In this article we will discuss about the development of eyes in all vertebrates.

The Optic Cup:

The optic vesicles push outward until they reach the epidermis, displacing the intervening mesenchyme, so that they come into direct contact with the inner surface of the epider­mis. Next, the external surface of the optic vesicle flattens out and invaginates inward, so that the vesicle is transformed into a double-walled, cup-like structure—the optic cup.

The invaginated wall of the optic cup is much thicker than the remaining external wall. The first develops into the sensory retina of the eye, and the second develops into the pigment coat of the retina (tapetum nigrum). The rim of the eye cup later becomes the edge of the pupil.

The cavity of the optic cup is the future posterior chamber of the eye, which is filled by the vitreous body. The opening of the eye cup is very large at first, but later the rims of the cup bend inward and converge, so that the opening of the pupil is constricted and reduced to its final relative dimensions.


The rim of the optic cup surrounding the pupil becomes the iris. The constriction of the pupil does not take place equally all around the circumference of the eye, but is delayed on the ventral edge of the eye cup. A groove, the choroid fissure, remains here, cutting through the otherwise approximately circular edge of the eyecup and reaching inward as far as the optic stalk.

This fissure serves for the entry into the posterior chamber of the eye of a blood vessel and of mesenchymal cells which are found later in the vitreous body. The fissure normally closes during embryonic life; however, it may persist, and as a result a gap is left on the ventral edge of the pupil. This deformity, which in uncomplicated cases does not substantially affect vision, is known as the coloboma of the iris. In humans it is caused by hereditary factors.

The size of the optic vesicles relative to the rest of the prosencephalon may vary considerably in different vertebrates. Even in one order among the vertebrates, such as the frogs, it was found by direct measurement that the mass of cells used for the forma­tion of the eye vesicles may, in different species, range from 10 per cent to 50 per cent of the volume of the prosencephalon.

As a general rule, the eye rudi­ments are large in bony fishes, reptiles, and birds, smaller in amphibians, and relatively very small in mammalian embryos. The determination of the optic cup is due to the action of the underlying roof of the archenteron on a part of the neural plate.


In the late neural plate stage the determination appears to be irrevocable, and parts of the optic vesicle cannot differentiate in any other way than by developing into retina, iris, or pig­mented epithelium of the eye. The eye rudiment may be excised and transplanted into any region of the embryo, and its development will continue more or less normally, producing a heterotopic eye.

A transplanted eye rudiment, although no longer capable of being transformed into other tissues, cannot develop normally unless it is surrounded by mesenchyme. Without mesenchyme in its environment, the differentiation of the optic rudiment remains extremely poor.

The determination of the eye as a whole does not mean that all the parts of the eye rudiment are also determined. The determination of the parts of the eye occurs much later. The eye rudiment in the neural plate stage and in the optic vesicle stage may be split in two, and each half will then develop into a complete eye.

This can best be demonstrated if a part of the eye rudiment is transplanted. It is found then that the remaining and the transplanted parts each develop into a small eye. Experiments have also been performed on the eye rudiment at the stage when the optic vesicle is being transformed into an optic cup and the two parts of the cup, the future sensory retina and the future pigment coat, become morphologically distinguishable.


If a piece of the presumptive pigmented epithelium is excised and transplanted into the vicinity of a normal eye of another embryo, it will develop into a complete eye, consisting of both a pigment coat and a sensory retina. However, a pre­requisite for this is that the piece not be too small. Very small fragments of the eye rudiment usually develop into pigmented epithelium only.

The ability of a part of the organ rudiment to develop as a whole is analogous to a similar ability of parts of the egg in the early stages of cleavage in some animals. The process observed in both cases is known as self-regulation. It is found in many organ rudiments; in fact, it can be considered as a common property of organ rudiments. The ability to self-regulate presupposes that the parts of the rudiment (or of the egg in the early cleavage stages) are not determined.

In the case of the eye cup, the absence of determination of parts can also be demonstrated in another way. If a suitable inductor is applied to the outer surface of the optic cup—that is, to the surface which normally differentiates into the pigmented epithelium—it may be induced to develop as sensory retina, so that the eye has two retinas, a normal one and an additional one.

