1. The present report is based on an experimental analysis of interactions of epidermis and dermis in the formation of the scale and its homologue, the feather. The method used consists of reciprocal exchange of epidermis and dermis from prospective scaled and feathered regions (tarsometatarsus and mid-dorsum), at stages before and during primordia formation. The recom-bined tissue-layers were grown on the chorioallantoic membrane.

  2. Results of examination of over 500 grafts have shown that both the inductive capacity of the dermis and the competence of the epidermis vary with developmental stage and with location on the body.

  3. Dermis from the tarsometatarsus acquires specific inductive capacity late in ontogeny; not until approximately the 13th day does the inductive stimulus reach a level of intensity sufficiently high to elicit scale formation in overlying ‘foreign’ epidermis. Feathers develop regularly in association with scales induced in the grafts.

  4. Dermis from the prospective spur region of the tarsometatarsus is excep-tional in that it possesses intense inductive capacity at 9 days, the earliest stage tested. The spur scale induced in ‘foreign’ epidermis of the graft is always free of associated feathers, as in the normal chick.

  5. Dermis from the mid-dorsum, like that of the spur region, is strongly inductive at the earliest stage tested (5 days) and continues strong throughout the entire age span used ( days). The inductive capacity of this feather-inducing dermis is not of short duration.

  6. The competence of epidermis from the mid-dorsum changes within a brief period of time between and days. This change can be detected only when the epidermis is brought into contact with tarsometatarsal dermis that has developed specific inductive capacity (13 – 15 days). At this time, the later -day epidermis, unlike that of earlier stages, produces no typical scales but continues to differentiate in a feather direction. The feathers, however, are not entirely normal.

  7. Epidermis from the tarsometatarsus exhibits a change in competence between 12 and 13 days. This change can be detected only in the structural aberrance of the resulting feathers. Those produced from 13-day epidermis are extremely aberrant. (Scales are never produced from tarsometatarsal epidermis in contact with mid-dorsal dermis, irrespective of the stage of either tissue component.)

  8. The change in competence with age, noted above for both mid-dorsal and tarsometatarsal epidermis, can be entirely masked if each is brought in contact with a stronger inductor, such as the dermis of the beak. Dermis from the beak ( days) induces a perfect beak in both -day mid-dorsal epidermis and 13-day tarsometatarsal epidermis.

  9. The remarkable ability of the epidermis to alter its course of differentiationat relatively late stages suggests that the morphological similarities between scale and feather may be more than superficial.

  10. Since tarsometatarsal dermis is capable of inducing both scales and feathers in the normal chick and in grafts, it appears to be bipotential. The possibility exists, however, that the epidermis of birds possesses inherent feather-forming tendencies which, in certain regions, become modified second-arily by the underlying dermis.

The foot of developing embryos of scalefooted varieties of the common domestic fowl (Silver Campine, White Leghorn and others) frequently exhibits feather primordia emerging from the scales of various regions. Many of the feather germs are rudimentary and disappear by the time of hatching, but some persist and differentiate into normal feathers. Such feathers are often inconspicuous and easily overlooked (Plate 1, figs. I-M).

Scales bearing feathers by no means represent an unusual phenomenon. Indeed, they were clearly identified by many of the earlier workers, particularly those interested in homologizing these two specialized epidermal derivatives (Jefferies, 1883; Davies, 1889; Bornstein, 1911; Blaszyk, 1935). More recently, the frequent association of feathers with the scales of those pedal components (digits) which arise from the wing ectoderm, after implantation of mesoderm from the prospective foot region, has been noted by Saunders, Cairns & Gasseling (1957) working with White Leghorns.

Although scales are the most common integumentary covering of the foot (tarsometatarsus, digits and plantar surface) and, often, adjacent portions of the tibiotarsus, they are not uniquely avian structures and do not distinguish birds from other vertebrates as does the possession of feathers. There are many birds in which the foot is partly or completely covered with feathers throughout life (Rawles, 1960). Even in closely related species, either the feathered or the scaled condition may predominate.

Such observations strongly suggest that the prospective skin of the foot, unlike that of other body regions, is not irrevocably committed to the formation of one type of specialized epidermal derivative; rather, it is potentially capable of giving rise to either or both feathers and scales. The fact that the two primordia are fundamentally alike in origin and in the early phases of their development points to a strong intrinsic relationship between feather and scale. This is not surprising, for feathers are generally believed to have evolved from the reptilian scale.

To shed some light upon this interesting and often observed relationship between the scale and its homologue, the feather, the following experimental study of the interactions between epidermis and dermis, before and during the formation of the primordia, was undertaken. The method employed involves exchanging the separated epidermis and dermis from small areas of skin, re-moved from prospective feathered and scaled regions at various early developmental stages. The ‘chimeras’ thus produced, each consisting of an epidermis and a dermis, but of different regional origin and of different developmental age, were then transplanted to the chorioallantoic membrane to continue their development and growth.

Most of the chick embryos used for this investigation were from a true-breeding flock of Silver Campines, a pigmented variety of domestic fowl. White Leghorn embryos from a reliable source were also used for comparison. Eggs were incubated at a temperature of 37–38°C. in a forced-draft incubator with humidity control.

The general operative technique involves four steps: (1) Isolation of small pieces of whole skin from prospective scaled and feathered regions. (2) Separation of the isolated piece of skin into its component parts, epidermis and dermis. (3) Recombination of epidermis and dermis in new and unusual contact relations. (4) Transplantation of the recombined epidermis and dermis to the chorioallantoic membrane (Text-fig. 1).

TEXT-FIG. 1.

Diagram showing preparation of grafts. D = Dermis; E = Epidermis.

TEXT-FIG. 1.

Diagram showing preparation of grafts. D = Dermis; E = Epidermis.

1. Isolation of skin from the foot (tarsometatarsus) and from the back (mid-dorsum)

The skin in its entirety was removed from the anterior and posterior faces (front and back) of the foot at stages prior to and during the formation of scales (9–15 days). This was done by making an incision down each side of the tarsometatarsus and across its proximal and distal ends, excluding the zone of junction with the tibiofibula and the digits. From each of the two strips of skin, small rectangular pieces of approximately 1–2·5 mm. long (the size of the piece depending upon the age of the foot) were isolated at various axial levels. A strip of skin of matching width was removed from the mid-dorsum, between the uropygium and the base of the neck, of embryos ranging in age from 5-to -days’ incubation, i.e., before and during the formation of feather primordia. From this narrow median strip, which included the central portion of the prospective dorsal feather tract, rectangular pieces of skin of approximately 1–2·5 mm. in length were removed at various axial levels.

2. Separation of the layers of skin with trypsin

The method of separating the epidermis from the dermis or ‘skin splitting’ with a dilute solution of trypsin held at 37°C. was introduced by Medawar (1941) working with Thiersch grafts. Both layers remained viable after separation and were found suitable for grafting. Later, Szabó (1955) made a very interesting discovery, namely, that the two layers of skin could be separated with greater ease and still remain viable, if the treatment with dilute trypsin was done at ca. 4° C. for a longer period of time. It appears that lowering the temperature slows down the rate of enzymatic action far more than the rate of diffusion. Thus, the enzyme is more evenly distributed and no maceration of the basal, germinative cells occurs.

In the present experiments with embryonic skin, Tzabó’ s method was used with slight modification; for instance, a 1 per cent, trypsin solution was used for a shorter period of time in the cold. In the hands of the present investigator, the ‘cold’ method was found to be infinitely superior to the ‘warm’ method for separating the epidermis from the dermis at all ages used. The difficulty experienced by Wessells (1962), for instance, in removing epidermis intact from the large scales of the anterior ‘shank’ of 12-day embryos, can be eliminated entirely by the use of cold trypsin. In the present experiments, epidermis was separated intact, with the greatest ease, from all regions of the foot of the older embryos, 12-to 15-days’ incubation age.

