CHAPTER XIII

VISION

The eye is enclosed in a globe of fibrous tissue, of which the front part, or cornea, being transparent, admits light. The epithelial layer which covers the cornea, conjunctiva, is also transparent. No bloodvessels enter these colourless tissues, unless as the result of inflammation due to infection or to exposure to sunshine or dust. For nutrition they are dependent upon the plasma which, exuding from, and returning to, the vessels which surround them, circulates in their tissue-spaces. In advancing years, when the circulation is less brisk, a ring of opaque tissue, arcus senilis, encroaches on the cornea. In the interior of the globe, just behind the cornea, is a projecting shelf, formed of a ring of tissue supported by buttresses, ciliary processes. It is continued inwards as the iris, a muscular curtain. The “hyaloid membrane” lines the back portion of the globe. Continued on the inner side of the ciliary processes, it splits into several layers, which pass, one in front of the lens, others to its edge, to which they are attached, and still another, very thin, behind it. Since it holds the lens in place, the anterior portion of the hyaloid membrane is known as its “suspensory ligament.” Thus the eyeball is divided into three chambers. The anterior is filled with watery lymph, aqueous humour. In it, resting on the anterior surface of the suspensory ligament of the lens, is the iris. The middle chamber contains the lens. The posterior chamber is filled with a liquid jelly, vitreous humour.

By the contraction of the circular fibres of the iris, the aperture of the pupil is diminished, limiting the light which enters the globe. This adjustment occurs when the illumination is bright. It is also brought into action for the purpose of cutting out divergent rays, which would not be clearly focussed when objects near at hand are looked at. The posterior surface of the iris and the inner surfaces of the ciliary processes are covered with dense black pigment. It is this pigment, showing through the uncoloured connective tissue and plain muscle-fibres of which the iris is composed, that gives their colour to grey and blue eyes. In many eyes the iris contains a brown pigment in its substance.

The back portion of the globe of the eye is covered with a curtain, the retina, formed by the spreading out of the fibres of the optic nerve in front of various layers of nerve-cells and the sensory cells of the organ of vision, rods and cones. The retina lies between the hyaloid membrane, which encloses the vitreous humour, and a layer of pigment which “backs” it, as a photographer backs a plate when he proposes to use it towards a source of light—to take a photograph of a window from within a room. The serrated margin of the retina is somewhat anterior to the equator of the eyeball. The pigment which backs the retina is contained in a sheet of cells which belongs to the pouch of brain that extended outwards towards the eye-pit (p. 334). Properly speaking, therefore, it is a layer of the retina.

Three sets of tissues take part in the development of the eyeball. (1) The epithelium covering the surface of the head is depressed as a pit, which gradually closes into a hollow sphere. This sphere, when its cavity is filled up, owing to the great elongation of the cells of its posterior half, becomes the lens. It breaks away from the rest of the epithelium of the surface, which clears to transparency as that part of the conjunctiva termed the “corneal epithelium.” (2) The retina, as already stated, is a hollow outgrowth from the interbrain. As this pouch approaches the lens, its anterior half is pushed back into the posterior half, forming a cup with a double wall. The anterior, or inner, sheet of the bowl of the cup develops into the nervous layers of the retina, the posterior sheet into its pigmented epithelium. (3) Connective tissues are transformed into the other constituents of the globe—cornea, iris, vitreous humour, etc. The globe is complete, except at a spot on the nasal side of its posterior pole where the optic nerve pierces it.

The bloodvessels of the retina, entering with the optic nerve, ramify on its anterior surface. Under ordinary circumstances we ignore the shadows which they cast, as we ignore the blind spot which coincides with the disc of insensitive tissue presented by the end of the optic nerve, and many other imperfections; but it was shown by Purkinje many years ago that by a very simple manœuvre they may be forced upon our notice.

By making use of _Purkinje’s figures_, it can be proved that the level in the retina at which undulations of light give rise to the impulses which evoke visual sensations coincides with the back of its anterior sheet—_i.e._, with the layer of rods and cones. A person stares fixedly at a white sheet in a dimly lighted room while an assistant, by the help of a lens, focuses a strong light on the front of his eyeball, to the outer side of the cornea. The rays, traversing the white of the eye, throw shadows of the retinal vessels on the layers behind them; but this not being the way in which light normally enters the eyeball, the person experimented upon supposes that he sees the shadows in front of him. He mentally projects them on to the white sheet. The pattern of his retinal vessels appears on the sheet in grey streaks. When the spot of light is moved, the shadow-pattern shifts, and in the same direction; since, as the retinal image is reversed, a movement from right to left is interpreted by consciousness as a movement from left to right. Given the angle through which the light is moved, and the apparent displacement of the shadows, it is a simple matter to calculate the distance behind the bloodvessels of the sensitive layer of the eye. So definite are Purkinje’s figures that the shadows of individual blood-corpuscles can be followed, and the rate at which they are moving in the capillaries of the retina calculated.

The retina is the organ of vision. Cornea, iris, lens, vitreous humour, are parts of the camera in which this sensitive screen is exposed; and of the retina, the sensitive layer is the layer of rods and cones. Interest therefore centres in these structures. They are disposed with the utmost regularity on the posterior surface of a thin, reticulated membrane—the outer limiting membrane. But rods and cones are only the outer halves of sensory cells, the inner portions of which, reduced to a minimum in thickness, except where they contain their nuclei, lie in the outer nuclear layer. Rods are the larger elements. Each consists of an outer segment, or limb, of relatively firm substance transversely striated, and liable to break into discs; and an inner limb of much softer substance, again divisible into two parts, the outer longitudinally striated, the inner granular. Cones are almost identical in structure with rods, save that their outer limbs are much smaller, their inner limbs rather fuller. In frogs and various other animals, but not in Man, each cone contains at the junction of its two limbs a highly refracting globule of oil, often brightly coloured, red, yellow, or green.