The latter is never as large as the normal one. A suitable inductor for this purpose is the sensory epithelium of an ear. The experiment therefore consists in the transplantation of an optic vesicle into the immediate vicinity of the ear vesicle to insure the close contact of the epithelia of the two organ rudiments.

With advancing development it is apparent that the rim of the optic cup becomes increasingly different from the deeper-lying parts. The constriction of the pupil takes place at the expense of a considerable thinning out of the wall of the rim of the optic cup. The thinned-out portion becomes the iris of the eye, while the remaining part, which stays considerably thicker, gives rise to the retina proper.

In the iris, large amounts of pigment are deposited in the outer epithelial layer. This layer is actually a part of the pigment coat of the eye cup. In addition to cells carrying pigment, this layer also gives rise to the smooth muscle fibers of the sphincter and dilator muscles of the iris. In the sensory retina, the cells start differentiating into the sensory and nerve (or ganglion) cells.

The first trace of this differentiation is seen in the arrangement of the nuclei of the cells in several layers. The nuclei situated in the layer nearest to the pigment coat belong to the future rod and cone cells. The rudiments of the rods and cones appear as cytoplasmic processes on the inner ends of these cells.

The remaining nuclei, arranged in two or more layers nearer to the cavity of the eye cup, give rise to the various types of intermediate and ganglion cells of the retina. Nerve processes arising from the ganglion cells of the retina grow out toward the brain, and the path which they take is along the stalk of the eye cup.

In this way the stalk of the optic cup becomes transformed into the optic nerve. On reaching the floor of the diencephalon, in all vertebrates except mammals, the nerve fibers do not enter the same side of the brain but cross over to the opposite side and there penetrate into the wall of the diencephalon and the mesencephalon.


Where the nerve fibers of the two eyes cross and bypass each other on their way to the contralateral parts of the brain arises the optic chiasma. In mammals, the crossing of the fibers of the optic nerve in the chiasma is not complete; a small proportion of the fibers enter the brain on the same side as the eye in which they originated.

The optic cup, even when fully differentiated, is not yet a complete eye. Certain accessory structures have to be added to it to make the eye fully functional. The most important of these structures is the lens, which serves for the refraction of light rays entering the eye.

The lens is not developed from the optic cup, but from the epidermal epithelium with which the optic vesicle comes in contact. As the outer wall of the optic vesicle begins to invaginate to become the retinal layer of the optic cup, a thickening appears in the epithelium which is in contact with the invaginating part of the optic cup. This thickening is the rudiment of the lens.

The Lens of the Eye:

The way in which the lens rudiment is separated from the remainder of the epidermis varies in different classes of vertebrates. In birds and mammals, the epidermal thickening folds in to produce a pocket which is for a short time open to the outside and later becomes a vesicle lying in the opening of the iris (in the pupil).

In amphibians and bony fishes, the thickening in the formation of which only the inner layer of epidermis takes part is nipped off from the epidermis as a solid mass, but later the cells of this mass rearrange themselves into a vesicle. In both cases the vesicle must undergo further differentiation before the lens can function as a refracting body.

This happens in such a way that the cells on the inner side of the lens vesicle elongate, become columnar at first, and later are transformed into long fibers. During this transformation, the nuclei of the cells degenerate, and the cytoplasm becomes hard and transparent. The fibers are arranged in the lens in a very orderly way, forming the spherical or ellipsoid refracting body of the lens.

Part of the lens epithelium remains unchanged and covers the sphere of fibers distally. The junction between the un­changed lens epithelium and the mass of fibers is the growth point of the lens. Here the epithelial cells are continuously transformed into fibers, so that the refracting body grows by apposition of new fibers.

Between the development of the optic cup and the development of the lens there exists, in most of the vertebrates studied experimentally, a direct causal relationship; the development of the lens is dependent on an induction from the optic vesicle. As the optic cup touches the epidermis, it gives off a stimulus of some kind, which causes the epidermal cells to develop into the lens rudiments.

Any epidermal cells are able to react to the induction of the optic vesicle, and without this induction the lens does not develop at all, or at least the development is defective. The dependence of lens development on the action of the optic cup can be shown by several types of experi­ments.

