When the epidermis is first segregated from the other parts of the ectoderm (neural plate, neural crest) during the process of neurulation, it is still a very complex rudiment. Most of it becomes the epidermis of the skin, but in addition, a number of other structures are derived from it. Some of these are – the lens, the cornea, and the cranial ganglia.
The epidermis itself gives rise to quite a large number of special differentiations, such as various unicellular and multicellular skin glands, including the sweat glands and the sebaceous glands, the hairs, feathers and scales, and various other special structures derived from these. The development of some of these parts involves histogenesis rather than organogenesis.
The Ciliary Cover:
One special differentiation of the epidermis occurs at a very early stage of development, and therefore deserves to be discussed here- it is the ciliary cover of the epidermis in amphibian embryos. Already in the late neurula stage some cells of the epidermis develop cilia. These are particularly abundant in anuran (frog and toad) embryos, but are also present in urodeles.
The cilia in amphibian embryos have been the subject of study with the electron microscope. At the same stages when cilia develop, other, non-ciliated, cells of the epidermis become secretory. These are the earliest functional differentiations of cells in any vertebrate embryo.
The origin and development of the ciliary cells is quite remarkable. The ectoderm of amphibian embryos at the end of gastrulation is composed of several layers of cells, of which the external layer contains distally abundant pigment granules. The cells of the deeper layers also possess pigment granules, but in lesser quantities.
At the time of the formation and closure of the neural plate, a very large proportion of the ectoderm of the embryo is removed from the surface—perhaps as much as one third of the total mass of the ectoderm. The rest of the ectoderm, now the epidermis, has to cover the whole surface of the embryo, and this surface increases even further when the embryo starts elongating in the post-neurulation stages.
The expansion of the epithelial layer occurs without appreciable increase in volume. It is thus another case of epiboly, as is the case with the ectoderm during gastrulation. The mechanism of expansion is, however, somewhat different. In both cases the expansion involves a re-arrangement of cells into a lesser number of layers, but in gastrulation the surface layer of ectodermal cells remains continuous, at least in Xenopus, and is not reinforced by egression of cells from the deeper layers.
The expansion of the epidermis during neurulation, on the other hand, involves the shifting of some cells from the interior to the surface, as well as a re-arrangement of the cells in the inner layer, so that the epidermis becomes a two-layered sheet of cells.
It is the cells which shift from the interior onto the surface that become the ciliary cells. This is evident because even before these cells reach the surface the basal granules of the cilia start forming in their cytoplasm.
At first the future ciliary cells extend to the surface only by attenuated ends, but later they expand distally, and their external surface becomes as large as that of the cells originally located on the exterior. The latter by this time are producing secretions in the form of mucous granules, close to their outer surfaces. The granules break out to the outside.
As the cells rising from the interior are less pigmented than the original surface cells, they are seen, under low magnification, as light colored patches on a dark background. Ciliary cells are scattered over the entire surface of the embryo, but no cilia are formed on the neural plate, which undergoes a completely different type of differentiation from that of the epidermis.
The basal granules of the cilia first appear in small clusters in the perinuclear cytoplasm of the cells. When the ciliary cells get access to the exterior, the basal granules move to the external surface of the cell and give rise to the shafts of the cilia as well as their “roots”.
In the late neurula stages the cilia, by their beating, cause the embryo to rotate within the egg membranes. The direction of the beat of the cilia is from the anterior end of the head all along the body toward the posterior end, so that when the embryos emerge from the egg membranes they are able to glide over the substrate, with the cilia serving as a locomotory mechanism before the muscle system starts functioning.
After the tadpoles begin swimming by lateral inflexions of the tail, the cilia persist for a while. At this stage the cilia Greate a current of water along the length of the body. The ciliary cover reaches its maximum development at the early tadpole stage. When the tadpole is fully formed (about stage IV) the cilia disappear.
The ciliary cells undergo degeneration, the cilia are lost and the cytoplasm becomes dehydrated. The differentiation of the ciliary cells is thus not only the first differentiation in the tadpole, but is also a terminal one – the cells die after their function is completed.