The layers in front of the rods and cones contain nervous elements accessory to them. In the “inner nuclear layer” are the ganglion-cells of the retina, homologous with the cells of the ganglia on the posterior roots of spinal nerves; but, in the retina, bipolar and extremely minute. On either side of the rather thick layer occupied by the nuclei of these ganglion-cells (and of cells of other types which, for the sake of clearness, we omit) is a felt-work of nerve-filaments in which their two extremities arborize. The most internal, or anterior, layer consists of a single sheet of rather large collecting cells and of their axons, which stream towards the optic nerve. Each cone has its proper ganglion-cell, collecting cell, and efferent fibre. Rods are served in groups by ganglion-cells and collecting cells. From this it may be inferred that a cone is a sensory unit, an inference confirmed, as we shall show presently, by direct evidence. The connections of the rods show that they also are sensory elements, although it may be doubted whether they are sensory units. The optic nerve contains a very large number of fibres—about a million—all small, but some distinctly larger than the rest. The largest very probably belong to the collecting cells of rods. But the retina certainly does not contain a million collecting cells. A considerable residue of fibres is therefore unaccounted for. It is supposed that they are afferent to the retina, but we have no knowledge regarding the nature of the impulses which descend from the brain.

The retinal pigment is not merely a backing for the sensitive screen. It undoubtedly plays an important part in vision. That it is not essential is evident from the fact that albinos, whose eyes appear pink owing to the absence of pigment, and the consequent showing through of the blood in the exceedingly vascular membrane which lies behind the retina, can see; although their visual sense cannot be described as normal. They are exceptionally sensitive to an excess of light. We shall return to this subject after describing the differences in manner of functioning which distinguish rods from cones, differences so marked as to justify us in speaking of two kinds of vision.

During twilight warm tones gradually fade out of the landscape; cold blues and greys predominate. A time arrives when scarlet poppies look black, although yellow and blue flowers and green leaves can still be dimly distinguished. In full daylight colours are seen at their brightest in the high lights; where the light is dim they tend to appear in different shades of grey. At night, if the sky is star-lit, all colours give place to a slightly bluish grey in the high lights, black in the shade. But a not very uncommon abnormality is night-blindness—inability to see at all when the light is not bright enough for the recognition of colours. In persons so affected the rods do not function; for it is with the rods that we see in weak light. They record differences in intensity between the lower limit of their sensitiveness and the higher degree of brightness, at which they are superseded by cones; but they afford no information regarding colour. Their monochrome is interpreted by the mind as a bluish grey, apparently because, since they are insensitive to red rays, the sensations of which they are the source are associated with the blue end of the spectrum. When the cones are stimulated very slightly, the reinforcing grey of the rods enables us to distinguish all other colours, save red, which appears black. In bright light the rods are in a permanent state of exhaustion; they do not contribute to vision. Rods respond to stimulation more slowly than cones. This fact enables us, by a very pretty experiment, to distinguish the two kinds of vision. A disc of green paper about the size of a threepenny-bit is pasted on a red surface. Held at arm’s length in a room lighted by a single candle, the disc looks dull green when the gaze is directed at it; but if the gaze be directed 2 or 3 inches to one side of it, it appears brighter than before, but less distinct and almost grey. The explanation of this is to be found in the fact that at the posterior pole of the eye there is a shallow cup—fovea centralis—which carries cones only, without rods. This small depression is the area of direct vision, the only spot at which we see things quite distinctly. At the fovea the nuclei and nerve-cells of the retina are withdrawn from in front of the cones to the margin of the cup, in order that they may not interfere with the passage of light. The pit and the ring round it contain some yellow pigment. Hence it is usually termed the “yellow spot.” When we are looking straight at the green disc, it is focussed on the yellow spot. It then excites a sensation of greenness; but since this is not reinforced by any rod-sensations, the green is dull. When it is focussed outside the yellow spot, it stimulates rods and the sparse cones which lie amongst them; and the rods being more sensitive than cones to light of low intensity, the disc looks brighter. If, while the observer is still gazing fixedly at a spot to the side of the disc, the red paper be waved rapidly, but gently, to right and left, a brightish grey cover seems at each movement to slip off the dark green disc, and to regain its position a moment later, with a jump. The grey rod-sensation, developing more slowly than the green cone-sensation, is, as it were, left behind. The two are separated at the moment when the paper starts to right or to left.

Astronomers have long recognized that one of the smaller stars which catches the attention when they are not looking directly at it may be invisible when the gaze is directed to the spot where it ought to be. It was visible when focussed on rods, but it is not visible when focussed on cones. In most birds the retina shows cones alone. To anyone who for the first time enters a dovecote at night the experience is very curious. A candle is for him a sufficiently strong illuminant, but it does not give light enough to enable the pigeons to see. Although evidently alarmed by the noise made by the intruder, they allow themselves to be taken down from their perches without making any attempt to escape. If, startled by the touch of a hand, they take to flight, they fly against the wall. Pigeons are night-blind. The retina of an owl bears chiefly rods, the outer limbs of which are exceptionally long.

The outer limbs of the rods are coloured reddish-purple. This colour is quickly bleached by light. If a frog which has been kept for a short time in the dark be decapitated, its head fixed for ten minutes in a situation in which a window is in front of it, then carried to a photographic dark-room, where an eye is taken out by red light, opened, and the retina removed, a print of the window will be seen upon it. Such an optogram may be fixed by dipping the retina in alum.

The retina is easily detached from its pigment-layer. If it has been bleached by exposure to light, it regains its “visual purple” when again placed in contact with its pigment. Evidently the visual purple is renewed from the pigment which lies behind (and around) the rods.

From the cells of the pigment-layer a fringe of streaming processes depends amongst the outer limbs of the rods and cones (Fig. 30). In a dull light the processes hang but a short way down; in a bright light they react almost to the outer limiting membrane. They supply pigment to the rods, but their relation to cones is not understood. It is clear, however, that the cones, although they are not coloured, are dependent upon the pigment-fringe, since they always remain in contact with it. Their inner limbs elongate in the dark, lifting them to the pigment, and shorten in bright light. These movements may merely indicate that the cones require a backing of pigment, but it would seem more probable that, like the rods, they absorb a substance which is sensitive to light, although we cannot recognize it by its colour.

