The most important and far-reaching of all the sensory avenues of knowledge concerning the external world is through the sense of sight, which pictures not only the changing scene, but also records place, movement, and distance of outside objects. Most of the other senses, such as taste, smell, touch, and temperature, furnish information concerning only the immediate environment. It is true that the radius from which impressions arrive is much enlarged by the sense of hearing, but hearing after all is confined to geographical distances even when augmented by the marvelous mechanical extensions that are made possible by radio. Although one must travel in order to “see” distant lands that lie beyond the horizon, as well as to come into contact with distant stimuli of any sort, it is quite possible to stand still upon a starry night and see the heavenly bodies that mark the very outposts of the known universe. What an infinitude of space is comprehended in the statement that one can see a star!

None of the human senses is so sorely missed as sight. After Beethoven became deaf and could no longer hear the musical harmonies which teemed in his fertile brain, he became an object of sympathy, but a blind astronomer or painter, living upon memories of what he can no longer see, would surely be a sadder figure.

Nevertheless, sight is by no means an universal or an indispensable endowment of animals, for many creatures are sightless, or “love darkness rather than light.”

Eyeless plants are more dependent upon light than animals are, and although all organisms alike owe their existence either directly or indirectly to the light of the sun and the marvel of photosynthesis, it is not always necessary for so elaborate a receptor as an eye to be involved in reactions to light.

Photoreceptors, which are structures adequate to respond to the stimulus of light, include much more than “eyes.” In fact protoplasm generally is more or less sensitive to light, and the function of sight may be regarded simply as a specialized extension of this peculiar type of irritability.

Photoreceptors That Are Not Eyes

If the eyeless earthworms did not retire to the safety of their burrows at break of day, after their nocturnal wanderings, the proverbial “early bird” would quickly eliminate them in the struggle for existence. They are able, however, by means of certain specialized photoreceptive cells in the skin, to distinguish light from darkness and so usually to escape such a fate.

Many protozoans, as well as larval forms of metazoans in which cellular elaboration of eyes is quite out of the question, nevertheless respond very definitely to the stimulation of light. They are said to be positively phototropic when they turn toward the source of light, and negatively phototropic, when they turn away. Usually these responses are beneficial to the animal concerned but not invariably so, for positively phototropic moths are known to be killed by flying into a flame.

As highly developed an animal as a vertebrate may possess photoreceptive integumentary cells located outside the eyes, as proved by the behavior of certain chameleon-like lizards, which normally respond to light by color changes in the skin, and which make this response when temporarily blinded, if a stimulating ray of light in an otherwise darkened room is focussed upon the skin.


Eyes are the photoreceptive organs par excellence. They may be described as of two different general sorts, namely, direction eyes, that distinguish light and dark and enable an animal to locate the source from which the stimulus comes, and image-forming eyes, that report to the brain a more or less definite picture, reflected from objects in the environment.

Direction Eyes

Direction eyes are typically shown in non-parasitic flatworms, or Turbellaria, which are found in the daytime out of reach of their enemies clinging to the underside of stones and sticks submerged in shallow water.

Direction eyes of a flatworm

The photoreceptive cells in direction eyes are packed closely together behind shield-like cups of pigmented cells that are not penetrated by light (Fig. 704). The angles at which these cups are placed on the two sides of the head is such that it permits the light, whenever it does not fall exactly parallel to the long axis of the body, to stimulate the photoreceptors on one side more than on the other side, so that the worm responds by turning until the stimulation received on both sides is equal. This results in orientation with reference to the source of light, and in a negatively phototropic flatworm, tends to carry it into darkness and safety.

Image-forming Eyes

Image-forming eyes are optical devices that not only differentiate between the presence or absence of light, but also receive reflected light in such a way as to transfer a picture to the brain.

There are two outstanding types of image-forming eyes, namely, mosaic and camera eyes. The former reaches its highest elaboration in the faceted eyes of insects, which far outnumber all other animals having eyes of any kind, but the latter is of greater present interest, because it is the type of the vertebrate eye.

Behind each facet of the compound eye of an insect are several parts which together constitute a separate optical instrument for receiving light that has been reflected from external objects. This unit of the compound arthropod eye is called an ommatidium. When clustered together, sometimes several hundred in a single eye, ommatidia produce a map of overlapping details that fit together, like the separate elements in a mosaic pattern, to form a single picture in the brain.

Usually the mosaic eyes of insects are set immovably in the head, but the convex exposure of their surfaces is such that the. marginal ommatidia may point outward at angles sufficient to include in the whole eye a wide range of vision.

In crustaceans it is usual for the mosaic eyes to be so mounted upon movable stalks that they may be turned in various directions, without the necessity of moving the rigid neckless head.