The secretory cells of the late neurula and early tadpole, on the other hand, remain as the sole outward cover of the skin of the tadpole after the ciliary cells are gone; eventually they are the first cells to be keratinized and lost at the time of metamorphosis of the tadpole into a frog.
The Organization of the Epidermis:
After the full expansion of the epidermis in the post-neurula stages, it consists in amphibians of two layers of cells – the outer covering layer or periderm and the inner so-called sensory layer. The latter name is used not because the layer as such has nervous functions but because some sensory organs are derived from parts of this layer.
In birds and mammals, the epidermis of early embryos also consists of two layers of cells, but the cells of the outer layer, the peridermis, are flattened and are shed eventually. In all vertebrates the inner layer proliferates, producing the stratified epithelium, the epidermis of the skin.
The innermost cells of the epidermis, the ones adjoining the basal membrane, become the generative (or Malpighian) layer of the epidermis. The skin is then composed of the epidermis and the layer of mesenchyme, partly derived from the neural crest and partly from the dermatomes, which gives rise to the dermis.
Many structures derived from the epidermis make their first appearance in the form of plate-shaped thickenings of the epidermal epithelium. Such thickenings are called placodes. That the ganglia of the cranial nerves are derived in part from placodes has been stated. When the lens rudiment first appears as a thickening of the epidermis, it bears a great similarity to the other placodes.
A pair of placodes, appearing in front of the anterior end of the neural plate and probably deriving their material from the neural folds itself, develops into the olfactory sacs. A placode appearing against the side of the hindbrain invaginates and produces a vesicle which is eventually separated from the epidermis.
This structure is the ear vesicle, the rudiment of the internal ear (the ear labyrinth), and the placode from which the ear vesicle is developed is the auditory placode. Parts of the epidermis adjoining the auditory placode also become thickened, and from these placodes develop the lateral line sense organs.
In the aquatic vertebrates (fishes, aquatic larvae of amphibians), the lateral line organs are distributed over the head (in several rows), and a row of the same organs stretches backward along the side of the entire body and tail. Wherever the lateral line organs are found, the cells of which they consist come from the placodes of the ear region.
In amphibians, the backward migration of the lateral line organ cells was demonstrated in a grafting experiment. Two embryos belonging to different species of frogs were cut transversely in halves, and the anterior half of a darkly pigmented species (Rana sylvatica) was grafted onto the posterior half of a lightly pigmented species. (Rana palustris). The darkly pigmented cells of the lateral line rudiment of the anterior half could then be observed to migrate into the posterior half and along the trunk and tail.
In the case of lateral line sense organs, the pathway of the migrating cells is dependent on the surroundings through which they migrate. If the anterior half of a frog embryo is transplanted onto another embryo whose own lateral line organ rudiment was removed previously, the lateral line cells of the anterior half grow out into the second embryo, and once they have reached the path of migration normally taken by the lateral line cells, they start moving along this path, even though their new direction is at an angle to the one they had been following before.
This experiment shows that the path of migration is determined by a factor lying outside the migrating cells themselves. It has been shown that the path of migration in this case is determined by the mesoderm.
The development of the placodes is probably always dependent on a stimulus from the tissues situated under the epidermis. This has been proved in some instances, as in the case of the lens of the eye. The ear is also dependent in its differentiation, a fact which can be deduced from experiments in which ear vesicles have been induced heterotopically as the result of the transplantation of various inductors.
The ear vesicle is among the structures that are often induced when the primary organizer is transplanted, but it can also be induced in experiments involving transplantation of adult tissues and of parts of the neural plate and neural tube. The latter experiments suggest that in normal development the ear vesicle is induced by the medulla oblongata.
This, however, cannot be the sole inductor, as the ear vesicles may develop in their normal position after the medulla oblongata is removed at an early stage, either alone or together with most of the central nervous system rudiment. It is concluded therefore that the development of the ear vesicle is dependent on multiple induction, and that similar stimuli are emitted both from the medulla oblongata and from the mesoderm developing from the roof of the archenteron.