The isolated pieces of skin were placed in a 1 per cent, solution of trypsin, Disco 1:250, in a balanced salt solution (Chick Ringer, Tyrode and Hanks’ solutions were all equally effective) held at approximately 4° C. for 1–3 hr., or until the epidermis begins to loosen at the edges. The time required for the separation depends, naturally, upon the age (thickness) of the skin. One hour or less was usually sufficient for 5-to 7-day skin from the mid-dorsum and 9-to 11-day skin from the foot. Older skin from the foot and the back usually separated easily after ca. 2–3 hr. With a little experience one can judge the time accurately by observation alone. While the skin can remain in the cold trypsin solution for a longer period of time without harm, the epidermis may float off, in which case the orientation may be lost. With proper timing the epidermis can be lifted off absolutely intact, without curling or rolling at the edges and completely free of adhering mesodermal cells. No picking or pulling is necessary. Furthermore, the dermis does not become swollen and sticky but remains firm and easy to handle. This was important for the present experiments in which both layers of the same isolate were used.

3. Recombination of the separated layers of skin

In making the exchange between the separated layers of skin of different origin and age, the dermis (or prospective dermis) of tarsometatarsal origin was overlain by epidermis (or prospective epidermis) of mid-dorsal origin and vice versa. In assembling the ‘chimeras’, the basal layer of the epidermis was placed over the upper surface of the dermis in normal contact relations. The reassembly of the epidermis and dermis was done in a cold 1:1 mixture of fresh egg albumen and a balanced salt solution, usually Ringer, thoroughly shaken and decanted. The following antibiotics were usually incorporated into the solution: streptomycin sulfate (0·05 mg./c.c.), penicillin (100 units/c.c.). The diluted albumen serves a dual purpose; it is required to inhibit further action of the trypsin and it reduces surface tension, making the subsequent handling of the very thin epidermis much easier. After epidermal-dermal recombination, excess fluid was carefully drawn off with a pipette and the piece of skin allowed to come to room temperature. (As the tissue-layers warm, they tend to stick together.) After 2–3 min., diluted albumen was added in sufficient quantity to float the graft on to a spatula, without displacing either of the component tissues.

4. Transplantation to the chorioallantoic membrane

By means of a small spatula, the graft of recombined layers of skin was transferred intact, dermal side down, to the exposed chorioallantoic membrane of host embryos of 10-to 11-days’ incubation. (The older membranes proved more suitable for this particular type of graft.) Before placing the graft, the membrane was lowered, by puncturing the air chamber at the blunt end of the egg, and moistened with a few drops of the diluted albumen mixture. To insure the necessary close contact of the graft with the membrane, excess fluid was carefully drawn off with a pipette. The opening or ‘window’ in the shell over the graft was covered with ‘parafilm’ and the egg returned to the incubator.

Because the chorioallantoic membrane was lowered, the epidermal surface of the graft is exposed to an artificially created airspace between it and the overlying egg shell. It was soon discovered that the feather germs arising from the epidermal surface of grafts thus exposed to air often failed to elongate normally, the tendency being to increase in diameter rather than in length (Plate 2, figs. H, I). Grafts composed of epidermis and dermis that have been separated and recombined are much more susceptible to harm from drying than are similar pieces of whole skin.

To prevent structural abnormalities due to drying, it was found necessary to keep the graft submerged during its 8 or 9 days’ period of growth on the chorioallantoic membrane. Therefore, as soon as the graft had become firmly attached to the membrane, approximately 12–18 hr. after grafting, fresh egg albumen plus a small amount of balanced salt solution (ca. 15 parts of albumen to 1 of salt solution, thoroughly shaken and decanted) was pipetted gently each day over the exposed surface of the graft (under magnification) in sufficient quantity only to keep it submerged. The embryo (host) cannot tolerate an excessive amount of fluid at one time. Examination of the graft each day has the added advantage of enabling one to follow closely the formation of the feathers and scales. With the daily addition of fluid, the membrane gradually rises bringing the graft nearer and nearer the shell level. To allow for this rise, it was necessary to create more space above the shell for the added liquid. This was done simply and satisfactorily by placing a ‘chimney’ over the original opening in the shell. The ‘chimney’ was made of a spiral of cotton cord coated with paraffin to preserve its form (Text-fig. 2). The above-described method permits normal differentiation of the feathers and scales. No difficulty was experienced in getting the grafts to ‘take’ on the membrane. It was extremely unusual not to recover a graft.

TEXT-FIG. 2.

‘Chimney’ over opening in shell, devised to provide extra space for daily addition of diluted albumen.

TEXT-FIG. 2.

‘Chimney’ over opening in shell, devised to provide extra space for daily addition of diluted albumen.

The majority of the 534 grafts obtained were allowed to continue their growth on the chorioallantoic membrane for the maximum period (8-9 days), i.e., until the host was nearly ready to hatch (19th day). Thus, the total age of the epidermis of the graft ranges from 1 week to several days under hatching age. A few grafts were removed for study at intervals between the third and sixth days after transplantation. Grafts selected for histological study were fixed in Bouin’ s fluid, sectioned serially at 8 μ, and stained with Ehrlich’ s acid hematoxylin, counterstained in some cases with eosin.

The experiments were adequately controlled. In the controls, the pieces of skin isolated from the mid-dorsum and the tarsometatarsus, at the various developmental stages used, were treated with cold trypsin to separate the two layers. The epidermis was then replaced immediately over the same dermis from which it had been separated. No exchange was made. The control piece of skin was grown as usual on the chorioallantoic membrane.

Before presenting the experimental findings, certain pertinent information concerning the normal development of the feather and the scale should be briefly mentioned.

The first indication of the site of origin of each derivative is seen in the meso-derm immediately beneath the ectoderm, as a condensation of the loosely arranged cells into localized thickenings. These aggregates of mesodermal cells, the primordia of the dermal papillae, soon induce specific proliferative reactions in the contiguous ectodermal cells (prospective epidermis). Increase in the cells of both the dermal mass and the overlying epidermis leads to the formation of the characteristic elevations (primordia) that gradually appear on the surface of the skin of the prospective feathered and scaled areas of the embryo (Plate 1, figs. B-D, G, H).

The primordia arise in an orderly, time-space sequence until a definite number are established. Those of the scales are uniformly distributed over the entire surface of the foot; those of the feathers are confined to specific areas, tracts, separated by apteria.

During ontogeny, there is a notable difference in the time at which the two types of primordia first appear. The feather papillae begin to show up externally in certain tracts towards the end of the 1st week of incubation. In the dorsal feather tract, for instance, between and 7 days (Plate 1, fig. G). The scale papillae are just beginning to become visible externally, on the anterior face of the foot, by approximately the 11th day of incubation; they are more clearly defined in the distal portion of the foot (Plate 1, fig. B). The surface of the opposite, posterior face of the 11-day foot is still smooth (Plate 1, fig. A). About 1 day later, primordia are becoming visible gradually on the surface of the posterior face. The marked difference in the advancement of the primordia of the scales of the two regions of the foot at day 13, are shown in Plate 1, figs. C and D. Thus, by the time the entire surface of the foot is more or less covered with recognizable scale papillae, the body of the embryo is completely covered with long, tapering, fully differentiated down feathers.

While the early phases of the formation of scale and feather primordia are fundamentally alike, the subsequent development of the two differs greatly. The feather germ elongates, as it grows rapidly outward, to form a tapering cylinder of epidermal cells filled with a dermal core or pulp containing blood vessels. The inner portion of the wall of the epidermal cylinder becomes divided into a series of parallel ridges, the primordia of the barbs and barbules; the outer part forms a protective sheath. The base of the cylinder gradually sinks below the surface of the skin into a tube-like follicle with the dermal papilla located in its base.

In contrast to the feather, the scale develops as a horizontal, epidermal plate composed of cells from epidermis overlying the upper (outermost) surface of the flattened dermal papilla. As the epidermis increases rapidly in thickness, the cells of its outer layers gradually become completely cornified to form a rigid, horny plate (Kerbert, 1877). The adjoining epidermis of the more basal portion of the papilla, the prospective interscale area, remains thinner and less highly cornified. The variety of size and shape exhibited by the scales and their-pattern of arrangement on the tarsometatarsus may be seen in Plate 1, figs. I-M.

Results of the interactions between the two recombined layers of tissue, of different regional origin and of different developmental age, are judged on the basis of the feather-like or scale-like properties exhibited by the epidermal outgrowths. In presenting the results, the data will be grouped into subdivisions according to the type of epidermal-dermal recombination and the age of the components.

Series 1. Epidermis from the mid-dorsum combined with dermis from the tarsometatarsus

In this series of experiments, epidermis from the mid-dorsum, a region that would normally give rise only to feathers, is placed over dermis from the tarso-metatarsus, a region that would normally produce scales. Thus, a unique contact relationship between the two component tissues of the skin is established. The information sought concerns the course of differentiation that the epidermis will now follow in response to inductive stimuli from the ‘foreign’ dermis.