The responsiveness of the rods to light is due to visual purple. As every lady is aware, colours, especially mauves and lilacs, are bleached by light. The chemical change affected by light in the colour of the outer limbs of the rods is the stimulant which originates impulses in the nerve-fibres connected with them, and it is generally believed that cones—the more highly specialized sensory cells—are stimulated in the same way. Visual purple is particularly abundant in all animals that range at night, with the exception of the bat. But its absence in the bat does not militate against the theory that it is the cause of night-vision, for it has been shown that a blind bat flies with almost as much freedom, and avoids obstacles—even threads stretched across the room—with as much skill as one that can see. It is guided by the bristles of its cheek. So, too, is the cat, which has the reputation of being able to see in the dark. Undoubtedly a cat’s eye is an exceptionally efficient organ in dim light, just as it is exceptionally sensitive to sunshine—it is provided with an iris which contracts the pupil almost to a pinhole—but the cat trusts to the bristles of its cheek for information regarding the things which block its path.

Most of the peculiarities which distinguish the reactions of the eye from those of other sense-organs can be explained by its mode of stimulation—the initiation of a nerve-current by a chemical change. No stimulus, if sufficiently strong, can be too brief. The retina reacts to an electric spark in the same way as a photographic plate; but, unlike the plate, the retina is restored to its previous condition of sensitiveness in about one-tenth of a second. A visual sensation lasts about one-tenth of a second. This prolongation of the sensation is, however, a mental, not a retinal, effect. The mind continues to see an object which has been illuminated by a flash until the retina is again in a condition to send brainwards a second impulse. Were our sensations coincident in duration with the stimulation of our sense-organs, we should live in a flickering cinematograph. When one is watching a moving point of light—the glowing end of a match, for example—the prolongation of sensation has its disadvantages; the moving point is interpreted as a streak of light. If the illumination be very brilliant, the object seen may give rise to a prolonged after-image. A glance at the sun leaves in the mind for seconds, or even for minutes, the image of a glowing disc. Sensations due to stimulation of the yellow spot last longer than those which originate in the peripheral retina. If, in a train, one is being carried at a certain pace, past a fence composed of upright palings, one sees the separate slats until the eyes are directed towards them, when they fuse into a continuous screen.

The phenomena of negative or complementary images are of retinal origin. The bright image of the sun, if the stimulus has not been too violent, gives place to a black disc. If one closes the eyes after staring at a window, a black surface crossed by bright lines is seen in place of a white surface with dark frames to the panes. If, after staring at a red surface, one looks at the ceiling, a green patch is seen; after yellow, blue. Every colour has its complement, which may be determined in this way. There is much uncertainty as to the exact terms in which this phenomenon is to be accounted for, but little doubt as to its being due to the peculiar mode of reaction of the retina to light. Chemical substances which have been used up have to be restored, and during the period in which they are coming back to what may be termed a neutral condition the retina delivers to the brain impulses of the opposite sign.

Contrasts which are experienced simultaneously are more difficult to understand than those which appear successively. In Fig. 31 the half of the grey cross which is surrounded by black appears brighter than the half which lies on white paper. A grey cross on a red background looks green; on a green background, red; on yellow, blue; on blue, yellow. If green is on red, it looks greener than if it is on white or black. These simultaneous contrasts are seen best when the strength of the colours is reduced by covering them with tissue-paper. It is as if activity of any one part of the retina is accompanied by activity of the opposite sign in the remainder. But it is unsafe, in explaining our various sensations, to lay too much stress on the mode of stimulation. The mind judges sensations in the light of previous experience. In anatomical language, the effect of sensations upon the personality depends upon the paths which impulses follow in the brain, and the associations which have been established by previous impulses which have followed the same paths. The retina enables us to distinguish tone and colour. By the variations in tone, the juxtapositions of light and shade, we recognize form. All streams of impulses which do not present tone-variations—do not, that is to say, reproduce the details of a scene—are interpreted in terms of colour. Every child discovers that the tedium of the intervals during which it is proper that his eyes should be closed may be relieved by pressing his knuckles against the lids. Although the world is shut out, a phosphene offers itself for his consideration—a yellow or white disc of irregular form with a red margin, changing into lilac bordered with green, and then into yellowish-green with a blue edge. Such, if my recollection can be trusted, were the pictures which I used to see as a boy; but no adjustment of pressure calls them forth with anything like the same vividness now.

All the senses show a tendency to rebound after activity, exhibiting contrast-phenomena; but the contrasts of vision are more marked and varied than those of the other senses, as everyone who is curious in the observation of his own sensations is aware. Negative after-images are generally referred to the retina; but various other kinds of after-image and contrast-phenomena must be attributed to the judgments passed by the mind upon the sensations which it receives; and not to physical changes in sense-organs. Positive after-images are well-marked appearances, although less common, perhaps, than the phenomena of reversal of sensation of which we have just written. On waking in the morning, one looks at the window; shifting the gaze to the ceiling, an after-image of the window appears, just as one saw it, with bright panes and dark frame. The “dark adapted eye,” being exceptionally sensitive, yields the same persistent positive after-image as the eye in its usual condition yields, after being directed towards the sun at mid-day. Movement-after-images can be explained only by referring them to misdirection of judgment. If the gaze is fixed on a rock close beside a waterfall, then shifted to a bank covered with grass or bushes, the part of the bank which occupies the lateral part of the field of vision appears to rush upwards, reversing the movement of the water. When the gaze has been fixed upon falling water—a narrow stream sparkling in sunlight—a central strip of the field moves upwards, the margins remaining stationary. If one stares at the spot on the surface of a basin of water on which drops are falling from a tap, and then looks at the floor, it is seen to contract towards the spot looked at, reversing the movement of the ripples in the basin. These observations reveal a fact of great importance in the physiology of vision. It is, probably, impossible truly to fix the gaze. The muscles of the eyeball keep the retinal field in constant movement—larger movements with minute oscillations superposed. When, as in watching a waterfall, movement has for a time taken a definite direction, its cessation is judged to mean reversal.