The induction of the ear vesicle probably proceeds in two stages-first, the presumptive ear ectoderm is acted upon by the underlying mesodermal mantle in the late gastrula and early neurula stages, and later, the determination is finally stabilized by the influence of the medulla which, as a result of the closure of neural folds, comes into close contact with the epidermis in the ear region.
A similar duplication in the sources of induction has been postulated for the nose rudiment—namely, an earlier induction by mesoderm and a later induction by the forebrain. It has been claimed, for instance, that the anterior portion of the neural plate if transplanted under the epidermis on the flank may induce a nose rudiment locally.
In more careful experiments it was found, however, that nose rudiments may develop in the absence of brain tissues, and thus the alleged inductions must be due to the nose rudiment material being grafted together with the presumptive forebrain. The two rudiments are thus induced simultaneously by the underlying roof of the archenteron and initially lie very close to each other.
The Olfactory Organ:
Although the nose rudiments seem to be determined at a very early stage (late gastrula), they first become discernible morphologically after the closure of the neural tube, in the form of two thickenings of the epidermis, the olfactory placodes, just anterolateral to the hemispheres of the telencephalon.
The central part of each placode becomes invaginated, and the olfactory placode thus becomes converted into an olfactory sac which is open to the exterior by the external naris.
Parts of the wall of the olfactory sac, especially the dorsal and lateral wall, are differentiated as olfactory epithelium. The primary sensory cells of the olfactory epithelium develop on their proximal ends nerve processes (axons) which converge to form the olfactory nerve. The olfactory nerve grows into the adjacent wall of the telencephalon, bridging the narrow gap between the olfactory organ and the brain.
In most fishes the olfactory organ retains essentially the same structure in the adult state, but in the group of Choanichthyes among the fishes and in the terrestrial vertebrates, the structure of the olfactory organ is further complicated by the internal nares, and in the mammals also by the nasolacrimal duct.
The internal nares or primitive choanae arise by a perforation of the nose sac cavity into the oral cavity. The actual perforation is preceded by the formation, from part of the ventral wall of the nose sac, of an elongated tube stretching backward and downward toward the oral cavity.
This tube is the nasal canal, and the epithelium lining it becomes thin and is thus fairly sharply segregated from the thicker epithelium giving rise to the sensory part. With the elongation of the nasal canal the sensory part appears to be a dorsolateral growth of this canal, although it is actually the older portion of the olfactory organ.
The internal (posterior) end of the nasal canal fuses eventually with the epithelium lining the oral cavity, and the intervening membrane becomes perforated as internal nares. Jacobson’s organ, where it is present, is another section of the olfactory sac retaining sensory function. It develops from a medioventral part of the sac.
The auditory placode, from which the internal ear is developed, initially shows considerable similarity to the olfactory placode and is also converted by invagination into a sac-like structure, but in the early stages there already are important differences.
In the amniotes the whole epidermal layer is involved in the formation of the auditory placode, and it later invaginates to form a sac which is, at least temporarily, open to the exterior. In the frogs, however, the auditory placode is formed by the thickening of the interior “sensory” layer of the epidermis, while the external covering layer is not involved at all.
As a result, when the placode invaginates, there is no opening or pit on the surface of the skin. In both cases, however, the opening of the sac becomes constricted and closed, so that eventually the rudiment of the ear takes the form of a completely closed vesicle, the ear vesicle. In the bony fishes the auditory organ is formed not by invagination but as a solid mass of cells on the inner surface of the epidermis and is hollowed out secondarily.
The ear vesicle is the rudiment of the most essential part of the internal ear, the labyrinth. When first formed, it is somewhat pear-shaped, the pointed end directed upward. This pointed end later gives rise to the endolymphatic duct. Soon the ear vesicle starts expanding, pushing away the surrounding loose mesenchyme. Parts of the wall of the vesicle become very thin, and the epithelial cells become flat. These parts will be the membranous areas of the labyrinth.