Human Eyes

A parallel between the typical camera eye of a vertebrate and a photographic camera holds good in numerous details. Both are dark chambers which admit the light through a focussing lens and which are guarded by a diaphragm and a shutter-like arrangement that regulate the amount of entering light. Both have a sensitive receiving surface, lining the side of the chamber opposite the window where the light comes in, and in both there is an inversion upon the sensitive screen of the image received. In one case the picture impressed upon the sensitive plate is “developed” after removal to an appropriate bath of chemicals. In the other instance the picture, received upon the sensitive retina of the eye, is transmitted by the optic nerve to the brain, which becomes its “developer.”

The eyeball, or “camera box” of the vertebrate eye, is safely lodged in a bony orbit of the skull, forming a protection to it from mechanical injuries, except for a small though necessary exposure on the front face. It is held loosely in place by a loose system of connective tissue fibers which permits considerable freedom of movement within the orbit.

Horizontal meridional section through the right eye of man, schematic

The walls of the eyeball consist of three concentric layers of diverse tissues. The outer layer, or tunica fibrosa, is skeletal in function, maintaining the rigidity of the spherical eyeball (Fig. 705). It is so tough that acids or alkalis, or even cooking, does not destroy it, and it is quite possible to pick a splinter off its exposed surface without serious injury.

The middle pigmented layer, or tunica vasculosa, provides a place for nutritive blood vessels, while the inside double retinal layer is sensory, containing among other elements the rods and cones that constitute the essential photoreceptive cells, to which all the other parts of the complicated visual apparatus are subsidiary.


The nervous elements of the eyeball arise from the wall of the embryonic brain. Early in development, before the medullary groove is closed into a tube, the diencephalon gives rise to two outpocketings (Fig. 706). Each of these soon differentiates into an enlarged distal portion, the optic vesicle, and a constricted proximal optic stalk which connects the vesicle with the diencephalic region. Next the terminal part of the vesicle invaginates to form a two-layered optic cup with its concavity toward the surface of the body. The inner layer of the cup eventually becomes the nervous portion of the retina, while the cup’s outer layer forms the pigment layer of the retina.

Outgrowth of the optic vesicles and formation of the lens in the chick

As the primitive double cup is forming, a portion of its ventral brim is involved in the invagination so that a deep notch, the choroid fissure, develops in this region (Fig. 707). This infolding also continues along the optic stalk to form a ventral groove which extends as far as the diencephalic wall.

The optic cup and lens in stereogram, showing choroid fissure

Eventually certain of the retinal neurons send out neurites which, after passing through the choroid fissure, run along the groove in the optic stalk and into the brain. These neurites form the optic nerve, while the optic stalk fades in importance and vanishes. Complete closing of the choroid fissure by the growth of surrounding cells about the optic nerve causes the latter to have the appearance of penetrating directly through the wall of the eyeball.

Development of the lens

Meanwhile, at the point where the optic cup comes nearest to the surface ectoderm, the latter gives rise to the lens. Lens formation is brought about by the invagination of the thickened ectoderm, which eventually pinches off a hollow vesicle that becomes entirely disconnected from the outside ectoderm. The cavity within the embryonic lens is gradually obliterated by the columnal growth of cells which make up its inner part and which eventually fill the space entirely (Fig. 708).

The retina and lens later become enveloped by mesenchymal tissues which give rise to the two outer layers of the eyeball, the tunica vasculosa and the tunica fibrosa.

Structure of the Human Eye

The outer tunica fibrosa of the human eye is made up of the sclera and the cornea (Fig. 705). The sclera is a tough, opaque layer of interwoven fibrous connective tissue. Commonly known as the “white of the eye,” it occupies five-sixths of the entire circumference, but is mostly out of sight within the orbit. It is pierced by the optic nerve and also by blood vessels.

The remaining one-sixth of the fibrous layer forms a transparent circular window, the cornea, over the front face of the eyeball and continuous with the white sclera. It is thinner in front (0.9 mm) than around the ring at the comeo-scleral margin (1.2 mm), where the muscles that rotate the eyeball in the socket are attached. The cornea is an important part of the focusing mechanism of the human eye, having a refractive power about two and one-half times that of the lens. Much of the importance of the lens, therefore, is associated with its use in accommodating the eye to objects at various distances from it.

The middle, vascular layer, or tunica vasculosa, which is in intimate contact with the layers next to it, both inside and out, is made up of three general parts continuous with one another, namely, the choroid, ciliary body, and iris.

The choroid, making up most of the middle layer except its anterior part, is rich in blood vessels and pigment cells. As a result of the heavy pigmentation of this part, light is absorbed instead of being reflected back and forth inside the eye. Toward the front of the eyeball the vascular layer becomes thickened into the ciliary body which is attached to the inside of the sclera, near the sclero-corneal junction, in the form of a ring. The portion of the vascular layer encircled by the ciliary ring and behind the cornea is the iris, which is not attached to the outer fibrous layer as are the choroid and ciliary body. In the center of the disc-like iris is an opening, the pupil, which always appears black in man because it is the only place through which light may enter into the dark camera-box, lined with nonreflecting tissue.