The anatomical unit of sensation is a cone. The fovea centralis, the only part of the retina capable of receiving sensations sufficiently discrete for reading, contains cones alone. If the gaze be directed but a very few millimetres on to the white margin of the page, letters lose their form. In the fovea the centre of one cone is 3·6 µ distant from the centre of the next. Two stars are visible as separate stars if they subtend an angle of at least 60 seconds with the eye. Their images on the retina are then 4 µ apart. Parallel white lines ruled on black paper, held at such a distance as causes them to subtend angles of 60 seconds with the eye, appear not straight but wavy, showing that their images are taken up, not by a continuous substance, but by the mosaic of cones. So far the explanation of the visual unit is strictly anatomical; but it must be added that trained observers can recognize the separateness of objects which subtend angles of much less than 60 seconds—not more than 5 or 6 seconds. This can be accounted for only on the hypothesis that images far closer together than the width of a cone produce a specific effect in passing across the anatomical unit.

In 1807 Thomas Young, the physicist, formulated a theory to account for =colour-vision=. He supposed that the retina contains three kinds of apparatus—_a_, _b_, and _c_—each especially responsive to a particular kind of light, all three slightly stimulated by rays of all colours. (Young imagined three kinds of nerve, but modern supporters of his theory suppose three different substances chemically changed by light.) A prism spreads out the rays which are combined in white light into a band in the order of their wave-lengths—those which have the longest wave-length (0·8 µ) and the slowest rate of vibration (381 billions to the second) at one end, those which have the shortest wave-length (0·4 µ) and the most rapid vibration (764 billions to the second) at the other: between these two extremes every intermediate grade of length and rapidity. These are a mere fraction—a small group—of the waves which the æther transmits, but they are all that we can see. The long, slow vibrations give rise to sensations which we describe as red; the short, rapid vibrations we describe as violet. Our names for the tints which intervene are singularly old-fashioned and unsatisfactory, but all persons agree that they recognize in the spectrum a certain number of definite colours. Some normal-sighted persons say twelve, others eighteen. It is largely a question of terminology.

Many considerations show that it is quite unnecessary to imagine that the retina is affected in a different kind of way by every kind of light, or by each of several groups of waves. If the red of the spectrum is mixed with yellow, we receive an impression of orange, which is identical with the impression produced by waves of the mean length of red and yellow; orange and green give yellow; yellow and blue, green. Any two complementary colours yield white. By taking three colours—say, red, green, and violet—we obtain, when they are duly mixed, not white light only, but light of any other tint, although not of spectral purity, since it is mixed with white. Young considered that all the conditions of colour-vision would be satisfied, all our various sensations provided for, if the retina contain three kinds of apparatus which light, according to its quality, affects in varying degrees; and with this theory of three kinds of apparatus—_a_, _b_, and _c_—the theory of three elementary or fundamental colour-sensations is indissolubly linked. The colour _x_ produces its intensest effect when _a_ is stimulated, with the least possible stimulation of _b_ and _c_; _y_ is the reaction of _b_, _z_ of _c_. Recent studies of the curves of intensity give us the tints of _x_, _y_, and _z_ as carmine-red, apple-green, and ultramarine blue.

The blending of sensations is illustrated with the well-known colour-top. But perhaps the most striking proof that three elementary colour-sensations are adequate to produce our visual world is afforded by photographs taken with the three-colour method. Three plates are exposed—(_a_) behind a red screen, (_b_) behind a greenish-yellow screen, (_c_) behind a blue screen. They are fixed in such a way that the portions acted upon by light are rendered insoluble, whereas the rest of the film can be dissolved away; _a_ is then stained red, _b_ greenish yellow, _c_ blue. The three are superposed, and the result appears to the eye as an exact reproduction of the subject of the photograph in all its hues. It shows every shade of orange and green and violet. It is as bright—that is to say, as full of white light—as the original.

Various objections may, however, be brought against Young’s theory. Of these, the most weighty are: (1) The retina does not contain three kinds of apparatus, as Young supposed; nor can we find three kinds of photochemical substances, as required by the theory in its modern form. If we could find them, a fresh difficulty would arise; for we have no reasons for supposing that one and the same nerve-ending can receive stimuli of three different kinds. (2) The theory offers no explanation of negative after-images—the complementary colours experienced when the eye is closed after staring at a brightly coloured object. (3) It does not adequately account for the various deficiencies of colour-blindness.

It is well recognized that there are various degrees of colour-blindness, and that the colour-vision of persons considered normal presents different grades of refinement. Nevertheless, the abnormalities of colour-blind persons are so marked that cases fall into definite classes. Those whose cones do not function—which means that their yellow spots are either undeveloped or diseased—see all things grey. They are totally colour-blind. Excluding these, the colour-blind may be grouped in one or other of two divisions—(_a_) those who confuse red and green, (_b_) those who confuse yellow and blue. One person out of every thirty-five is red-green blind. The proportion is even higher if males only are considered, showing how very unfortunate is our choice of warning signals. A man who is red-green blind cannot tell the port from the starboard light. Blue-yellow blindness is, on the other hand, extremely rare. According to Young’s theory, colour-blindness is due to the absence of one of the three sets of visual apparatus. But cases do not altogether conform to this hypothesis. We knew an amateur water-colourist, since deceased, who derived intense pleasure from the beauties of Nature, and showed no mean skill in reproducing them with his brush, notwithstanding the fact that he was red-green blind. Each night his sister arranged his paint-box for him, and only rarely did he use vermilion to fill in a foreground of lush green grass. But this mistake, when he made it, did not destroy his own satisfaction in the picture. It was clear that red had a value for him, although he confused it with green. It is impossible for a normal person to see through the eye of one who is colour-blind, and there is no other means of comparing his sensations with our own. The mistakes which the colour-blind make in sorting coloured objects and in naming mixtures of light selected from various parts of the spectrum show the range of their deficiency, but give us no information regarding the qualities of the sensations which they retain.