One type of experiment is to remove (excise) the eye rudiment before it can reach the epidermis. Such an operation usually leads to the absence of the lens. Another pertinent experiment is to remove the epidermis which normally would have formed the lens and to replace it with a piece of epidermis taken from another part of the body, from the head or even from the belly.

In this experiment it was observed that the epidermis, if in contact with the optic vesicle, develops into a lens. A third type of experiment is to transplant the optic vesicle, without the epidermis normally covering it, under the epidermis in an abnormal position. In this case the local epidermis may be caused to develop a lens.

For a long time it was believed that an intimate contact between the optic vesicle and the epidermis is indispensable for lens induction. A thin layer of cellophane inserted between the optic vesicle and the epidermis in a chick embryo completely stopped the inducing action of the optic vesicle. Insertion of a porous membrane, which presumably allowed for the passage of macromolecules, between the eye vesicle and the epidermis in a frog embryo also precluded lens induction.

However, experiments have been reported in which a partial screening of the epidermis from the eye vesicle in a chick embryo by a thin slice of agar did not prevent the complete development of the lens. The inducing agent could thus either get through or get around the agar. This suggests that the inducing agent is a chemical substance.

Apart from this, not much is known as to the nature of the stimulus responsible for the induction of the lens. It has not been possible to extract a lens-inducing substance from the optic vesicle. On the other hand, some of the “abnormal” inductors of neural plates are known to induce “free” lenses, that is, lenses without an eye. The thymus of the guinea pig seems to be especially suitable for this purpose. That the stimulus in this case is exactly the same as that exercised by the optic cup remains to be proved.

There is one peculiar similarity between induction of the neural plate and induction of the lens, concerning the distribution of the cytoplasmic ribonucleic acid in the components participating in the process. The eye vesicle at the time when it comes in contact with the presumptive lens epidermis contains a large amount of ribonucleic acid.

At the same time, the presumptive lens epidermis has little ribonucleic acid, and in this respect it is no different from the rest of the epidermis. After the contact is established, the ribonucleic acid in the cells of the eye vesicle is found to be concentrated near the outer margin of the cells, that is, where the cells touch the epidermis.

Large amounts of ribonucleic acid now appear also in the presumptive lens cells, at first only at their proximal ends, where they are in contact with the eye vesicle, but later in the outer parts of the epidermal cells. In subsequent stages, the ribonucleic acid content in the retinal cells decreases, while it continues to increase in the cells of the lens rudiment. There is no proof, however, that the ribonucleic acid actually passes from the eye cup cells into the lens rudiment cells.

In another experiment, performed on Xenopus embryos, the proteins of the eye cup cells were marked by supplying them with radioactively tagged phenylalanine (containing 14C). Subsequently, radioactive material was found to have passed into lens rudiments induced by the marked eye cups which were transplanted into normal embryos.

Although this experiment is open to the same criticisms as the experiments with radioactively marked inductors of the neural plate, the results are compatible with the assumption that a chemical substance passes from the eye cup into the reacting epidermis during lens induction and that this substance is a protein. There would thus be a similarity in this respect between the mechanism of neural plate induction and that of lens induction.

The relation between lens development and optic cup development is complicated considerably by the fact that in a few amphibians—Rana esculenta, Xenopus laevis, Rhacophorus schlegelii, and others to a lesser extent — the lens shows a certain degree of independent development (“self-differentiation”) even in the absence of the optic cup, that is, when the optic cup has been previously removed.

The degree of independent development varies from a tiny nodule of epidermal cells to a rather typical lens with fiber differentiation. The development is never completely normal; as lenses without eyes undergo a far-reaching degeneration once the initial stages of development have occurred. Nevertheless, the independent development of lenses shows that the eye cup is not the only part which may be involved in lens development.

Experiments in which free lenses have been induced by abnormal inductors (thymus) suggest that the “self-differentiation” of lenses, when it occurs, is due to some other influence on the differentiation of the epidermal cells and that this influence is responsible, in some species, for “indepen­dent” lens development (i.e., lens development which is independent of the eye cup).

What these influences might be may be concluded from experiments in which belly ectoderm was not capable of developing a lens when transplanted just before the formation of the optic vesicle but was able to react if transplanted in the neurula stage. It follows that to be able to react the epidermis must be in position some time before contact with the optic vesicle is established.