Beginning with a relatively thin edge next to the choroid, the ciliary body gradually increases in thickness until the maximum is reached near the margin of the iris (Fig. 705). The most conspicuous parts of this body are the ciliary processes and the ciliary muscles. The ciliary processes are 70 to 80 radiating ridges composed chiefly of connective tissue, blood vessels, and pigment cells (Fig. 709). The sides of these ridges and the valleys between them furnish the points of attachment for the suspensory ligament that stretches from them to the capsule surrounding the lens.

Anterior half of the eye seen from within

The ciliary muscles have to do with changing the shape of the lens so as to focus images sharply upon the retina. These muscles, smooth and involuntary, are innervated by fibers of the oculomotor nerve. Their action, however, is much more rapid than that of most smooth muscles. They include both circular and radial fibers, with the origin ends of the latter attached to the sclera near the junction of the sclera with the cornea. Thus contraction of the ciliary muscle fibers pulls the choroid forward and reduces the diameter of the ciliary body.

The iris serves as a delicate diaphragm surrounding the pupil. Within it are two antagonistic muscles, the sphincter pupillae and the dilatator pupillae, the contractions of which serve respectively to decrease and to increase the size of the pupil. In this manner the iris regulates the amount of light admitted into the eye. By covering the outer more curved edge of the lens it also cuts down or prevents the occurrence of spherical or chromatic aberration, which would blur or confuse the images cast upon the retina. When a person “goes to the light” to examine an object carefully, it is usually not because of a scarcity of light available, but because increased light by narrowing the pupil cuts out side lights and thus sharpens the vision, just as closing the diaphragm in a camera reduces spherical aberration and sharpens the image.

Pigment of various kinds, abundant in the iris, gives the characteristic “color of the eyes.” A thick pigment epithelium forms the posterior surface of this structure in all individuals. In the absence of other pigment, light reflected from this layer appears blue. Whenever any other eye color, such as gray or brown, is shown, it is due to the deposition of additional pigment in the outer parts of the iris where they conceal the blue derived from the pigment epithelium. The eyes of albinos appear pink because, in the complete absence of all iris pigments, the color of the blood shows through this region.

The two muscles of the iris, its pigment epithelium, and the innermost layers of the ciliary body do not properly belong to the tunica vasculosa, but are ectodermal derivatives of the embryonic optic cup. In the iris, the outer layer of the cup gives rise to the muscles, while the inner layer of the cup becomes the pigment epithelium.

The retina, the inner or sensory layer of the eyeball, is derived from the two layers of the optic cup and is therefore really a part of the brain. As mentioned earlier, the outer laver of the cup gives rise to the non-nervous pigment epithelium of the retina, which rests against the choroid. The inner layer of the cup forms the nervous portion of the retina, composed of three sets of neurons in series with one another, namely: photoreceptors, intermediate neurons, and ganglionic neurons (Fig. 710). The retina terminates abruptly near the periphery of the ciliary ring, although derivatives of the optic cup continue along the inner portion of the ciliary body and iris. The irregular anterior margin of the retina is known as the ora serrata (Fig. 709).

Diagram showing the three principal layers of the retina, with two rods and one cone in the outer layer

The photoreceptors, rods and cones, are next to the retinal pigment cells, from which fine cytoplasmic processes extend down between the outer segments of these receptor cells (Fig. 711). Clones are primarily concerned with color vision, while rods are chiefly useful in colorless vision at low light intensities. According to the most reliable estimates the retina includes 7,000,000 cones and 100,000,000 or more rods, all packed closely together like matches in a box, each one registering a single point of reflection from outside illumination. In the human eye, rods and cones differ more in size than in shape, yet their names are still descriptive of their form.

Diagram showing detail of structure of the retina

Vision is dependent upon photosensitive pigments found in the outer segments of rods and cones. The existence of such a substance, called visual purple or rhodopsin, has been known since 1876. It is abundant in rods and was at one time thought to occur in small quantities in cones. It is now believed that cones have a somewhat different pigment, visual violet or iodopsin. The production of both of these pigments is dependent upon a supply of vitamin A. It has long been known that improper diets may result in night blindness, or loss of sensitivity to dim light. Recently it has been shown that this condition results from the inability to synthesize visual purple in the absence of vitamin A. According to the photochemical theory of Hecht, the photosensitive pigments are decomposed by light with the formation of various products, including at least one which can initiate impulses in the receptor cells of the retina.

Neurites of the rod and cone cells synapse with bipolar intermediate neurons which in turn connect with large ganglionic neurons (Fig. 710). The neurites of these ganglionic neurons form the optic nerve, after running across the inside of the retina. The point where these fibers converge to leave the eyeball is known as the optic disc, or the blind spot. Since photoreceptive rods and cones are not present there, rays of light striking it are not seen. The blind spot is like a photographic film with a patch of its surface scratched off. The presence and extent of the blind spot in the reader’s eye may be easily demonstrated by reference to Figure 712.