The test of colour-sensitiveness usually employed is the grading of a large number of wools of different tint. The order in which the colours should be arranged is not a matter of opinion. They must be placed in the order in which they occur in the spectrum—_i.e._, arranged according to their wave-lengths. In the cases of colour-blindness which are most frequently met with the defect may be described as due to an absence of the sense of redness, or as an absence of the sense of greenness. The two conditions can be distinguished. But since the eye is not dark for red (although in certain cases vision is very weak for the red end of the spectrum) or dark for green, the abnormality cannot be adequately accounted for on structural grounds. It is not explicable on the hypothesis that one of three sets of responsive sense-organs (or nerve-fibres) or photochemical substances is absent from the eye. Again, it is generally agreed that the sensations of white, yellow, and blue of the red-green colour-blind are similar to those of normal persons. This is not in harmony with the theory of the omission from their eyes of one of three pieces of colour-apparatus.

Professor Hering, of Leipsic, adopting the generally accepted view that light effects chemical changes in substances contained in the retina, to which changes stimulation of nerve-endings is due, formulated a theory of colour-vision which many physiologists prefer to Young’s. He imagines that the retina contains three kinds of pigment, each of which is, as he believes all living substance to be, in a constant state of change. It is at the same time being built up and destroyed. Using the terms which connote the opposite directions of metabolism, the pigment is simultaneously undergoing anabolism and katabolism; the two processes, when the retina is at rest, maintaining equilibrium. When light acts upon either of the substances, it hastens, according to its quality, either the one process or the other; and the chemical change, whether it be constructive or destructive, stimulates the endings of optic nerves. Hering assumes, therefore, that there are six elementary qualities of visual sensation—red, green, yellow, blue, white, black. Red, yellow, white are due to anabolism of the visual substances; green, blue, black are due to their katabolism. The installation of yellow amongst the unanalysable colours is a relief to many minds. It is almost impossible to think of yellow as a compounded colour. White also, we feel, is not a compounded colour, despite our knowledge that a prism scatters from it all the hues of the rainbow. Black, many persons assert, gives them a definite sensation, and not merely a sense of rest. (Parenthetically, it may be observed that the _feeling_ that a colour is pure or mixed is not to be trusted. It may be based upon the chromatic aberration of the eye, or it may be reminiscent of the paint-box. We know that we cannot make yellow by mixing red and green pigments, hence we feel that it is pure. Of green we are not by any means sure; gamboge and Prussian blue come into our minds.) Except when the light which falls upon the retina is giving rise to one of the four pure colour-sensations, all three substances are simultaneously affected, although one may be undergoing katabolism while the other two are being built up, or _vice versa_. Hering accounts for simultaneous contrast by assuming that the activity of any one part of the retina induces an opposite kind of change in the remainder, and especially in the vicinity of the primarily active part. When a certain patch is developing a sensation of red, the rest of the retina develops a sensation of green.

The great merit of the theory is, however, to be found in its offering an explanation of complementary after-images. The green patch seen with closed eyes after one has stared at a red object is due to the rebound of metabolism. In returning to a condition of chemical equilibrium the retinal substance acts as a stimulant which evokes the antagonistic colour. But it is a theory which makes very large assumptions. It assumes, for example, the possibility of the existence of a substance which is built up by light from one end of the spectrum, and decomposed by light from its centre. Not that Hering regards the existence of three retinal substances as essential to his theory. He is prepared to transfer to the brain the seat of the substances, or the substance, which, by their, or its, anabolism and katabolism, produces antagonistic colour-perceptions; but in this he is abandoning physiology for metaphysics. We have no warrant for imagining that there exists in the brain any substance which, by undergoing physical changes of various kinds, produces various psychical effects. The problem to be solved is physiological. Rays of light of different wave-lengths excite the retina to discharge impulses which are variously distributed in the brain. The effects which they produce in consciousness depend upon their distribution. The impulses to which the longest rays give rise evoke sensations of red, those due to the shortest, sensations of violet. And what is true of the retina as a whole is true, apparently, of each individual cone. In what way does light act upon a cone? It is one of the most fascinating problems in physiology. Round it our thoughts revolve whenever we are trying to form conceptions of the nature of stimulation, sensation, and perception. Each of the two theories which we have expounded above helps to group together certain of the more striking phenomena of colour-vision, but neither gives a satisfying explanation of their causation.

The sensitiveness of the retina is in a remarkable degree adjusted to the intensity of the light. When a dark room is entered, the pupil dilates; but one’s power of distinguishing objects continues to increase after the pupil has reached its maximum size. At the end of ten minutes the eye may be twenty-five times as sensitive as it was when the room was entered. This _adaptation to darkness_ is due in large degree to the substitution of rods for cones as the organs on which vision chiefly depends. But it cannot be wholly due to this, since it occurs when one is working with a red light. Probably the red used in a “dark-room” is not sufficiently near the end of the spectrum to be completely without influence upon visual purple, but it is a colour to which rods are comparatively insensitive. Other evidence also points to an adaptation of cones as well as of rods.