Data obtained from a study of 319 grafts recovered from this type of epidermal-dermal recombination, made at successive developmental stages, before and during the formation of the feather and scale primordia, are summarized schematically in Text-fig. 3.

TEXT-FIG. 3.

Schematic representation of results of dermal-epidermal recombinations of experimental Series l. Broken heavy lines indicate occasional formation of normal feathers and scales. A = Dermis from anterior tarsometatarsus; P = Dermis from posterior tarsometatarsus.

TEXT-FIG. 3.

Schematic representation of results of dermal-epidermal recombinations of experimental Series l. Broken heavy lines indicate occasional formation of normal feathers and scales. A = Dermis from anterior tarsometatarsus; P = Dermis from posterior tarsometatarsus.

(a) 5-to -day mid-dorsal epidermis over 9-to 11-day tarsometatarsal dermis

In this group of recombinations, epidermis from the mid-dorsum at all stages tested, 5-to -days, gives rise to typical down feathers when placed over dermis from either the anterior or the posterior face of the foot (tarsometa-tarsus) of embryos of 9-to 11-days’ incubation age. Although the dermis is necessary for further growth and differentiation of the overlying epidermis (neither epidermis nor dermis alone is capable of differentiation), no specific effect of its presence can be detected in the grafts. The recombined epidermis continues to differentiate as it would have done under normal contact relations. Representative grafts are shown in Plate 2, figs. A-F. The feathers composing the down plumage of these grafts, to all intents and purposes, may be considered normal in structure and color. The feather cylinder is long and tapering, composed of eleven to sixteen barbs, each bearing two rows of barbules, and the whole complex enclosed by a feather sheath, as in the normal unhatched chick. Plate 2, fig. G shows a feather from a graft after removal of the feather sheath and the feather parts allowed to dry. It will be observed that the barbs and barbules spring apart, as they become dry, to form a downy mass characteristic of the young, hatched, chick.

In the normal Silver Campine, the down plumage of the dorsal feather tract is composed of both white and pigmented feathers. The white are confined chiefly to lateral stripes bounding a median, broader, pigmented (slate gray) stripe (Rawles, 1959). Since the prospective white-striped areas are often irregular in contour, their inclusion in the original isolate is not unusual.

When included, white down feathers appear among the pigmented on the graft surface (Plate 2, figs. B, F, H).

(b) 5-to -day mid-dorsal epidermis over 12-day tarsometatarsal dermis

The first indication of a change in the response of mid-dorsal epidermis to specific inductive stimuli from the underlying foot dermis begins to show up when combined with dermis of embryos of 12-days’ incubation age. Typical scales, for the first time, appear on the surface of a few of the grafts, and the feathers are, on the whole, less normal in general appearance than those which developed in contact with younger foot dermis (9–11 days). Since the control grafts, in which epidermis is separated from the dermis and immediately replaced, prove that the method employed permits normal differentiation of feathers, the occurrence of aberrant feathers, therefore, implies that the older, 12-day, foot dermis is participating more actively in the epidermal-dermal reaction than previously (9–11 days).

For the first time, also, differences in the response of the epidermis to dermis from the anterior, as opposed to the posterior, face of the foot can be detected. The only grafts that develop scales are those from combinations in which dermis from the lower, anterior face of the foot was used.

Mid-dorsal epidermis (5-7 days) combined with dermis from the distal portion of the anterior face of the foot, where the scale primordia are more advanced, now and then gives rise to scales with associated feathers (Plate 3, figs. A, D). Many of these feathers are relatively long and tapering. Scales were not observed in similar combinations with epidermis of the later stages, to -days (Plate 3, figs. G, H, J). Aberrant feathers are the only indication of an effect of the stronger dermal stimulus.

With the exception of the spur scale, no other scale formation can be identified in grafts from combinations including dermis from the posterior face of the 12-day foot. Epidermis, of all stages tested, in contact with dermis from this posterior region of the foot continues to develop feathers, many of which are relatively normal in general appearance (Plate 3, figs. B, C, E, F, I).

The dermis of the 12-day foot appears to be in a transitional state in regard to the intensity of the inductive stimulus. Only in certain areas has it yet become sufficiently strong to exert a specific effect. Slight individual differences in developmental age at this particular period may account, in part, for the observed variation in epidermal response.

(c) 5-to -day mid-dorsal epidermis over 13-to 15-day tarsometatarsal dermis

By approximately the 13th day of incubation, the scale-inducing capacity of the dermis, of both anterior and posterior faces of the foot, has increased in intensity to the extent that scales develop regularly in grafts from combinations including mid-dorsal epidermis of 5-to -day embryos, which previously had differentiated primarily in a feather direction. No further increase in the in-tensity of the scale-inducing stimulus can be clearly demonstrated in combinations including older dermis from the foot of 14- and 15-day embryos. Thus, epidermis from the mid-dorsum of 5-to -day embryos underlain by dermis from the foot of 13-, 14-and 15-day embryos all produce scales (Text-fig. 3). In form and structure, the scales induced from mid-dorsal epidermis are typical of the region of the foot from which the dermis was removed. Feathers, showing varying degrees of aberrance, occur frequently in close association with the scales. While these outgrowths from the scale may often appear quite stunted and abnormal, they nevertheless possess the distinguishing features of a feather, barbs and barbules, as confirmed in histological section.

Grafts obtained from various epidermal-dermal combinations within this age group are shown in Plate 4, figs. A-F. From an examination of the figures, it will be observed that the scales induced by dermis from the anterior face (front) of the foot (Plate 4, figs. A-C), are consistently larger and heavier than those induced from the opposite, posterior face (back) of the foot (Plate 4, figs. D-F), as in the normal chick. The pronounced regional specificity of the scales is clearly demonstrated in all grafts of this category. It will be observed that the feathers which arise from the margins of the large, median scales of the front of the foot are fewer and more atypical than those arising from the smaller, median and lateral scales of the back of the foot (compare Plate 4, figs. B and C with Plate 4, figs. D-F).

The induction of feathers more often from epidermis overlying the primordia of the smaller scales, as seen in Plate 4, figs. D and E, for example, demon-strates differences in the inductive capacity of the dermis, within a relatively small area, at the same chronological age. This is even more strikingly shown in Plate 4, figs. G and H, in which the dermal primordium of the spur induces a normal spur scale. The epidermis in contact with primordia of the small round scales, immediately surrounding the spur, forms feathers. Since the smaller scales of the lateral and posterior regions of the foot appear later in ontogeny and are, therefore, of younger developmental age, it would seem that the intensity of the dermal stimulus of these particular areas has not yet increased sufficiently to induce scales in the overlying mid-dorsal epidermis.

The scales that develop in chorioallantoic grafts do not overlap as tightly as those of the normal foot. Since the control grafts show this same surface configuration, it is unlikely that it results from the techniques of skin splitting and recombination, per se. In part, it may be due to the fact that the raw edges of the grafted tissues show a strong tendency to encircle the point of entry of the blood vessels from the membrane, exerting tension on the graft. Also, the absence of the normal underlying bony framework probably contributes to their rather ‘puffy’ appearance.

The pigmentation of the scales deserves comment. In the normal Silver Campine embryo, the scales are very lightly pigmented by melanocytes that differentiate only in the epidermis of the dorsal or upper surface of the scale.

The ventral portion of the large overlapping scales does not become pigmented. The epidermis between the smaller scales is also unpigmented. The heavy pigmentation observed in the scales of the grafts, induced from mid-dorsal epidermis of the Campine, is the result of multiplication of a few melanoblasts that have already entered the epidermis by the time the isolation was made. These melanoblasts, which under normal circumstances would pigment the down feathers, now differentiate and produce pigment in the scales, but only in those specific portions of the scales that would normally support pigment cell growth and differentiation. (For pertinent information concerning the selective association of embryonic pigment cells, the reader is referred to Weiss & Andres, 1952; and Rawles, 1955, 1960.)