The blind spot of the eye

In the stratum of the intermediate cells there are cross-connecting neurons, horizontal cells and amacrine cells, which join together various retinal neurons. There are also non-nervous supporting fibers of Muller extending through nearly the entire thickness of the retina.

It will be observed that the rods and cones, unlike other sensory receptors, point away from the source of stimulation. Light, upon reaching the retina, first encounters the non-receptive ganglionic and intermediate cells before coming to the receptive rods and cones. This inversion of the retina, characteristic of all vertebrates, may be explained, as Balfour suggested, by the embryonic history of the retina. According to Balfour the rods and cones when still in the surface ectoderm faced the outside world from whence the stimulus of light came (Fig. 713). With the invagination of the central nervous system they would therefore face the cavity of the brain. As the cavity between the two layers of the optic cup is really a continuation of the brain cavity, the receptors in the nervous portion of the retina would face this space. Thus the photoreceptors head toward the pigment layer of the retina, or away from the source of light for the completed eye.

Diagrammatic transverse section through the head of a hypothetical vertebrate embryo

In the posterior part of the retina in the direct line of the visual axis, is a small yellowish area, the macula lutea. In its center is a conical depression, the fovea centralis, where the nervous portion of the retina is reduced to a layer of receptor neurons, as a result of the spreading apart of the intermediate and ganglionic layers (Fig. 714). The floor of the fovea, occupied by several thousand long slender cones closely packed together, is the region of greatest visual acuity and color perception, due in part to the great number of photoreceptors, all cones, located here. It is also possible that the spreading of the inner layers of the retina permits a freer passage of light to the photoreceptors. Recently Walls has suggested that the sides of the fovea may act as refracting surfaces which so enlarge the image that more photoreceptors are involved. Thus a clearer image results from the increased resolving power in the foveal area. Whenever we look directly at any object, the image of that object falls on the fovea, while the light from neighboring objects, entering the eye at an angle, falls upon the rods and scattered cones that are distributed in the retinal areas outside the fovea.

Section through the human fovea

Light entering the pupil of the eye at an angle instead of directly in line with the visual axis falls upon portions of the retina outside of the fovea. It must, therefore, filter through and between the intervening, intermediate and ganglionic cells of the retina before it reaches the receptive rods and cones whose receptive ends extend between the pigmented cells of the outer retinal layer. The amount of light reaching the rods and cones under all these difficulties is regulated, as occasion demands, by the iris, with its adjustable pupillary aperture.

The lens is a cellular biconvex structure in which the component parts become transparent. In conjunction with the cornea it serves to refract the rays of light which enter the eye. Thus these rays are concentrated or focused upon the retina, producing a sharply defined replica of the external scene, much smaller in size than that which is presented to the eye. The anterior face, which is somewhat less convex than the posterior, is in contact with the pupillary margin of the iris. Surrounding the lens is a highly elastic capsule to which the zonular fibers (fibers of Zinn), which make up the suspensory ligaments, are attached near the equator of the lens. Thus these fibers extend from the ciliary body to the lens capsule.

In the resting eye, the short zonular fibers exert a tension, which somewhat flattens the lens, especially its anterior surface. In this state, images of distant objects are sharply focused on the retina. When the ciliary muscles contract, they reduce the diameter of the ciliary body thereby decreasing the tension on the zonular fibers. The elastic lens capsule is thereby allowed to modify the shape of the plastic lens so that it is more spherical, a change that is most pronounced in the anterior part of the lens (Fig. 715). By these delicately controlled changes in the shape of the lens, images of near objects are brought into sharp focus on the retina.

Diagram illustrating the process of accomodation in the human eye

There are three chambers, or cavities, in the eyeball (Fig. 705). The large space behind the lens, and nearly surrounded by the retina, is the vitreous chamber, containing the vitreous body, a transparent, jelly-like mass. Extending through this body, from the concavity into which the lens fits to the vicinity of the “blind spot,” is the hyaloid canal, through which an artery runs during embryonic development. Between the cornea and the lens are two spaces which communicate with one another through the pupil. In front of the iris is the anterior chamber, behind it the posterior chamber, both of which are filled with a watery lymph, the aqueous humor.

Communicating lymph spaces are also present in the eyeball. Around the outer margin of the anterior chamber is a loose spongy tissue containing many spaces of Fontana which communicate with the chamber. Nearby, at the corneo-scleral junction, in a circular channel, the canal of Schlemm, which communicates with neighboring scleral veins by a score or more of small branches (Fig. 715). Although there are no openings from the spaces of Fontana into this canal, it is believed that fluid from the anterior chamber entering the spaces of Fontana may pass into the canal and thence into the blood stream. Possibly excess fluid is removed from the eyeball along this pathway.