_Accommodation of the eye for distance_ is brought about by a mechanism which allows the lens to change in shape. It becomes more convex when a near object is looked at than it was when adjusted for an unlimited distance, which is its condition when the eye is at rest. Adjustment for near objects involves muscular action, and is accompanied by a sense of effort, however slight. Whilst the eye is at rest the lens is mechanically compressed against the anterior layer of its suspensory ligament. Accommodation for near vision is effected by the ciliary muscle, which is placed in the shelf of tissue which projects into the interior of the eyeball. This muscle is made up of a ring of circular fibres, and to the outer side of this, of fibres which radiate backwards and outwards. The longitudinal, or radiating, fibres obtain their purchase by attachment to the firm wall of the globe just beyond the cornea. They spread into the front of the loose chorioid membrane which lines the eye behind the retina. By the joint action of these two sets of plain muscle-fibres the suspensory ligament is slackened, and the extremely elastic lens, previously compressed, bulges forwards. The radius of curvature of its anterior surface changes from 10·3 millimetres for distance to 6 millimetres for vision at the “near point.” It was stated, in connection with the development of the lens (p. 374), that the cells of the posterior half of the hollow sphere out of which it is formed grow forwards into extremely long fibres, which traverse its whole thickness. These fibres are bent like the segments of a carriage-spring. Their anterior ends rest against the flattened ligament of the lens; the vitreous humour, which is always under tension, compresses their posterior ends. When removed from the eye, the lens becomes rounder than it is _in situ_, even when accommodated for near objects. But in later life it grows stiff. It ceases to bulge forwards when its ligament is slackened. Hence it becomes necessary to aid the presbyopic eye with convex glasses when it is used for near objects, although for distant vision it remains as effective as ever. If the ciliary muscle is constantly and completely relieved of the labour of accommodation, it grows lazy, or rather wastes from want of use. A person who relies on spectacles loses his power of accommodation; but ophthalmologists agree that self-focussing, if it give rise to a sensation of strain, is bad for the eyes. In myopic persons the eyeball is too deep; objects are focussed in front of the retina. In hypermetropia (“long sight”) the eyeball is too shallow; objects are focussed behind the retina. Concave glasses correct the one condition, convex glasses correct the other. Glasses are also very commonly called for to neutralize another defect—regular astigmatism—which may be present by itself, or may accompany insufficient length or too great length of the optic axis. It is due to unequal curvature of the cornea. Usually the curvature is sharper in the vertical than in the horizontal meridian (_cf._ p. 269); as a consequence, points in a vertical line are focussed in front of points in a horizontal line. Cylindrical glasses, not lenses, are required to correct this defect. And here it may be well to call attention to the fact that rays of light are more sharply refracted by the surface of the cornea than they are by the crystalline lens. The lens has a high index of refraction (1·45), but it does not lie in air (the index of refraction of which is 1), but between two humours which have about the same index as water—namely, 1·336. The bending by the combined action of the cornea and the lens of rays of light which come from a source so distant that they may be considered as parallel brings them to a focus on the retina, when the lens is at its flattest. When the lens is at its roundest, rays which diverge from a point only 5 inches in front of the eye are focussed on the retina. The lens is therefore essential for accommodation, but, after its removal for cataract, vision, even for near objects, is rendered possible by the use of convex glasses.

A star or a distant gas-lamp is seen as a point of light with rays. Usually this figure, which has given origin to the expression “star-shaped,” shows three greater rays alternating with three lesser rays. Such an image is not produced by a point of light near to the eye, since it is due to the puckering of the lens when flattened against its ligament. It brings into evidence the three axes on the front of the lens and the three axes which alternate with them on the back, with regard to which the lens-fibres are disposed.

As an adaptation of living tissues to optical purposes the eye is above admiration, yet it presents many =defects=, which an optician corrects in the instruments which he manufactures. A remarkable fact in the physiology of vision is our unconsciousness of the imperfections of its organ. An unusual experiment is needed to bring them to our notice. If we look through a common glass lens uncorrected for unequal refraction of rays of different wave-lengths, we recognize that a bright object is shown with a colour-fringe, yet we take no cognizance of the colour-fringes which surround the images of all bright objects focussed upon our retinæ. If we think about the matter, we recognize a feeling that blue in a window of stained glass appears farther away than red; but this might well be due to association. Blue glass is chiefly used for the sky. If we look at a bright object through purple glass, we her red with a blue fringe or blue with a red fringe, according as the eye is focussed for red or for blue. The purple glass having absorbed all intermediate rays, we become aware that we cannot focus the two extreme ends of the spectrum at the same place. Since a greater effort of accommodation is needed to focus red, we judge that the bright object is nearer to us when it appears red than when it appears blue.

Spherical aberration is another fault of the lens. The rays which enter its margin are brought to a focus sooner than those which pass through its centre. This is due to the fact that its surfaces are regularly curved, whereas a glass lens is corrected by grinding it flatter towards the margin. This defect is partly corrected by the cornea, which has an ellipsoidal surface, and partly by the greater density of the centre of the lens. Yet it is still necessary for the eye to be “stopped down” by the iris when a near object is looked at, although less light is entering the eye than when it is directed to the horizon—a condition which would lead a photographer to open his iris-diaphragm.

Of all the imperfections of the eye which the mind ignores, the most remarkable is the gap in the field of vision, due to the gap in the sensitive layers of the retina, which occurs where the optic nerve enters it—the blind spot. Hold this page of the book 10 inches from the face, keeping the lines of print horizontal. Close the left eye and look at X with the right eye. The black disc disappears, because its image is focussed on the blind spot. Since the picture on the retina is reversed, it is clear that the optic nerve enters the globe to its inner side, and slightly above its horizontal meridian. But, unless we employ an unusual test, we are quite unconscious of the fact that a definite hole is punched in the picture. The mind fills it in, and the way in which it does so is extremely suggestive. It lies about it—in a downright ingenuous fashion if it is confident of credence, in a more subtle way if a simple falsehood is likely to be challenged. In place of the black disc make nine conspicuous crosses:

Hold the paper in such a position that _X_ falls upon the blind spot. It ought to disappear, but the mind assures you that there is a cross at that spot. The mind completes the field. In place of the crosses use noughts and crosses, thus:

Now let _X_ fall on the blind spot, and allow the eye to go just a little out of focus. The four marginal crosses draw inwards:

The mind contracts the field. Still denying the gap, but not having sufficient data from which to invent an object, the fraudulent nature of which would not be found out the instant that the gaze is shifted, the mind lies regarding the position on the paper occupied by surrounding objects.