(d) 8-to -day mid-dorsal epidermis over 13-to 15-day tarsometatarsal dermis

The data of the preceding combinations have clearly demonstrated that dermis from the foot of 13-to 15-day embryos is fully capable of eliciting scale formation in mid-dorsal epidermis within a wide age range between 5 and days. Shortly afterwards, however, by approximately 8-to -days of incubation, there is a notable change in the response of the epidermis to this scale-inducing influence. Feathers, rather than scales, are now the predominant structures arising from the surface of the grafts. Representative cases are shown in Plate 4, figs. I–O. It will be observed that the feathers are numerous and rather equally spaced over the surface of the grafts. While many of the feathers elongate and approach the normal in general appearance, they are, on the whole, less perfect in structural detail than those which developed from epidermis of the same age underlain by younger foot dermis (9–12 days). As noted previously, here, also, regional differences in the intensity of stimuli from the dermis of the two faces of the tarsometatarsus are reflected in the general appearance of the resulting epidermal derivatives. If one compares Plate 4, figs. I and J with Plate 4, figs. L and M, for instance, it will be seen that the more typical feathers arise from epidermis in contact with dermis from the posterior face of the foot, particularly from the regions bearing the primordia of the smaller scales (Plate 4, figs. H, I). No differences in the tissue interactions could be detected between the White Leghorn and the Silver Campine fowl.

The epidermis covering the grafts is characteristic of the feathered regions of the body. It is relatively thin, smooth and translucent in the living state (before fixation). No typical scales could be recognized from examination of the grafts in toto. In histological section, however, evidence of aberrant scale formation was occasionally revealed by the columnar arrangement of the basal cells of the epidermis in some areas, accompanied by a thickening of the over-lying stratum corneum. Therefore, on the basis of their histological characteristics, these scale-like epithelial areas may be regarded as aberrant scales.

Grafts showing aberrant scale formation may be seen in Plate 4, figs. K and L. (The graft shown in Plate 4, fig. L, is exceptional in that it did not develop under the ‘window’ but tightly pressed against the egg shell. Thus, mechanical factors undoubtedly account, in part, for the extreme aberrance of the resulting feathers.) The scale-like portions of the epidermis are the heavily pigmented areas or spots. The epidermis of these areas developed in contact with the primordia of the prospective dorsal or outer face of the scale ridges, where the inductive stimulus appears to be stronger. If the primordia of the scales had developed normally, the proximal portion of one pigmented area would have overlapped the distal portion of the next, etc., thus covering most of the white area lying between (compare Plate 4, figs. K and L, with Plate 4, fig. C, which shows normal scale formation). In grafts of this type, that show scale-like formation, the feathers often tend to be even more aberrant, but, nonetheless, they are distinctly feathers and more numerous than those occasionally found in scaled regions of the normal chick.

The restriction of scale formation to this very aberrant expression appears to be correlated with the height of the feather primordia, i.e., the age of the epidermis, at the time of recombination. When the primordia (dome-like elevations) rise 0·1–0·2 mm. above the surface of the skin of the mid-dorsum, the reaction is completely in a feather direction. Not even aberrant scales form (Plate 4, figs. N, O).

Since the grafts from this entire group of recombinations are primarily feathered, the indication is that by 8 to days when the feather primordia have already begun to elongate, the course of epidermal differentiation is so strongly fixed in a feather-forming direction that it cannot be readily changed. The only indication of any response to the older 13-to 15-day dermis is seen in varying degrees of abnormality of feather structure: the feathers are not entirely normal.

Series 2. Epidermis from the tarsometatarsus combined with dermis from the ‘mid-dorsum

This series of experiments is the reciprocal of the preceding, Series 1. Contact is now established between epidermis from a region that normally forms scales, and dermis from a region that normally forms feathers. The question to be answered, as before, is : What course of differentiation will the epidermis follow in response to inductive influence from the underlying ‘foreign’ dermis?

It was not found practicable to recombine epidermis, separated from the 14- and 15-day embryos, with mid-dorsal dermis. By this late age the scale ridges are steep and overlapping, and the epidermis has become very thick. Due to the overlapping of the larger, median scales, it is impossible to stretch out the epidermis flat enough to make continuous contact between the basal, germinative cells and the underlying dermis. Similarly, the control grafts of skin after approximately 13 days do not, as a rule, develop normally.

Results obtained from an examination of 215 grafts obtained from this series of epidermal-dermal recombination, at successive developmental stages, are shown schematically in Text-fig. 4. In every recombination of this series epidermal differentiation is solely in a feather-forming direction. No evidence of scale formation whatsoever could be detected in the epidermis of any of the grafts of this series. The absence of scales, or scale-like formations, was confirmed by histological examination of serial sections of fifty grafts representative of the various age combinations. As the epidermis increases in age, however, the resulting feathers become progressively more aberrant, irrespective of the age of the underlying dermis.

TEXT-FIG. 4.

Schematic representation of results of dermal–epidermal recombinations of experimental Series 2.

TEXT-FIG. 4.

Schematic representation of results of dermal–epidermal recombinations of experimental Series 2.

(a) 9-to 11-day tarsometatarsal epidermis over 5-to -day mid-dorsal dermis

Structurally normal down feathers developed regularly from tarsometatarsal epidermis of 9-to 11-day embryos, from either aspect of the foot, when under-lain by dermis from the mid-dorsum of 5-to -day embryos. Typical cases are shown in Plate 5, figs. A-F. The feathers of the grafts possess the distinguishing features of a normal down feather: they are composed of numerous barbs bearing rows of barbules, enclosed by a sheath. It will be observed that the general quality of the feathers of this group of grafts, derived from epidermis of tarsometatarsal origin, differs from that of feathers derived from epidermis of mid-dorsal origin. The former tend to be shorter, thicker, and of a coarser texture than the latter (compare Plate 5, figs. A-F with Plate 2, figs. A-F). Such qualitative differences are most likely correlated with differences in keratinization of the epidermis of the two regions of the body, back and foot.

Feathers derived from epidermis of tarsometatarsal origin are also more lightly pigmented than those derived from epidermis of mid-dorsal origin. This is attributed to differences in the number of prospective pigment cells in the epidermis of the two regions at the time of isolation. Thus, in color as well as in quality, the feathers of the grafts resemble those that develop often from the foot region of a normal chick in association with scales (Plate 1, figs. I-M).

The general characteristics of the feathers of all of the grafts of this particular category are indistinguishable, irrespective of the age of the dermis used in the recombination.

The undifferentiated mesenchyme (prospective dermis) of the mid-dorsum of the early 5-to 6-day embryos (Plate 1, figs. E, F) is just as effective in calling forth a specific feather-forming response in the overlying tarsometatarsal epidermis, as that of later stages in which the primordia of the dermal papillae are clearly visible as condensations or thickenings in the dermis (compare Plate 5, figs. C, E, F, with Plate 5, figs. A, B, D). Of especial interest is the graft shown in Plate 6, fig. A, in which undifferentiated mesoderm from the mid-dorsum of a 5-day embryo induced normal feathers in undifferentiated ectoderm of the 9-day tarsometatarsus.

(b) 12-day tarsometatarsal epidermis over 5-to -day mid-dorsal dermis

Feathers continue to arise from the older 12-day foot epidermis overlying mid-dorsal dermis of the various ages tested, but, on the whole, they are less normal in general appearance than those which developed from combinations with foot epidermis of earlier stages (compare Plate 5, figs. G, K, with Plate 5, figs. A, E, F). While some of the feathers elongate and are structurally normal (Plate 6, fig. B), others fail to attain such structural perfection.

The first feathers to emerge from a graft are more normal that those that follow. Plate 5, fig. H, for example, shows a graft removed after 4 days’ growth on the chorioallantoic membrane. The feather primordia are normal to the unaided eye and in histological section. The thin epidermis of this graft appears to be developing under tension caused by the formation of a vesicle in the under-lying mesoderm (dermis). Vesicles occur frequently in this series of epidermal-dermal recombinations and occasionally reach enormous proportions (Plate 5, fig. K). The graft shown in this figure is the same type of graft shown in Plate 5, fig. H, but it was left on the membrane for four additional days’ growth. The space between the feathers indicates that they were forced apart by the expand-ing vesicle. The four normal feathers were the first to emerge. Those which began to form later show varying degrees of structural abnormality.