Accessory Parts

In addition to the eyeball itself there are certain accessory structures, namely, extrinsic muscles, glands, and shutter-like eyelids and eyelashes, that are parts of the complex vertebrate eye region. The extrinsic muscles, that enable the eyeball to face in various directions without moving the entire head, have already been considered. They are inserted on the outside of the eyeball far enough forward so that, when contracted, they do not pull directly against their point of attachment, but against the curving surface of the eyeball, thus minimizing the liability of tearing loose (Fig. 716).

Diagram to show that the eyeball muscles are attached so far forward that they pull against the side of the eyeball

Eyelids are transverse protective folds of the skin that close like shutters over the front face of the eye. The inner surface of the lids, the conjunctiva palpebrarum (Fig. 717), is a reddish mucous tissue continuous with the conjunctiva bulbi that extends over the eyeball, making a thin transparent skin on the face of the cornea through which light must pass on its way to the retina. The upper lid in man is larger and more movable than the lower, and the aperture between the lids, or the palpebral fissure, is rather wider than in most mammals, showing some of the white sclera as well as the circular transparent cornea.

Vertical section through the upper eyelid and outer part of the eyeball

The edges of the palpebral fissure are supplied with a double row of eyelashes, larger above than below, which guard the sensitive conjunctival surfaces against dust particles and similar unwelcome intrusions.

In the inner angle of the eye there is present a vertical fold of the conjunctiva bulbi, called the plica semilunaris, that in many vertebrates becomes extended into a movable third eyelid, or nictitating membrane, lying under the other two and closing over the eyeball from the inner angle outward (Fig. 718).

Front view of left eye showing plica semilunaris, caruncula lacrimalis, and lacrimal apparatus

At the medial end of each eyelid is a small opening, or punctum lacrimale, which leads into a slender canal, or lacrimal duct. The ducts from the two eyelids lead into a nasolacrimal canal, through which the excess of tears produced by the lacrimal gland is ordinarily drained into the nasal cavity. Medial to the plica and between the two lacrimal ducts is a small reddish elevation of the conjunctiva, the caruncula lacrimalis, which probably has to do with regulating the escape of tears through the nasolacrimal canal.

Between the outer skin and the inner conjunctiva palpebrae each eyelid is reinforced by a stiffening fibromuscular layer, the tarsal plate, containing numerous Meibomian glands, that pour out an oily secretion at the inner edge of the eyelid (Fig. 717). The oily film produced by these glands serves constantly to seal the inner margin of the moving lids to the surface of the eyeball and, when the eye is completely closed, to hold the margins of the two lids temporarily together.

The lacrimal glands open inside of the upper lids by several short ducts at the outer angle of the eye. The occasional occurrence in man of lacrimal glands opening inside of the lower lids is a reminder of the evolutionary journey they have made in order to arrive at their present position. In amphibians and reptiles these glands open inside of the lower lids.

Tears are a watery secretion from the lacrimal glands and are continually produced, flowing in the form of a thin film over the exposed surface of the eyeball, to drain eventually through the puncta lacrimalia into the nasolacrimal canal, at the inner angle of the eye. They serve not only to keep the conjunctival and nasal membranes moist, but also to clean the surface of the eyeball of foreign particles that may accidentally find lodgment there.

Weeping, which is accompanied by an overflow of tears, is peculiar to mankind. It is apparently a phyletically recent acquisition connected with certain emotional states that are not present in lower animals. Consequently, as a comparative anatomist would expect, a human baby cannot weep, in spite of the presence of lacrimal glands, until it is about six weeks old, although it may repeatedly demonstrate its ability to “cry.”

Median Eyes

An ancestral median eye, either the parietal or the pineal body arising from the dorsal diencephalic region of the brain, is laid down embryonically in nearly all vertebrates.

In lamprey eels both structures, parietal and pineal, are represented as a pair of organs with the parietal organ on the left and the pineal on the right side respectively, or more commonly with the parietal organ crowded around into a position anterior to that of the pineal body or beneath the latter (Fig. 719).

Median longitudinal section through median eyes of Petromyzon

The parietal organ degenerates in the cyclostomes, most fishes, and amphibians, but develops into a structure resembling a true optical organ with a retina and considerable structural complexity in certain lizards, particularly Sphenodon.

The extinct stegocephals had a conspicuous foramen through the top of the skull, like that in the skull of Sphenodon and certain anurans and lizards, which was probably for the accommodation of some sort of a median eye.

Only a trace of the parietal organ remains in birds, while among adult mammals it entirely disappears, in many cases in the embryo as well as in the adult.

Comparative Anatomy


Eyes are absent in amphioxus, but numerous photoreceptive cells occur in the nerve cord, thus rendering these primitive chordates sensitive to light as it penetrates through the semi-transparent tissues of the body. Each of these cells has a pigment cup associated with it (Fig. 720).