During this time, the head mesoderm lies immediately underneath the presumptive lens epidermis, and the influence of the head mesoderm was concluded to be the factor which prepares the epidermis for the subsequent induction emanating from the optic vesicle. Presum­ably, in some species, the “preparation” goes so far that the development of the lens may start even if the eye vesicle is not present.

Our conclusion may perhaps be stated in another way – we may say that the development of the lens is a result of induction emanating from two sources, the head mesoderm and the eye vesicle. Normally, both are necessary for successful lens de­velopment.

There is some evidence, however, that the relative importance of the two inductors may be changed by environmental factors. Keeping the developing embryos at a low temperature seems to favor induction by the head mesoderm, so that induction by the eye vesicle becomes unnecessary.

The differentiation of the lens cells into lens fibers in normal development is caused by the same induction as that involved in the development of the lens as a whole. It has been proved that contact with the presumptive retinal layer of the eye can induce lens fiber differentiation. The epithelial cells of the lens rudiment are all capable of fiber
differentiation if they are exposed to a suitable stimulus.

Contact with retinal tissue or with the sensory epithelium of an ear vesicle (compare what has been said on the induction of the retina) may cause the formation of an additional mass of lens fibers, so that under experimental conditions lenses may develop having two independent masses of fibers.

Accessory Structures of the Eye:

The other accessory structures of the eye which are present in all vertebrates having functional eyes are the choroid coat, the sclera, and the cornea. The choroid coat and the sclera develop from mesenchyme accumulating around the eyeball, in the way mesenchymal cells accumulate around many organs giving rise to their connective tissue capsules. 

In the case of the eye, the interior layer of mesenchymal cells gives rise to a network of blood vessels surround­ing the pigment epithelium. The outer layer of mesenchyme forms a fibrous capsule around the eye, which serves for its protection and for the insertion of the eye muscles. The capsule may either remain fibrous or develop cartilage or even bone (in reptiles and birds).

The cornea originates, in part, from mesenchyme, but the epidermal epithelium also has an essential role in its formation. The connective tissue part of the cornea is continuous with the sclera, while the corneal epithelium is continuous with the skin epidermis or with the epithelium of the eyelids, where such are present. Both the epithelium and the connective tissue of the cornea become transparent, so that the light rays may enter the eye.

The development of the cornea can easily be traced in living amphibian embryos. Initially the epidermis covering the eye is pigmented, for the epidermal cells contain granules of pigment derived from the egg. In the cells of the presumptive cornea these pigment granules become dissolved, and later, when the chromatophores develop in the connective tissue of the skin, the cornea remains free of them.

Transformation of the skin into the cornea is caused by an induction, the source of which is the eyeball. This can be proved by transplanting the eyeball heterotopically or by replacing the normal cornea by skin from another part of the embryo. The stimulus can be given off by both the eye cup and the lens. If the lens alone is transplanted, the epidermis over it loses its pigment and differentiates as cornea. If the eye is removed, the cornea does not develop at all.

Induction of the cornea presents an interesting peculiarity as compared with neural plate and lens induction in amphibians. The competence to differentiate as cornea is found in the skin not only during a short period of embryonic development but for a long time, long after the normal differentiation of the cornea has taken place. Also, the eyeball retains its inductive ability for a long time, probably permanently.

Moreover, the persistence of the cornea is dependent on the continuous presence and influence of the eyeball. If, in a late amphibian larva or an adult, the eye is removed, the cornea soon loses its transparency, is invaded by chromatophores, and becomes more or less normal skin. On the other hand, a fully differentiated piece of skin will lose its chromatophores and become transparent cornea if it is transplanted over the eye.

In the development of the eye, induction takes place repeatedly, and some parts after having been induced themselves become a source of inducing stimuli.

A whole chain of inductors can thus be noted:

i. The roof of the archenteron induces the neural plate and therefore also the eye cup rudiment which is part of the neural plate.

ii. The eye cup rudiment, becoming the optic vesicle, induces the lens (acting together with head mesoderm).

iii. The lens induces the cornea (acting together with the optic cup).

Parts developing as a result of induction, and inducing in their turn, may be called secondary, tertiary, etc., inductors, or organizers of the second grade, third grade, and so forth.