(c) 13-day tarsometatarsal epidermis over 5-to -day mid-dorsal dermis

Epidermis from the 13-day foot, which has already begun to form scales, responds to the inductive stimulus from the underlying dermis of mid-dorsal origin by changing its course of differentiation from scale formation to feather formation. Since the great majority of feathers arising from the surface of the grafts fail to elongate to any appreciable extent, and therefore appear as round to ovoid protuberances, their feather properties are not readily distinguishable to the unaided eye (Plate 5, figs. I, J). The identification of barbs and barbules in histological section, however, clearly proves that these structures are, in fact, feathers. Sections showing some of the various types of feather formation occurring in this kind of graft are shown in Plate 6, figs. C-G. It is interesting to note that the feather cylinder, although it fails to elongate, passes through the normal developmental sequence of sinking down into the dermis and forming a tube-like feather follicle with the dermal papilla located at its base (Plate 6, figs. C-E). The degree of structural aberration varies greatly. The large feather, for instance, shown in Plate 6, fig. G, developed nine barbs with barbules. The smaller, more atypical (aberrant) feathers (Plate 6, figs. C-E), also develop barbs and barbules, but they are fewer in number. In some cases, the growth of the feather primordium appears to have been arrested at a very early stage, when barb ridges are just beginning to form (Plate 6, fig. F). This may be due to cornification of the overlying epidermis which normally would be occurring at approximately this time and which might arrest further development.

The results of the two series of recombination experiments are in agreement as regards the participation of both tissue-layers in the formation of feathers and scales. Moreover, they reveal interesting and significant differences in both the inductive capacity of the dermis and the competence of the epidermis. These differences appear to be related to the location of the dermis and epidermis on the body of the embryo as well as to the stage of development.

Inductive capacity of the dermis

The intensity of the inductive stimulus from dermis of tarsometatarsal origin reaches a level sufficiently high to elicit scale formation from epidermis, that would normally produce feathers, by approximately the 13th day of incubation. At this developmental stage, and through the subsequent 2 days tested, typical scales are induced in epidermis of mid-dorsal origin (5 to days) which had produced only normal feathers when placed in contact with dermis from the foot of earlier age. This change in the effectiveness of the dermal stimulus becomes manifest within a relatively brief period of time, between the 12th and 13th days (Text-fig. 3). Even though, prior to this time, no specific influence of the dermis can be detected in the grafts, its presence is necessary. Without the underlying dermis, no epidermal differentiation, of any kind, occurs. (This has also been demonstrated by others; see Saunders, 1958, for references.) Thus, dermis from the foot of 9-to 11-day embryos, although without influence on the eventual type of epidermal derivative formed, appears to play an important rôle in the further growth and maintenance of the overlying epidermis (compare McLoughlin, 1961).

Within the region of the foot, the dermis acquires the ability to induce scales in certain areas slightly earlier than in others. For instance, dermis from the anterior face of the tarsometatarsus (but not the posterior) occasionally induces scales at day 12. At this developmental stage, however, the general trend of epidermal differentiation still remains primarily in a feather direction.

Therefore, for a relatively long age period, from the 9th until approximately the 13th day, no marked change in the intensity of the dermal stimulus from the tarsometatarsus can be detected in the grafts.

There is no question that dermis from the 13-to 15-day tarsometatarsus is fully capable of calling forth a specific scale-forming response in mid-dorsal epidermis. Yet, interestingly enough, at these stages, when the inductive capacity of this dermis appears strongest, it cannot, even now, entirely overcome the prevailing disposition of the mid-dorsal epidermis to produce feathers. The number of feathers associated with scales induced in this epidermis from a region that would normally produce feathers, exceeds the number generally found on the normal foot.

There is, however, one small area of the dermis of the lower posterior portion of the foot, the spur primordium, which is exceptional in showing intense inductive capacity at a very early stage. Dermis, or prospective dermis, from the spur region is strongly inductive at the earliest stage used (9 days) and remains strong throughout the entire following age series. Dermis from this particular area, between the ages of 9 and 15 days, induces, without fail, a normal spur scale in overlying epidermis of mid-dorsal origin, at all ages tested (5 to days). The spur scale of the grafts, like that of the normal embryo, is always entirely free of associated feathers.

Dermis from the mid-dorsum, like that of the spur, also has intense inductive capacity within the entire age span used in the present experiments (5 to days). Inductive stimuli from early, prospective dermis, of 5-to -day em-bryos, before the dermal condensations are visible, are already sufficiently strong to call forth a specific feather-forming reaction in overlying epidermis of tarsometatarsal origin at all ages tested, 9 to 13 days. This dermis continues to be strongly inductive throughout the remaining age periods used, 7 to days. Thus, there seems to be no relation between the appearance of dermal condensations (primordia of the dermal papilla and pulp of the feather) and the time at which the inductive influence becomes effective (compare Sengel, 1958).

In the present experiments there is no loss of the inductive stimulus from mid-dorsal dermis between the ages of 7 and days (95 cases), as reported by Sengel (1958) working with recombinations grown in vitro. Sengel’ s conclusion, that the inductive action of mid-dorsal dermis is of brief duration (beginning between and 7 days and ceasing to exist between 7 and days), is based entirely on negative results from a total of eight explants. The possibility, therefore, that unfavorable environmental conditions, rather than the actual loss of inductive capacity, might cause the observed absence of growth and differentiation in explants, led the present writer to repeat these particular experiments, using the chorioallantoic membrane as a transplantation site. Eight recombinations were made, therefore, in which epidermis from the mid-dorsum of 5-to -day embryos was placed over dermis from the mid-dorsum of 7-to -day embryos. Normal down feathers developed from the surface of the grafts in every case. The graft shown in Plate 6, fig. M, is a representative example. Experiments were also made in which mid-dorsal dermis of the same stages (7 to days) was combined with epidermis from the lateral apteria (featherless areas) of embryos of 7 to days. Again, the mid-dorsal dermis induced normal feathers in the overlying epidermis of the grafts (ten cases). The fact that normal down feathers developed consistently in overlying epidermis from all three different regions tested (tarsometatarsus, mid-dorsum and apteria) when brought into contact with dermis of mid-dorsal origin, between the stages of 7 and days, would seem sufficient proof that the mid-dorsal dermis at this period has, in-deed, not lost its inductive capacity.

Since the dermis of the spur and the mid-dorsum are similar in that they are both strongly inductive at early developmental stages and give no evidence of change within the wide age span tested, it became important to test the induc-tive capacity of dermis from yet another area of the body. The beak was chosen for the reason that it represents a highly specialized epidermal derivative of an entirely different type.

Experiments were performed in which dermis from the beak of 5-to -day embryos was combined with: (a) epidermis from the tarsometatarsus (11 and 13 days), and (b) epidermis from the mid-dorsum (5 to days). Since the complete data from these experiments will be reported in another communication, it will suffice here to state briefly that, without exception, a normal beak was induced in the overlying epidermis, irrespective of the age or origin of this epidermis (Plate 6, figs. J, K, L). No change in the effectiveness or intensity of the inductive stimulus from the beak dermis was detected between the earliest (5 days) and the latest (days) developmental stages used. In this respect,the dermis of the beak is like that of the mid-dorsum and spur and unlike the dermis of the foot.

The beak dermis appears, then, to be an extremely strong inductor. It is capable, for instance, of inducing a perfect beak from 8-to -day mid-dorsal epidermis which, even in contact with 13-and 15-day foot dermis, remains refractory to the scale-inducing stimulus and continues to develop in a feather direction. Moreover, dermis from the beak induces a perfect beak in 13-day foot epidermis, which under the influence of dermis from the mid-dorsum was no longer completely responsive but, instead, gave rise to abnormal feathers.

The experimental findings, therefore, with respect to the inductive capacity of dermis from the various regions of the body tested, suggest that the several types of dermis might be ranked as follows: (1) dermis from the foot (tarso-metatarsus), weakest; (2) dermis from the mid-dorsum, stronger; and (3) dermis from the spur and beak, strongest.

Competence of the epidermis

The ability of the epidermis to react to specific stimuli of the dermis (inductor) changes very little within the range of developmental stages used in the present experiments. No alteration in competence of epidermis from the mid-dorsum can, in fact, be demonstrated until it is tested in combination with foot dermis that has finally developed the capacity to induce scale formation (13-15 days). When this is done, a distinct change in competence is observed within a brief interval of time between approximately and days. At this time, the 8-to -day epidermis, unlike that of 5 to days, does not produce typical scales, but continues to differentiate in a feather direction (Text-fig. 3). At this rather late stage, when the feather primordia are distinctly visible extern-ally on the surface of the mid-dorsum as dome-like elevations, the epidermis appears to have become so strongly ‘fixed’ for feather formation that its course of differentiation cannot readily be changed. That the response of this 8-to -day epidermis is modified, however, by the scale-inducing dermis of the 13-to 15-day foot, is evident in the properties of the resulting feathers. They exhibit varying degrees of structural aberration not observed when this same epidermis developed in contact with 9-to 12-day foot dermis that had not yet acquired effective inductive strength.