Photoreceptors of amphioxus


The eyes of cyclostomes are degenerate rather than primitive. The eyeball is for the most part small and buried under a thick skin. It lacks cornea, iris, lens, lids, and ciliary apparatus, although Geotria macrophthalmus, a fresh-water cyclostome from South America, having unusually large eyes as its name indicates, is an exception. There is no differentiation into rods and cones in the retina, for only elongated rods are present, indicating that rods, which are adapted for the reception of lights, shadows, and the movements of external objects rather than for the reception of colors, are phylogenetically older than cones.


The elasmobranch eye is provided with a large rounded cornea, that aids the spherical lens in focusing. Partial compensation for the difficulty in seeing both forward and backward with an eye placed on the side of a rigid neckless head, is furnished by the lens that projects through the pupil against the cornea, so that urays” of light coming at an extra wide angle are caught and concentrated upon the retina.

Eyelids, which are plainly simple folds of the skin, are present in many elasmobranchs but not in other fishes. The outer sclerotic layer is frequently reinforced by cartilage. The eyelids of the hammer-head shark, Sphyrna, are circular, which is perhaps the primitive form of all eyelids. It is easy to see how closure of the eyelids would be greatly facilitated by modification into upper and lower lids.

A tapetum lutidum, a layer composed of pigment and light-reflecting crystals, is found in the center of the choroid layer in some cartilaginous fishes. In the outer part of the choroid some teleosts have a silvery or greenish-golden layer called the argentea, that likewise acts as a reflector.

The eyes of teleosts vary greatly in size, being large in pelagic carnivorous fishes and certain deep-sea forms dwelling in regions of dim light, but small in bottom feeders. The eyeball is usually much flattened on the front face, so that the optical axis is shorter than the diameter through the equator. No eyelids or glandular devices to keep the eyeball moist are present or needed, but the unblinking eyes are in some measure shielded from the impact of water during locomotion by their lateral position.

Periophthalmus, the ambitious tropical climbing fish that crawls out of the water and lies in wait for flying insects upon the aerial roots of mangrove trees, has so far improved on the traditional fish eye as to anticipate the winking disappearing eye of the frog, which can be depressed into a protective orbit in the skull, or popped out at will to view the surroundings.

Certain deep-sea fishes possess so-called telescopic eyes, with elongated eyeballs, enormous spherical lenses, and rounded corneas directed upward and pointing towards the source of the dim light above them (Fig. 721).

Median section through the telescopic eye of a deep-sea fish, Agropelecus

Focusing is accomplished as in a camera by shifting the position of the lens with reference to the sensitive retina, while in higher forms the same result is brought about by changing the shape, but not the position, of the elastic lens. The eyes of fishes arc normally nearsighted, that is, they are accommodated to near objects when at rest, so that focusing by shifting the position of the lens is called for only when more distant objects are to be brought into clear vision. In either case it is not easy to see at a distance through the denser light-absorbing medium of water. The ciliary processes and muscles are small.

The movement of the lens within the eyeball of a teleost fish, but not of an elasmobranch, is probably aided by the processus falcijormis, a sickleshaped organ containing blood vessels, nerves, and muscles, and having an enlarged end, the campanula Halleri (Fig. 722). It extends from the choroid through the retina to the back of the lens. There is apparently no focusing device in the eyes of elasmobranchs.

Diagrammatic vertical section through the eye of a teleost, Salmo

Most fishes are practically color-blind, for a histological examination of the retina reveals a great scarcity, or an entire absence, of color-receiving cones. Some marine teleosts have a fovea.

Dark-adapted and light-adapted retinas of Ameiurus

In fishes there are marked changes in the retina when an animal is transferred from darkness into light, or the reverse happens (Fig. 723). In a dark-adapted animal the retinal pigment has receded from around the rod and cone endings and collected in the bases of the pigment cells. The cones have elongated somewhat, while the rods have shortened, with the result that these two types of cells are of approximately equal lengths. When such an animal is exposed to light the pigment migrates away from the bases of the pigment cells, the cones contract, and the rods elongate into the pigmented area. Many believe that by these changes the highly sensitive outer segments of the rods are made more accessible in dim light or darkness, yet protected from excessive exposure in strong light. In this manner the retinal changes supplement the activity of the iris in regulating the amount of light which reaches the photoreceptors. This theory does not explain the behavior of the cones and is open to other objections, yet various other theories which have been proposed are equally inadequate. Further data are needed before we can hope to understand the complex activities of the retinal elements.


In aquatic urodeles generally, the eyes are small, without lids, and often sunken into the skin. These animals apparently see with considerable difficulty, and then only nearby objects that are in motion. Even the anurans, those amphibian aristocrats which are adapted to life on land, possess eyes in many particulars simpler than those of fishes, for, although eyelids and eye glands are present, there is no tapetum or argentea associated with the choroid, and the retina possesses no fovea.

The small lens in the amphibian eye is no longer spherical but ovoid. It is located entirely behind the iris, giving space for the anterior chamber of the eye, while the cornea is so rounded out as to have the focusing value of a second lens, making the animal shortsighted in air when at rest, but farsighted when submerged under water, since the cornea and fluid in the anterior chamber have practically the same refractive index as that of the water outside, with the result that the cornea fails to focus the light as it passes through. As in fishes, the position rather than the shape of the somewhat inelastic lens is changed in the process of focusing, the ciliary muscles and processes being rudimentary.

