision beyond the twenty feet limit. * It is common knowledge that we judge distance of objects by comparing their apparent size with what we con sider their real size. And this is an other useful factor in perspective vision beyond twenty feet. That to a certain extent we judge the relative position of objects as we are accustomed to find them, is perhaps not so generally sus pected, but is nevertheless true. For example, a half mile away, we cannot by our natural optical means really tell whether a tree is in front of a building or in its wall. Nevertheless, knowing from experience that trees grow in front of buildings and not in their walls, we are strongly under the im pression that we really see it in front of the building.
S T E R E O S C O P I C AND P E R S P E C T I V E VISION
In this connection, it m a y be m e n tioned t h a t the a p p a r e n t direction of m o v e m e n t of objects and their m o t i o n in general, is of s o m e aid in b o t h per spective and stereoscopic vision, as is particularly the case w i t h the " m o v i n g picture." A s to stereoscopic vision as a factor in perspective vision, it is b u t necessary here to repeat t h a t it is in itself an i m p o r t a n t g u i d e in e s t i m a t i n g the depth of objects, since the solid appearance of objects decreases with distance. I a m n o w a w a r e that it will be ob jected to the idea of r e g a r d i n g m i n o r e l e m e n t s , here e n u m e r a t e d , as g e n u i n e factors of stereoscopic a n d perspective vision, on t h e g r o u n d t h a t they a r e largely mental in their n a t u r e , and not real optico-physiologic p h e n o m e n a (such as sense of c o n v e r g e n c e , accom m o d a t i o n , e t c . ) . T o this I m u s t point out that n e i t h e r the sense of solidity nor the sense of depth are purely ocu lar. Not all the factors of stereoscopic and perspective vision combined could produce either the sense of d e p t h . o r solidity w i t h o u t the association of oilier senses, particularly the tactile and m u s cle senses (of the body in general ). Suf fice it to mention here, the well known observation that those sceing-for the first time must develop a sense of depth and solidity by learning to associate their ocular impression's with their tactile and muscular experiences. In other words. the senses of depth and solidity are neither essentially ocular, nor tactile, nor muscular in their n a t u r e ; but are a psychic complex, outgrowing principally from the above three senses, but em bracing in addition other mental reac tions to bodily experience. It may now be of interest, for the pur pose of e m p h a s i z i n g the relativity <>f the senses of depth and solidity, to briefly examine the combinations of elements
239
which enter into the various forms of image formation. B e g i n n i n g with the p h o t o g r a p h , the "flattest" i m a g e of all, we find t h a t w e are by no m e a n s justified in c o n s i d e r i n g it devoid of all stereoscopic and per spective effects. A n d it is indeed sur p r i s i n g h o w m a n y of the e l e m e n t s are present. T o begin with, in p h o t o g r a p h s in which t h e light is well han dled, the s h a d o w factor does produce a s t r o n g stereoscopic effect. A s to per spective effects, all factors, except those of a c c o m m o d a t i o n and c o n v e r g e n c e play, are p r e s e n t . T h e sense of depth due to the b l u r r i n g of objects which are out of focus is especially m a r k e d in p h o t o g r a p h s , for which reason a good p h o t o - p o r t r a i t rivals a p o r t r a i t paint ing in stereoscopic effects. In p a i n t i n g s this factor is a b s e n t , since the artist a c c o m m o d a t e s , and therefore paints all objects within t w e n t y feet with equal clearness. T h i s is quite a defect, and is one reason w h y p a i n t i n g s lose in perspective the closer they are looked at. It is h o w e v e r partly c o m p e n s a t e d by objects b e i n g r e p r e s e n t e d "complete"-; i. e., as seen with both eyes. In addition, the artis tic h a n d l i n g of a t m o s p h e r i c effects in landscapes ( a n d o t h e r stimuli to i m a g inafon lacking in the p h o t o g r a p h ) , m a k e s the effect of depth in p a i n t i n g s far superior to p h o t o g r a p h s . T h e stereoscope, already discussed, differs from the image formation in ac tual life by lack of a c c o m m o d a t i o n play, and indeed lack of a c c o m m o d a t i o n al together, since it is replaced by the use of convex lenses. As to actual life, it is b u t necessary to repeal that all factors of stereoscopic and perspective vision are p r e s e n t u p to the t w e n t y feet limit. B e y o n d t h a t everything depending upon con vergence and a c c o m m o d a t i o n is, of course, absent.