Epidermis from the foot (tarsometatarsus) exhibits even less change with age, in response to underlying dermis of mid-dorsal origin. Between the ages of 9 and 13 days the epidermal response is entirely in a feather-forming direction (Text-fig. 4). Scales could not be identified in any of the grafts, in toto or nr histological section. There is a notable difference, however, in the structural quality of the resulting feathers; those arising from the earliest tarsometatarsal epidermis tested (9–11 days) are, as a rule, structurally normal; those from the 12-day epidermis often show varying degrees of aberrance, while those from the 13-day epidermis are, with few exceptions, very aberrant.

The age differences noted above in the response of epidermis from the mid-dorsum and from the foot can be either erased or completely concealed if each is placed in contact with dermis from the beak, a stronger inductor. In response to beak dermis (5-days), a normal beak is formed from both 8-to -day mid-dorsal epidermis and from 13-day foot epidermis (Plate 6, fig. K).

Since a strong inductor, such as the dermis of the beak, can either mask or erase existing age differences in epidermis that show up only when inductors of lesser strength are employed, the question arises as to whether other differ-ences residing in the epidermis might also be revealed under the influence of a ‘weak’ inductor, such as the dermis of the foot. Results obtained from com-bining epidermis of the same age (5- days) from three entirely different regions of the body, mid-dorsum, beak and lateral apteria, with dermis from the 13-day foot, have shown that, indeed, other differences can be detected. Scales, for instance, are induced in the overlying epidermis of all of these com-binations, but the number of feathers developing in association with the scales varies greatly. Scales induced in both beak and apterous epidermis bear only an occasional feather (Plate 6, figs. H, I), as found on the normal foot; scales induced in mid-dorsal epidermis bear more feathers than found on the normal foot (Plate 4, figs. E, F). Thus, the prospective epidermis from the mid-dorsum possesses propensities for feather formation that appear to be absent in epi-dermis of both the beak and the apteria.

The diverse epidermal specializations (feathers, scales, beaks, spurs, etc.) that develop from the skin of defined regions of the body of birds are all the final products of interaction of ectoderm and mesoderm, two tissues of different embryonic origin. From the results of the present experiments and those of others along similar lines (Cairns & Saunders, 1954; Sengel, 1958; Gomot, 1959), it is evident that the ectoderm (prospective epidermis) acquires its specific properties through reaction to inductive stimuli arising in the underlying meso-derm (prospective dermis). Thus, the reaction is a mutual one and dependent upon the properties of both epidermis and dermis. During this interaction, many different factors are brought into play, variation in any one of which may have a pronounced effect on the final product. In the present experiments, it has been shown that both the inductive capacity of the dermis and the response of the epidermis vary with developmental stage and with location on the body. These differences in the reacting tissues, relating to age and region, can be demonstrated in terms of the final product, only after exchanges are made between the two tissue components. It is not possible to predict the results of untried recombinations. A case in point is the experiment in which epidermis from the mid-dorsum of 8-to -day embryos was combined with dermis from the tarsometatarsus of 13-day embryos. On the basis of the results of this particular recombination alone, it would have been concluded that the epidermis at this late stage is ‘fixed’ or determined for feather formation, since feathers rather than scales developed from the grafts. However, when this same epidermis is brought into contact with dermis from a different region, such as the beak, it now responds by forming a normal, featherless beak. It is important, therefore, to test the same epidermis with dermis of different origin and age (and vice versa) before drawing a hard and fast conclusion concerning the potentialities of either tissue.

The inductive capacity of dermis from the different regions of the body tested—tarsometatarsus, mid-dorsum, beak, spur—has been compared in terms of relative strength, i.e., intensity of the dermal stimulus. Dermis from the tarsometatarsus appears to be not only the ‘weakest’, but also the latest in ontogeny to acquire demonstrable inductive properties. Not until approxi-mately the 13th day does it reach a degree of intensity sufficiently high to elicit a specific scale-forming response in overlying epidermis. Dermis from the spur primordium is strongly inductive at 9 days; dermis from the beak and mid-dorsum at 5 days, the earliest stages tested. In these regions the dermis has, undoubtedly, also gradually acquired its specific inductive capacity. Induction is not to be thought of as a single, sudden event but rather as a sequence of consecutive interactions which may extend over varying periods of time. It is likely that ectodermal-mesodermal interaction begins to take place when the two layers are first brought into contact in the early blastoderm.

The dermis of the foot (tarsometatarsus) is unique in that it induces both scales and feathers. In normal development, feathers are often found in asso-ciation with scales, particularly in the earlier stages of scale formation. Simi-larly, in the grafts, feathers are nearly always found in association with those scales induced in ‘foreign’ epidermis as a result of contact relationship with foot dermis. It is evident, therefore, that the tarsometatarsal dermis is of bi-potential character, capable of transmitting both stimuli, feather-forming and scale-forming, one of which dominates. However, the possibility cannot be dismissed that the epidermis of birds possesses inherent feather-forming ten-dencies which in certain regions become modified secondarily by the under-lying dermis. Since the dermis of the foot region appears to be a relatively ‘weak’ inductor and, furthermore, acquires its inductive capacity late in onto-geny, it may not be capable of ‘erasing’ this inherent feather-forming capacity of the epidermis.

The remarkable ability of epidermis to alter its course of differentiation at relatively late developmental stages, suggests that the morphological similarities between the scale and the feather may be more than superficial. It is possible that relatively few metabolic differences exist between them, thus permitting one to be converted into the other with relative ease. Nor is it impossible that these differences are of a quantitative rather than a qualitative nature.

Etude des interactions tissulaires dans Vécaille et la plume au moyen de recombinaisons dermoépidermiques

  1. Le présent travail est basé sur une analyse expérimentale des interactions de l’ épiderme et du derme dans la formation de l’ écaille et de son homologue, la plume. La méthode utilisée consiste en l’ échange réciproque de derme et d’ épiderme des régions présomptives porteuses d’ écailles et de plumes (tarso-métatarse et region médio-dorsale) à des stades précédant et accompagnant la formation des ébauches. Les tissus recombinés ont été cultivés sur membrane chorio-allantoïdienne.

  2. Les résultats de l’ examen de plus de 500 greffes ont montré que le pouvoir inducteur du derme et la compétence de l’ épiderme varient tous deux avec le stade de développement et avec l’ emplacement sur le corps.

  3. Le derme du tarsométatarse acquiert tardivement son pouvoir inducteur spécifique au cours de l’ ontogenèse; c’ est seulement vers le 13e jour que le stimulus inducteur acquiert un niveau d’ intensité assez élevé pour provoquer la formation d’ écailles dans l’ épiderme ‘étranger’ sus-jacent. Des plumes se développent régulièrement en association avec les écailles induites dans les greffons.

  4. Le derme de la région présomptive de l’ ergot du tarsométatarse est ex-ceptionnel par le fait qu’ il possède un pouvoir inducteur intense au 9e jour, au stade le plus précocement testé. L’ écaille de l’ ergot induit dans l’ épiderme ‘étranger’ du greffon est toujours exempte de plumes associées, comme chez le poulet normal.

  5. Le derme médio-dorsal, comme celui de la région de l’ éperon, est fortement inducteur au stade le plus précoce testé (5 jours) et continue à l’ être fortement pendant toute la période considérée (5 a 8 jours ). Le pouvoir inducteur de ce derme inducteur de plumes n’ est pas de courte durée.

  6. La compétence de l’ épiderme médio-dorsal se modifie pendant un court laps de temps, entre 7 jours et 8 jours Cette modification ne peut être décelée que lorsque l’ épiderme est mis en contact avec du derme tarso-méta-tarsien qui a acquis son pouvoir inducteur spécifique (13 – 15 jours). A ce moment, l’ épiderme le plus âgé (8 – 8 jours ), à la différence de celui des stades plus précoces, ne produit pas d’ écailles typiques mais continue à se différencierdans le sens plumaire. Les plumes, néanmoins, ne sont pas entièrement nor-males.