The nictitating membrane in a frog’s eye, unlike that of higher vertebrates, is derived secondarily from the large lower eyelid and is lubricated by the Harderian glands, which open into the conjunctival sac beneath it along the lower eyelid. Phylogenetically these are the earliest form of lacrimal glands.

The iris of the eye in many amphibians is brilliantly colored, being frequently golden, or shot with yellow flecks.

“Which like the toad, ugly and venomous
Wears yet a precious jewel in his head.”

The pupil shows much variation in shape, ordinarily being round, but transversely oval in Rana and Bufo, vertically elliptical in Alytes, and somewhat triangular in Bombinator.

The rods in the retina are considerably more numerous than the cones. Migration of retinal pigment is extensive in these animals but the rods and cones do not change in length to the same extent as in fishes. In urodeles the length of the rods is not altered by changes in illumination.


The eyes of reptiles are always decidedly lateral in position, so that they have little if any common field of vision. A lizard, for example, may spookily roll one eye upward and the other backward or forward at the same time, thus seeing independently in two different directions at once. With the necessity for adaptation to life on land, the reptilian eye is safeguarded against increasing dangers by means of glands and well-developed eyelids. In addition to the small upper and the larger and more movable lower lids, there is a transparent nictitating membrane inside of the paired eyelids next to the eyeball. Harderian glands supply this third eyelid along the ventral border, while true lacrimal glands for the first time appear in the region of the outer angle of the palpebral fissure. Both Harderian and lacrimal glands find an outlet for the excess of their secretions in the nasolacrimal canal. That the differentiation of tear glands is a comparatively recent acquisition among reptiles is indicated by the fact that Sphenodon, which represents the most ancient of surviving reptilian types, is without tear glands.

Snakes, geckos, and certain limbless lizards, living in intimate contact with the ground, lack movable eyelids. Their staring unwinking eyes are protected by fixed transparent goggle-like windows of skin, shed in ecdysis. Such a transparent window in the closed lower eyelid allows light to enter and is at the same time a protection against blowing sand.

Considerable advance over the amphibian retina is shown in the relative number of cones as compared with rods. In fact diurnal lizards have no rods, while some diurnal turtles have very few. Crocodiles, which have a retina rich in rods, are well adapted for nocturnal vision. The diurnal lizards have better developed foveas than are found in any mammals including man. The iris of the reptilian eye is frequently highly colored, and in some turtles may even be of a different color in the male and female of the same species. The pupil is usually round, but it may be slit-like, either vertically or horizontally.

The reptilian lens is less convex than that of amphibians. Further, for the first time in vertebrate evolution, the lens is elastic so that focusing is effected by the improved method of changing the shape of the lens, rather than by shifting its position as in aquatic vertebrates. The ciliary processes are well developed and the large ciliary muscles are composed of striated fibers, a condition found in no other vertebrates except birds.


In every vertebrate class except birds there are certain species that are either blind or possess only rudimentary eyes, but the sense of sight is absolutely indispensable to these highly modified aviators in the struggle for existence. Although glands and muscles of the eyeball are better developed in mammals, both in complicated structure and efficient working, no other vertebrate eye excels that of birds, particularly birds of prey. The accommodation apparatus in the bird’s eye is especially rapid and effective. Chickens, with their eyes focused closely on the work of picking up small grains of food from the ground, become instantly aware of a predatory hawk, sailing like a vanishing speck high overhead. The same hawk can drop with fatal precision upon a tiny field mouse from a height that seems incredible to the possessor of human eyes.

The eye of birds is relatively very large, each eye occupying fully as much space in the skull as the entire brain in some cases (e.g., the owl). If proportionately as large as that of a bird, the eye of an average-sized man would weigh, according to Tiedeman, as much as five pounds.

Schematic section through the eyeball of a bird, Strix

The eyeball of most birds is not spherical, since it is constricted in the ciliary zone by a sclerotic ring in such a way that the corneal region becomes projecting and very conical and the posterior part larger and more flattened (Fig. 724).

In most birds there is a preponderance of retinal cones. The eye of some birds possesses two foveal depressions (Fig. 725), although the majority of these animals have only one.

Section through the head of a swallow, Tachycineta, showing the two foveae of each eye

The ciliary processes are large and numerous, frequently numbering more than 100. As in reptiles the ciliary muscles are striated. Both of these features are probably associated with the excellent power of accommodation which birds possess.

Projecting into the vitreous body of the eye is an erectile fanlike organ of several folds, the pecten, that bears a superficial resemblance to the processus falciformis of the teleost eye, but is not homologous with it. The pecten arises from the point of exit of the optic nerve while the falciform process is a choroid outgrowth. The initial stages of the pecten appear among certain reptiles (Fig. 726), while embryonic traces of it still persist in the mammalian eye. Its function is not known with certainty but it seems probable that it plays a nutritive role and may also regulate the pressure of the fluids within the eye.