Other parts, particularly those of the medioventral wall of the vesicle, remain thick or even become thicker; the cells in these areas become columnar and give rise to patches of sensory epithelium which form the maculae of the internal ear. Even before it is subdivided into membranous and sensory parts, the ear vesicle gives off on its median surface a group of cells which become the acoustic ganglion (ganglion of nerve VIII).
The expansion of the ear vesicle is unequal, and thus it becomes constricted in some places and bulges out in others. As a result, the shape of the organ becomes increasingly complicated, so that it eventually deserves its name—the labyrinth.
The sacculus is subdivided by a constriction from the utriculus. The utriculus becomes drawn into three mutually perpendicular folds, the rudiments of the semicircular canals. The sides of the folds eventually stick together and become perforated, while parts of the original cavity along the edges of the folds remain open and become the semicircular canals, opening at both ends into the cavity of the utriculus.
A hollow outgrowth of the sacculus forms the rudiment of the lagena in lower vertebrates, and in higher vertebrates this outgrowth becomes very elongated and coiled to give rise to the cochlea. As the ear vesicle changes its shape and produces the various parts of the labyrinth, the sensory areas become subdivided and further differentiated until each of the maculae have taken up their final positions in the fully developed labyrinth.
As the ear vesicle expands to produce the labyrinth, it becomes surrounded by mesenchymal cells which later give rise to cartilage and produce the cartilaginous ear capsule which surrounds and protects the inner ear. There is a direct causal relationship between the ear vesicle and the development of the ear capsule; if the ear vesicle is removed, the ear capsule does not develop, and if a foreign ear vesicle is transplanted in the tail-bud stage, the local mesenchymal cells may aggregate around it and produce an additional cartilaginous capsule.
The mesenchyme that is used for the ear capsule is of mesodermal origin and is derived from the sclerotomes. Mesenchyme of neural crest origin, as well as subcutaneous mesenchyme, is apparently not capable of reacting to induction by the ear vesicle. As a result, an ear vesicle transplanted heterotopically does not always cause a good capsule to be developed around it.
The most complete capsules develop around ear vesicles transplanted in the immediate vicinity of the normal ear, between the ear and the eye, where the grafted vesicle can draw on the same supply of mesenchyme as the normal ear vesicle.
However, the sclerotome mesenchyme of the trunk, that is, the mesenchyme giving rise to the cartilages of the vertebral column and the ribs, reacts to the ear vesicle by forming large masses of cartilage, which may partially surround the grafted ear vesicle.
The concentration of mesenchyme on the surface of the ear vesicle may be partly the result of its expansion, which would lead to the mesenchyme being compressed against its surface. However, there is no doubt that mesenchymal cells may travel considerable distances to reach the ear vesicle and to invest it with cartilage.
In experiments on transplantation of the ear vesicle to the trunk region it can be seen that thick bars of cartilage grow out from the vertebral column to the ear vesicle, presumably indicating the pathway which had been followed by the mesenchyme. In the same experiments it can also be seen that the total amount of cartilage in the area is greatly increased, so that the ear vesicle must either stimulate the proliferation of skeletogenic mesenchyme or increase the proportion of mesenchymal cells which become chondroblasts.
The reverse occurs in the case of the removal of the ear vesicle. Not only does the ear capsule not develop, but there is no superfluous cartilage in the area. In the absence of the ear vesicle the proliferation of the pre-cartilage cells falls short of the normal, or else the cells that should have become cartilage cells differentiate along other paths.
The development of the ear provides another example of a chain of inductions:
1. The primary inductor—the roof of the archenteron, consisting of presumptive chordomesoderm—causes the development of the hindbrain.
2. The hindbrain, as a secondary inductor, stimulates the development of the ear vesicle (in conjunction with the direct action of the mesoderm on the presumptive ear ectoderm).
3. The ear vesicle, as a tertiary inductor, causes the formation of the cartilaginous capsule.
One might have expected that the development of the middle ear would be related to the development of the inner ear, but this is not the case. The middle ear, consisting of the eustachian tube, the ear ossicles (columella in the frog) in the cavity, and the tympanic membrane, develops normally after the removal of the ear vesicle.