  7. L’ épiderme du tarsométatarse présente une modification de sa compétence entre 12 et 13 jours. Ce changement ne peut être décelé que dans la structure aberrante des plumes résultantes. Celles produites à partir d’ épiderme de 13 jours sont extrêmement aberrantes. (11 ne se forme jamais d’ écailles à partir de l’ épiderme tarso-métatarsien en contact avec du derme médio-dorsal, quel que soit le stade de chacun des deux constituants tissulaires.)

  8. La modification de la compétance avec l’ âge, notée ci-dessus pour l’ épi-derme à la fois médio-dorsal et tarso-métatarsien, peut être entièrement masquée si chacun est mis en contact avec un inducteur plus fort, tel que le derme du bec. Le derme du bec (5–8 jours ) induit un bec parfait à la fois dans l’ épi-derme médio-dorsal de 8 à 8 jours ; et l’ épiderme tarso-métatarsien de 13 jours.

  9. L’ aptitude remarquable de l’ épiderme à modifier le cours de sa différencia-tion à des stades relativement tardifs suggère l’ idée que les similitudes morphol-ogiques entre l’ écaille et la plume peuvent être plus que superficielles.

  10. Comme le derme tarso-métatarsien est capable d’ induire à la fois des écailles et des plumes chez le poulet normal et dans les greffons, il s’ avère être bipotentiel. Il est possible, néanmoins, que l’ épiderme des Oiseaux possède des tendances propres à la formation des plumes qui, dans certaines régions, se trouvent secondairement modifiées par le derme sous-jacent.

I wish to acknowledge, with gratitude, my indebtedness to my husband, Mr John S. Spurbeck, for the superb photographs and other illustrative material, prepared with his usual skill and care. I wish also to thank Professor Dietrich Bodenstein and Doctor Irwin R. Konigsberg for many helpful suggestions.

Blaszyk
,
P.
(
1935
).
Untersuchungen Uber die Stammesgeschichte der Vogelschuppen und Federn und Uber die Abhangigkeit ihrer Ausbildung am Vogelfuss von der Funktion
.
Gegenbaurs Jb
.
75
,
483
567
.
Bornstein
,
F.
(
1911
).
Ueber Regeneration der Federn und Beziehungen zwischen Federn und Schuppen
.
Arch. Naturgesch
.
77
,
1
11
.
Cairns
,
J. M.
&
Saunders
,
J. W.
, Jr
. (
1954
).
The influence of embryonic mesoderm on the regional specification of epidermal derivatives in the chick
.
J. exp. Zool
.
127
,
221
48
.
Davies
,
H. R.
(
1889
).
Die Entwicklung der Feder und ihre Beziehungen zu anderen Integumentgebilden
.
Gegenbaurs Jb
.
15
,
560
649
.
Gomot
,
L.
(
1959
).
Contribution a l’ étude du développement embryonnaire de la glande uropygienne chez le canard
.
Arch. Anat. micr. Morph, exp
.
48
,
63
141
.
Jeffries
,
J. A.
(
1883
).
The epidermal system of birds
.
Proc. Boston Soc. nat. Hist
.
22
,
20341
.
Kerbert
,
C.
(
1877
).
Ueber die Haut der Reptilien und anderer Wirbelthiere
.
Arch. mikr. Anat
.
13
,
205
62
.
Mcloughlin
,
C. B.
(
1961
).
The importance of mesenchymal factors in the differentiation of chick epidermis. II. Modification of epidermal differentiation by contact with different types of mesenchyme
.
J. Embryol. exp. Morph
.
9
,
385
408
.
Medawar
,
P. B.
(
1941
).
Sheets of pure epidermal epithelium from human skin
.
Nature, Lond
.
148
,
783
.
Rawles
,
M. E.
(
1955
).
Pigmentation in autoplastic and homoplastic grafts of skin from fetal and newborn hooded rats
.
Amer. J. Anat
.
97
,
79
128
.
Rawles
,
M. E.
(
1959
).
An experimental study on the development of regional variation in the plumage pattern of the Silver Campine fowl
.
J. Morph
.
105
,
33
54
.
Rawles
,
M. E.
(
1960
).
The integumentary system
.
In Biology and Comparative Physiology of Birds
, Vol.
1
, pp.
189
240
.
New York and London
:
Academic Press
.
Saunders
,
J. W.
, Jr
.,
Cairns
,
J. M.
&
Gasseling
,
M. T.
(
1957
).
The role of the apical ridge of ectoderm in the differentiation of the morphological structure and inductive specificity of limb parts in the chick
.
J. Morph
.
101
,
57
87
.
Saunders
,
J. W.
, Jr
. (
1958
).
Inductive specificity in the origin of integumentary derivatives in the fowl
.
In The Chemical Basis of Development
, pp.
239
53
.
Baltimore
:
The Johns Hopkins Press
.
Sengel
,
P.
(
1958
).
Recherches expérimentales sur la différenciation des germes plumaires et du pigment de la peau de l’ embryon de poulet en culture in vitro
.
Ann. Sci. Nat., Zool
.
20
,
432
514
.
Szabo
,
G.
(
1955
).
A modification of the technique of ‘skin splitting’ with trypsin
.
J. Path. Bad
.
70
,
545
.
Weiss
,
P.
&
Andres
,
G.
(
1952
).
Experiments on the fate of embryonic cells (chick) disseminated by the vascular route
.
J. exp. Zool
.
121
,
449
87
.
Wessells
,
N. K.
(
1961
).
An analysis of chick epidermal differentiation in situ and in vitro in chemically defined media
.
Dev. Biol
.
3
,
355
89
.
PLATE 1

Figs. A-B. Longitudinal sections through skin of tarsometatarsus (distal) of normal 11-day embryo; posterior face (A), anterior face (B). Scale primordia visible on anterior surface as low elevations, x 100.

Figs. C-D. Longitudinal sections through skin of tarsometatarsus (distal) of normal 13-day embryo; posterior face (C), anterior face (D). Note marked difference in size and height of scale primordia of the two opposing surfaces. (See text.) x 100.

Figs. E-H. Transverse sections through skin of mid-dorsum of normal embryos of 5, 6, 7, and 8 days, respectively. Note changes in epidermis and dermis leading to formation of feather primordium as a domelike elevation on surface of skin (H). (See text.) x 180.

Figs. I-K. Side view of tarsometatarsus of normal Silver Campine (I), and White Leghorn (J and K), embryos (near hatching), showing scales bearing normal and rudimentary down feathers. Scales of anterior face are largest and overlap distally. Smaller median scales of opposite posterior face overlap proximally, x 7.

Figs. L-M. Surface view of skin removed from posterior proximal portion of tarsometa-tarsus of 18-day White Leghorn (L), and Silver Campine (M), embryos showing feathers developing from many smaller scales. Note that feathers point in proximal direction (towards top of figure), x 7.

PLATE 1

Figs. A-B. Longitudinal sections through skin of tarsometatarsus (distal) of normal 11-day embryo; posterior face (A), anterior face (B). Scale primordia visible on anterior surface as low elevations, x 100.

Figs. C-D. Longitudinal sections through skin of tarsometatarsus (distal) of normal 13-day embryo; posterior face (C), anterior face (D). Note marked difference in size and height of scale primordia of the two opposing surfaces. (See text.) x 100.

Figs. E-H. Transverse sections through skin of mid-dorsum of normal embryos of 5, 6, 7, and 8 days, respectively. Note changes in epidermis and dermis leading to formation of feather primordium as a domelike elevation on surface of skin (H). (See text.) x 180.

Figs. I-K. Side view of tarsometatarsus of normal Silver Campine (I), and White Leghorn (J and K), embryos (near hatching), showing scales bearing normal and rudimentary down feathers. Scales of anterior face are largest and overlap distally. Smaller median scales of opposite posterior face overlap proximally, x 7.

Figs. L-M. Surface view of skin removed from posterior proximal portion of tarsometa-tarsus of 18-day White Leghorn (L), and Silver Campine (M), embryos showing feathers developing from many smaller scales. Note that feathers point in proximal direction (towards top of figure), x 7.

PLATE 2

Fig. A. Graft showing normal down feathers produced from 5-day mid-dorsal epidermis in contact with dermis from 9-day tarsometatarsus (anterior, proximal), after 8 days on chorioallantoic membrane (CAM), x 15.