Is it quite fair to the mind to say that it lies about the blind spot? The mind judges sensations in the light of experience. An association of previous sensations teaches me that the wall of the room is not pierced by a round hole a foot in diameter opening into outer darkness. Many sensations to me the fact that the designs on a wall-paper succeed one another with unbroken regularity. Fixing my gaze on one of them, I cannot by any effort of attention efface the pattern which happens to be focussed on the blind spot. I know that I shall see it the instant that I move the eye. If I let my eye roam until the face of my wife falls on the blind spot, its image disappears. I know its lineaments far better than I know the pattern on the wall-paper, but I cannot fill it into the picture. Her hands are visible, and the work which is resting in her lap, but in a mysterious way the background draws together where the face should be. My mind refuses to pass a false judgment; but it also refuses to see that there is a gap.

This exceedingly instructive observation teaches the relativity of sensations. It shows that a sensation has no objective value until judgment has been passed upon it by the mind. The meaning of this we express in figurative language, none other being available. We speak of a new sensation as being compared with sensations previously received—taken into the picture-gallery of the mind, and placed in its due position amongst the infinitely numerous records which are stored there. If we try to make a nearer approach to correlating physical with psychical activity, we say that sensation has no value save that which it acquires from its temporal relation in the sequence of sensations to which attention is directed, and that this value depends upon the relation which similar sensations have possessed in former sequences. There is no gap in binocular vision. An object focussed on the inner (nasal) side of the right eye, where the blind spot is situate, is focussed on the outer (temporal) side of the left eye. The left eye sees the object to which the right eye is blind. Since we have almost invariably used two eyes in the past, experience teaches that there is no gap in the field of vision. Hence the new group of sensations which alleges that there is a gap must be corrected. The field must be filled up in the way which experience shows to be most likely. The retina is a sheet of rods and cones, each of which has a nervous connection with the brain proper to itself. The retinal field is associated with the brain-field. But this does not imply that we may think of the mind as having a spatial distribution on A or button B in the retina causes bell A′ or bell B′ to ring in the brain, but it does not follow that perception A′′ or perception B′′ will be heard in the mind. It will be heard if this is the association established by custom, since mind is the product of experience. But the new sensation is creating precedent as well as being judged by it.

Point A in the right retina is associated by experience with point _a_ in the left, and point B with _b_. These are termed _corresponding points_, because they are similarly stimulated in binocular vision. The mind, therefore, judges that it receives the same information from each pair of corresponding points. The position of corresponding points will be understood if the right retina is imagined as put inside the left, precautions being taken to make the yellow spots coincide, and to avoid twisting the retinal cups in taking them out of the eyeballs. Great care is taken to maintain the points in correspondence during the various movements of the two eyeballs. In addition to the four recti muscles which move the eyeball upwards, downwards, to right and left, two oblique muscles give it the requisite amount of rotation. We have learned to give the same value to the impulses from two corresponding points. But under changed conditions the correspondence changes. When a squint develops in childhood, it follows one of two courses; either the obliquity of one of the eyeballs increases until it looks towards the nose, and its images cease to interfere with the images in the dominant eye—they are ignored by the mind—or a fresh correspondence is established between points in the oblique eye and points in the eye which looks straight forward. If we are severely critical, we find, from a study of the form of the eyeball, that it is impossible that the same rods and cones should occupy corresponding points in different positions of focus and with different degrees of convergence of the eyeballs. To permit of this the retinal cups would need to change in shape. But again mechanical correspondence is of little consequence. In the light of experience the mind judges that points correspond. When we are gazing at a flat surface, the mind judges that corresponding points are giving it similar information. It does not see a flower on a wall-paper twice as bright or twice as red with two eyes as with one. If the eyes are normal, the impression received through the two is precisely the same as the impression received through either singly. But when we are looking at solid objects, the image on one retina is not the same as the image on the other. One eye sees farther round the object on the one side, the other on the other; and it is just this disparity in the pictures, aided by the feeling that the eyes are converging, that gives the impression of solidity. Correspondence of points, on the other hand, is not necessarily sufficient by itself to convince the mind that the pictures presented by the two eyes are identical. When a flat triangle such as this is regarded with the two eyes, its black lines fall on corresponding points; but the figure is associated in the mind with other sensations—sensations of movement and touch. Notwithstanding the identity of the retinal images, the mind tries to see them as disparate. The figure troubles the eyes. At one moment the meeting-point of the three central lines projects forwards, at the next it recedes. That similarity of retinal images counts for something is shown by closing one eye. The uncertainty of shape of the figure is rendered more troublesome. It changes still more rapidly from convex to concave. When the point seems to be in front of the page, the accommodation of the eyes is adjusted for nearness; when behind the page, for greater distance. But the illusion that the object occupies three dimensions is not dependent upon the sense of contraction of the ciliary muscle. When the paper is moved towards the eye, its centre recedes; it is left behind until the ciliary muscle has had time to contract. When it is moved away from the eye, it projects until the ciliary muscle has had time to relax. Accommodation follows judgment, not judgment accommodation. The mind is extremely suspicious of the veracity of its newsagents. Disparateness of images, convergence of the eyeballs, shifting of accommodation for the various levels of an object in space, should be indisputable evidence of solidity or of hollowness. Conversely, the absence of either factor should be conclusive proof of flatness. But the mind does not trust to isolated sensations; it looks for associations of sensations. When the finger hints, “I could touch that sharp point,” it is useless for the eye to aver that there is no point to be touched.

If two exactly similar photographs are placed in a stereoscope, the fact that the eyes are not converged gives to the common picture an appearance of depth, notwithstanding the fact that corresponding points on the two retinæ are stimulated. If the two photographs have been taken, as they should be taken for this purpose, with a double camera, the disparity of the retinal images immensely enhances the impression of solidity.