This may be because the middle ear is derived from the branchial apparatus, which is an essential part of the vertebrate organization, deeply rooted in the basic mechanism of vertebrate development and thus not in need of stimulation from the inner ear. In this case the functional apparatus, the organ of hearing, is made up of two parts not causally connected in development but linked together only through the medium of their definitive functioning.
Besides the organs developing from placodes, a number of structures are developed from the epidermis which in their early stages can be classified as “outgrowths” or, more correctly, as protrusions. These are the unpaired fin fold, the external gills (in aquatic vertebrates), the “balancer” (in the larvae of salamanders), and the paired limbs.
The Fin Fold:
The unpaired fin fold is a structure found in all fishes and in the larvae of amphibians. It is a vertical fold of skin which starts in the posterior head region or the anterior trunk region, extends backward all along the back and dorsal side of the tail, bends over to the ventral side of the tail, and can be traced forward, along the ventral side of the tail and the belly to the middle of the trunk. The fold consists of epithelium and connective tissue.
In fishes, parts of the fold are later invaded by skeletogenous tissue which produces fin rays, thus transforming these parts into the unpaired fins of the adult. Parts of the larval fin fold in between the unpaired fins of the adult disappear. In amphibians which metamorphose into terrestrial adults (all Anura and some of the Urodela), the fin fold disappears at metamorphosis, but it may persist in neotenic species or in some purely aquatic salamanders (Cryptobranchus).
The fin fold first appears as a longitudinal thickening of the epidermis, which is seen as a ridge on external inspection. The thickening increases by shifting upward and toward the midline of the adjacent strips of the epidermis.
The cells moving in from the right and left flanks remain separated as two layers of epithelium, except at the crest of the ridge. There is, however, no hollow in between the two layers. The fold is hollowed out shortly thereafter by the two epithelial layers separating in the middle, and then connective tissue cells of neural crest origin penetrate into the fold.
Although the neural crest cells enter into the formation of the fin fold in a later stage, they are actually responsible for the determination of the whole structure. If the neural crest cells are removed shortly after their formation, or if the neural folds, from which they arise, are cut away, the fin fold is not developed in the region of the defect. If a piece of the neural fold or a mass of the neural crest cells is transplanted under the epidermis in any part of the body, a fin fold develops at the site of the transplantation.
The ability to induce the development of the fin fold is found in the trunk neural crest cells only. The neural crest cells of the head cannot induce a fin fold, although the epidermis of the head region is fully competent to react by developing a fin fold, if it is exposed to the action of the necessary inductor. The extent to which the fin fold extends anteriorly is thus dependent on the extent to which the neural crest possesses the ability to induce the fold.
The External Gills:
The external gills are protrusions of epidermis, with connective tissue, blood vessels, and muscle inside. They develop as outwardly directed pockets above the gill clefts and are found in some fishes (Polypterus, the lungfishes, Misgurnus) and in amphibian larvae.
The original out-pushing forms the shaft of the external gill. On this shaft secondary branches develop which are formed at first as solid, outwardly directed thickenings of the epidermis of the shaft and are subsequently hollowed out. Sometimes the shaft is so short that the gill appears to be a bunch of filaments which in their turn may develop branches.
The pattern manifested in the development of the gills has been found in the Urodela to be dependent not on the epidermis but on the inner layers of the embryo. A piece of epidermis taken from the flank may be transplanted over the gill region to replace the local epidermis, and if the transplantation is carried out early enough (soon after the closure of the neural folds), the normal development of the external gills is not impeded.
The epidermis of the gill region may be lifted and replaced again after it has been rotated 90 or 180 degrees. If such an operation is carried out before gill development begins, the external gills will appear in their normal positions, as if nothing had happened. If the mesoderm of the gill region is included in the graft which is implanted in inverse orientation, the gills will still develop in their normal positions although the development may not go as smoothly as when only the epidermis is involved.