Diagrammatic horizontal section through the eye of Chameleon, showing rudiment of pecten, and sclerotic cartilages

Since the iris of the bird’s eye contains striated muscle fibers, it is capable of more intensive contraction than is possible in the case of any other vertebrate eye, the iris of which is fitted only with smooth muscles. The color of the iris varies considerably both with age and sex, and is characteristic for different species. In many parrots, for example, it is white, while in cormorants it is green, in swifts blue, and in Vireo olivaceous red. The whole expression of the otherwise immobile face of the bird is centered in the lively iris, with its invariably round pupil.

The nictitating membrane is well developed and is particularly useful to flying birds, being shut like transparent goggles over the face of the eyeball, thus preventing a flow of blurring tears during flight, which normally would be incited by the stimulating contact of air against the rapidly moving cornea.

Harderian glands are usually large in the bird’s eye, while the smaller lacrimal glands occupy the same position as in reptiles, at the outer angle of the palpebral fissure.

Owls have binocular vision, that is, both eyes are trained upon the same field simultaneously. In all birds the act of directing the eyes towards the source of optical stimulation is greatly facilitated by the fact that the extremely mobile head is mounted upon a particularly flexible neck.


The mammalian eyeball is nearly spherical. The upper eyelids, unlike those of other vertebrates, are larger and more movable than the lower eyelids, while two rows of eyelashes, lubricated by ciliary glands, are added to the equipment. The nictitating membrane is poorly developed. The eyelids of many mammals, such as mice, rabbits, and cats, are sealed at birth, opening only after several days.

For the most part the tear glands move around to an externodorsal position, although Harderian glands along the ventral margin of the eyelids are present in whales and such semiaquatic forms as the otter, hippopotamus, and seal.

There is no bony sclerotic ring, but a stiffening cartilage is present in the sclerotic wall of the monotreme, Echidna. The sclera of whales is greatly thickened and resistant, possibly to withstand pressure from the surrounding water.

The choroid of the mammalian eye is very rich in blood vessels, whose turgor may in a measure compensate for the lack of skeletal stiffening in the walls of the eyeball. In many mammals, particularly ungulates, cetaceans, and carnivores, there is a light-reflecting tapetum lucidum within the choroid layer, but in higher vertebrates including man it is lacking. The eyes of a cat show this reflecting device particularly well at night when lights from an automobile flash into them so that the eyes seem to glow like balls of fire.

The color of the iris varies among mammals generally as it does in man. Thus, there are blue-eyed goats, yellow-eyed cats, and brown-eyed dogs. The pigment determining human eye-color does not reach its final shade until five or six years of age. Aristotle, whose mind was occupied with many things, took time to look into babies’ eyes and to note that they are always blue at first. Smooth muscles are present in the iris, as well as in the ciliary apparatus.

In certain ungulates, such as goats, gazelles, camels, and coneys, the edge of the iris shows a peculiar modification, the umbraculum (Fig. 727), consisting of pigmented, projecting, granular fringes which permit a lessened amount of light to enter through its ragged edges even when the pupil is wide open. Many heavy-headed ungulates have a pupil in form of a transversely oval aperture that enables the animal to sweep the horizon without swinging the head. In most cases, however, the pupil is round, although cats have a vertical slit-like opening in the iris, adapted to nocturnal explorations, while the seal has a curious pear-shaped pupil with the wide end next the nose.

Umbraculum in the iris of a llamas eye

The lens in the mammalian eye is still less convex than that of reptiles or birds. It is more spherical in water forms and is relatively largest in nocturnal and crepuscular animals, such as bats, cats, and mice. Nocturnal animals are further characterized by the absence or paucity of color-receiving cones in the retina.

Detwiler says that “in mammals and man it is very questionable whether pigment migration has been demonstrated definitely in any instance.” He also doubts the occurrence of any contraction and elongation of rods and cones.

Stereoscopic perception of distance through triangulation resulting from binocular vision is especially essential to animals of prey. Among mammals it is present in some carnivora and in higher primates. It is also found in a few sharks and rays, toads, and carnivorous birds (e.g., hawks, owls, and some gulls). Lower vertebrates with lateral eyes and two fields of vision sacrifice the more accurate judgment of distance for a wider field of vision.

It should be remembered that a little more than half of the energy of extra-terrestrial light is appreciable to the photoreceptors of the human eye. About 43 per cent of the spectrum at the infrared end, and 5 per cent at the ultraviolet end, is “out of sight.” The 52 per cent remaining, that is, between ether waves of 397 millionths of a millimeter in length at the violet end and 760 millionths of a millimeter in length at the red end, represents the entire output of photic energy for which the human retina is adequate. Within these extremely minute limits lies the whole visual splendor and variety of our color world.