It is impossible to exaggerate the dependence of sensation on =judgment=. At birth a child commences the long process of education which enables it to associate the sensations derived from its retinal images with the movements which place it in contact with things. It discovers that, when it is necessary to make the eyes converge, the object is near at hand. It also associates the voluntary action of contracting its ciliary muscle with nearness. Unconverged and unaccommodated eyes come to mean distance. So, too, do indistinctness due to absorption by the atmosphere, blueness due to the same cause, a small image on the retina. But there are obvious limits to its power of ascertaining the distance of an object, and therefore, conversely, of its power of estimating size. We have no idea of the size of the retinal image of the sun. Very few people would be prepared to believe that the angle which the sun subtends with the eye barely exceeds half a degree. (The first finger, viewed in profile, at arm’s length, covers one degree of arc.) A disc of paper of the right size, placed at the right distance, looks far too small to represent the sun. The most brilliant of orbs bulks larger than this in our minds. Everyone who for the first time looks at the sun through well-smoked glass, or, better, through a flat-sided vessel filled with ink and water, is astonished that it looks so small. Nor are we prepared to accept the evidence of a camera that the sun at the zenith does not produce a smaller image on the retina than the sun when rising above the horizon. Yet if a photographic plate is exposed to the rising sun, and again, without changing its focus, to the sun at the zenith, the two images are practically equal. There is a slight difference due to the greater refraction of rays passing tangentially through the atmosphere, but it is so slight as to bear no relation to the difference between our two judgments of size. When the sun is rising behind trees and houses, we compare it with objects which we know to be large and distant; yet it looks almost as large when rising out of the sea. One of the causes of the illusion is our conviction that the sky is flattened; and this, again, is due partly to its paler tint—its less substantial blueness—near the horizon, and partly to our impression that it is spread out over a flat earth. When the sun is in what we deem to be the more distant part of the vault of heaven, we judge it to be farther from us, and therefore larger than when it is above us. Yet the last word has not been said in explanation of a phenomenon which has been studied by mankind since the dawn of science. Helmholtz attributed the apparent greater distance, and consequent greater size, of the sun and moon when near the horizon to the indistinctness of their discs. When its image is so reflected from the zenith as to cause the moon to appear to rest upon the horizon, it does not, he said, increase in size. In answer to Helmholtz’s explanation, it may be objected that, when at midnight he brought the full moon down from the zenith, he did not bring with her the conditions of light and colour by which she is customarily surrounded when floating on the horizon. If, when watching the moon which has just risen, vast in diameter, out of the sea, one interposes between it and the eye a sheet of paper in which a small hole has been made, and looks at the moon with one eye through the hole, it instantly shrinks to the size which it appears to have at the zenith. It is not even necessary to blot out the whole of its trail of light on the sea. At the same time, it appears to retreat to a great distance. This shows how complicated are the associations upon which judgments of size and distance are based, and to how small an extent they are determined by the size of the image on the retina. This observation is most surprising if made one or two nights after full moon, when twilight is already dim at moon-rise.

Our estimate of the distance away from us of an object on the horizon is based upon the time and effort which experience tells us we should need to spend in reaching it. The untried appears shorter than the tried. Anyone who compares his feeling of the number of yards he would have to climb up a pole reaching to the zenith with his feeling of the number of steps he would need to take to reach the horizon will recognize that the horizon appears to him to be the farther away.

In representing a solid object an artist conveys theidea that light is falling obliquely upon it. One side of the object, therefore, is more strongly illuminated than the other. By depth and gradation of shade he indicates the extent to which the thing projects forwards, if solid, or falls back, if hollow. He makes the margin of a ball hazy, in the expectation that the spectator will look at the spot nearest to him—an artifice which he may easily press too far, since the eyes wander restlessly over a flat surface. In representing distance he is dependent upon giving to the various objects in his picture sizes equivalent to the sizes of their images on the retina, making them brighter or paler and more or less distinct. Yet he cannot hope to simulate the convincing evidence of distance which is afforded by our sense of the degree of convergence of our eyes. Hence, as Francis Bacon pointed out, a picture appears more real when one eye is closed than when both are open. Its middle distance at once falls back.

Innumerable are the illustrations which may be given of errors of sensory judgment, but none are more striking than the various figures which may be drawn with converging or diverging lines. The mind under-estimates acute and over-estimates obtuse angles. It is impossible to convince oneself that in Fig. 36 the line A bisects a symmetrical arch. Equally difficult is it to believe that in Fig. 37 the line with diverging terminal segments and the line with converging terminal segments are of exactly equal length. In the Ruskin Museum at Sheffield there is a sketch by the master of the façade of a church which shows a vertical tower to one side of a triangular pediment, or, rather, this is what the sketch was meant to show, and does show, when measured on an architect’s table. In effect the tower appears to be leaning towards the pediment. Errors of judgment of this type have been attributed to the curvature of the lines of a rectilinear image on the retina, the mind judging the distance between two points by the length of the chord, and not the length of the arc which joins them. This is very simply illustrated by the example of the apparently greater length of a filled space than of a vacant one.

A B looks longer than B C. If A B C be represented as a curved line, the arc A B will, of course, be longer than the chord B C. But it is not safe to suppose that the mind compares the length of an arc with the length of a chord. Judgment is based upon experience, and probably the illusion is due to more subtle causes than the curvature of the retina. The mind does not look at the retina. If it did, it would find the reversal of the picture the least of the inaccuracies which it had to correct. It would find it very difficult, for example, to superpose in its stereoscope the photographs of a vertical tower taken simultaneously by the right eye and the left. The curved images on the retina of the vertical lines which define the angles of the tower, as seen with one eye, could not be made to correspond with the images focussed by the other eye. The Greeks felt this when they settled the form of a column. The canon of the swelling entasis and increasing taper above it did not destroy the appearance of uniform thickness which the shaft presented. It gave to the eye just the slight help which it needs to enable it to picture the shaft as of the same thickness from base to capital.