If, however, the endoderm of the gill region is rotated together with the other germ layers, the developing gills are dependent on the new orientation of the graft, even if it is disharmonious with the other parts of the embryo. It is therefore the endoderm that determines the position of the developing external gills.
The part played by the endoderm in the development of the external gills may also be tested by completely removing the endoderm of the gill region, while leaving the other two layers intact. The result is that the external gills do not develop at all.
The Balancers and Adhesive Organs:
The balancers are tentacle-like organs present in the larvae of many species of urodele amphibians (newts and salamanders). The organ is situated, one on each side, just behind the angle of the mouth underneath the eye. It is a slightly curved cylindrical process, consisting of epithelium and a connective tissue core.
The connective tissue is especially dense just underneath the epithelium, where it forms a cylindrical supporting membrane. Proximally this membrane is attached to the quadrate. The epithelium on the tip of the balancer produces a mucous secretion and is therefore slightly adhesive.
A newly hatched larva of a newt or salamander uses the balancers for support, when resting on the ground, and to prevent the body from falling on one side in the stages when the forelimbs are not yet developed. When the forelimbs become functional, they take over the support of the body, and the balancers gradually degenerate.
The balancers would not deserve our attention if it were not that these simple organs lend themselves to some experiments of considerable interest. The epidermis of the balancer is normally derived from a part of the ectoderm lying just outside the neural fold in the vicinity of the eye rudiment.
Other parts of the ectoderm, however, possess the ability to develop into a balancer when stimulated by an inductor. Balancers have often been induced in experiments on the “primary organizer” and on the inducing ability of the archenteron roof. The development of balancers is an indication that the inductor possesses the regional specificity of an archencephalic inductor. The part of the organizer actually responsible for the induction of the balancer in normal development is probably the archenteron roof, but the adjacent portion of the neural plate, once it has been determined, also possesses the ability to induce a balancer.
The larvae of anurans do not possess balancers, but tadpoles of frogs and toads develop a different organ with somewhat similar biological functions. It is the adhesive organ, also known as the oral sucker. Actually, the adhesion which is affected by the organ is not due to a sucking action but to the secretion of a sticky slime. The adhesive organ is a glandular structure consisting of very elongated columnar cells which produce a copious secretion of mucoprotein.
The swimming powers of a newly hatched tadpole are quite limited, and after swimming for a few seconds the tadpoles tend to become attached either to some submerged objects (to the glass in an aquarium) or even to the surface film of the water. The sticky secretion of the adhesive organ keeps them suspended; otherwise, they would be lying most of the time on the bottom, where the oxygen supply may be considerably depleted.
When fully developed, the adhesive organs vary from one species to another. Most often they are in the form of a V-shaped furrow with thickened edges (especially in toad tadpoles), but in some species of frogs they have the shape of two conical projections or even the form of one unpaired conical outgrowth (in Xenopus).
It is interesting to note that after the ciliae of the epidermis the adhesive organs are the first part of the body of a frog embryo to become functionally differentiated. The differentiation of these organs begins in some toad species immediately after neurulation, and they start secreting before the tadpole can swim.
This is advantageous for the animal because, after hatching, the tadpoles for some time remain hanging on the outer surface of their egg membranes. At this stage the tadpoles possess a limited power of locomotion owing to the ciliary action of their epidermis. As the tadpole develops and its swimming ability improves, the adhesive organs cease secreting and degenerate.
The Paired Limbs:
The early rudiments of paired limbs are similar to the gill rudiments in that they are out-pushing’s of the epidermis which are filled with a mass of mesenchymal cells. In the case of the limbs, however, it is the differentiation of the mesenchyme into parts of the skeleton and muscle of the limb that deserves the greatest attention. The limbs will therefore be dealt with in conjunction with the organs derived from the mesoderm.
There is still one more structure of great importance which is derived from the epidermis, namely, the mouth invagination (the stomodeum). Its formation and further development can be most conveniently treated together with the other parts of the alimentary canal, which are derived from the endoderm.