Fig. B. Graft showing normal down feathers produced from 7-day mid-dorsal epidermis in contact with dermis from 1012-day tarsometatarsus (anterior, distal), after 8 days on CAM. xl5.

Fig. C. Graft showing normal down feathers produced from 6-day mid-dorsal epidermis in contact with dermis from 10-day tarsometatarsus (anterior, distal), after 712 days on CAM. xl5.

Fig. D. Graft showing normal down feathers produced from 8-day mid-dorsal epidermis in contact with dermis from 11-day tarsometatarsus (mid-posterior), after 9 days on CAM. x 12.

Fig. E. Graft showing normal down feathers produced from 812-day mid-dorsal epidermis in contact with dermis from 10-day tarsometatarsus (posterior, proximal), after 8 days on CAM. xl5.

Fig. F. Graft showing normal down feathers produced from 6-day mid-dorsal epidermis in contact with dermis from 11-day tarsometatarsus (mid-posterior), after 7 days on CAM. xl5.

Fig. G. Normal down feather from graft of same epidermal-dermal combination shown in Fig. C, above. Feather sheath removed, feather parts allowed to dry. Note numerous barbs, each bearing two rows of barbules. x 15.

Fig. H. Graft showing feathers produced from 512-day mid-dorsal epidermis in contact with dermis from 10-day tarsometatarsus (mid-posterior), after 8 days on CAM. Graft was not kept ‘submerged’ . Short, thick feathers show effect of drying, x 15.

Fig. I. Graft showing feathers affected by drying. Compare feathers with those of graft shown in Fig. C, above, of same epidermal-dermal combination but ‘submerged’ during development, x 15.

PLATE 2

Fig. A. Graft showing normal down feathers produced from 5-day mid-dorsal epidermis in contact with dermis from 9-day tarsometatarsus (anterior, proximal), after 8 days on chorioallantoic membrane (CAM), x 15.

Fig. B. Graft showing normal down feathers produced from 7-day mid-dorsal epidermis in contact with dermis from 1012-day tarsometatarsus (anterior, distal), after 8 days on CAM. xl5.

Fig. C. Graft showing normal down feathers produced from 6-day mid-dorsal epidermis in contact with dermis from 10-day tarsometatarsus (anterior, distal), after 712 days on CAM. xl5.

Fig. D. Graft showing normal down feathers produced from 8-day mid-dorsal epidermis in contact with dermis from 11-day tarsometatarsus (mid-posterior), after 9 days on CAM. x 12.

Fig. E. Graft showing normal down feathers produced from 812-day mid-dorsal epidermis in contact with dermis from 10-day tarsometatarsus (posterior, proximal), after 8 days on CAM. xl5.

Fig. F. Graft showing normal down feathers produced from 6-day mid-dorsal epidermis in contact with dermis from 11-day tarsometatarsus (mid-posterior), after 7 days on CAM. xl5.

Fig. G. Normal down feather from graft of same epidermal-dermal combination shown in Fig. C, above. Feather sheath removed, feather parts allowed to dry. Note numerous barbs, each bearing two rows of barbules. x 15.

Fig. H. Graft showing feathers produced from 512-day mid-dorsal epidermis in contact with dermis from 10-day tarsometatarsus (mid-posterior), after 8 days on CAM. Graft was not kept ‘submerged’ . Short, thick feathers show effect of drying, x 15.

Fig. I. Graft showing feathers affected by drying. Compare feathers with those of graft shown in Fig. C, above, of same epidermal-dermal combination but ‘submerged’ during development, x 15.

PLATE 3

Fig. A. Graft showing scales and associated feathers produced from 512-day mid-dorsal epidermis in contact with dermis from 12-day tarsometatarsus (mid-anterior), after 9 days on chorioallantoic membrane (CAM), x 15.

Fig. B. Graft showing feathers produced from 5-day mid-dorsal epidermis in contact with dermis from 12-day tarsometatarsus (mid-posterior), after 9 days on CAM. x 15.

Fig. C. Graft showing feathers produced from 6-day mid-dorsal epidermis in contact with dermis from 12-day tarsometatarsus (mid-posterior), after 8 days on CAM. x 15.

Fig. D. Graft showing scales and associated feathers produced from 6-day mid-dorsal epidermis (White Leghorn) in contact with Campine dermis from 12-day tarsometatarsus (anterior, distal), after 9 days on CAM. x 15.

Fig. E. Graft showing feathers produced from 7-day mid-dorsal epidermis in contact with 12-day dermis from tarsometatarsus (posterior, distal), after 9 days on CAM. x 15.

Fig. F. Grafts showing feathers produced from 6-day mid-dorsal epidermis in contact with dermis from 12-day tarsometatarsus (posterior, distal), after 8 days on CAM. x 15.

Fig. G. Graft showing feathers produced from 712-day mid-dorsal epidermis (White Leghorn) in contact with 12-day dermis (White Leghorn) from tarsometatarsus (mid-anterior), after 9 days on CAM. x 15.

Fig. H. Graft showing feathers produced from 8-day mid-dorsal epidermis in contact with 12-day dermis from tarsometatarsus (mid-anterior), after 8 days on CAM. x 15.

Fig. L Graft showing normal feathers produced from 8-day mid-dorsal epidermis in con-tact with 12-day dermis from tarsometatarsus (posterior, distal), after 9 days on CAM. x 15.

Fig. J. Graft showing feathers produced from 812-day mid-dorsal epidermis in contact with 12-day dermis from tarsometatarsus (anterior, distal), after 9 days on CAM. x 15.

PLATE 3

Fig. A. Graft showing scales and associated feathers produced from 512-day mid-dorsal epidermis in contact with dermis from 12-day tarsometatarsus (mid-anterior), after 9 days on chorioallantoic membrane (CAM), x 15.

Fig. B. Graft showing feathers produced from 5-day mid-dorsal epidermis in contact with dermis from 12-day tarsometatarsus (mid-posterior), after 9 days on CAM. x 15.

Fig. C. Graft showing feathers produced from 6-day mid-dorsal epidermis in contact with dermis from 12-day tarsometatarsus (mid-posterior), after 8 days on CAM. x 15.

Fig. D. Graft showing scales and associated feathers produced from 6-day mid-dorsal epidermis (White Leghorn) in contact with Campine dermis from 12-day tarsometatarsus (anterior, distal), after 9 days on CAM. x 15.

Fig. E. Graft showing feathers produced from 7-day mid-dorsal epidermis in contact with 12-day dermis from tarsometatarsus (posterior, distal), after 9 days on CAM. x 15.

Fig. F. Grafts showing feathers produced from 6-day mid-dorsal epidermis in contact with dermis from 12-day tarsometatarsus (posterior, distal), after 8 days on CAM. x 15.

Fig. G. Graft showing feathers produced from 712-day mid-dorsal epidermis (White Leghorn) in contact with 12-day dermis (White Leghorn) from tarsometatarsus (mid-anterior), after 9 days on CAM. x 15.

Fig. H. Graft showing feathers produced from 8-day mid-dorsal epidermis in contact with 12-day dermis from tarsometatarsus (mid-anterior), after 8 days on CAM. x 15.

Fig. L Graft showing normal feathers produced from 8-day mid-dorsal epidermis in con-tact with 12-day dermis from tarsometatarsus (posterior, distal), after 9 days on CAM. x 15.

Fig. J. Graft showing feathers produced from 812-day mid-dorsal epidermis in contact with 12-day dermis from tarsometatarsus (anterior, distal), after 9 days on CAM. x 15.

PLATE 4

Fig. A. Graft showing scales and feathers, produced from 5-day mid-dorsal epidermis in contact with 15-day dermis from tarsometatarsus (anterior, distal), after 9 days on chorioallantoic membrane (CAM), x 15.

Fig. B. Graft showing scales and associated feathers, produced from 612-day mid-dorsal epidermis in contact with 13-day dermis from tarsometatarsus (anterior, distal portion), after 8 days on CAM. x 15.

Fig. C. Graft showing scales with exceptionally few associated feathers produced from 7-day mid-dorsal epidermis in contact with 13-day dermis from tarsometatarsus (anterior, distal), after 9 days on CAM. x 15.

Fig. D. Graft showing scales and feathers produced from 6-day mid-dorsal epidermis in contact with 14-day dermis from tarsometatarsus (posterior, distal), after 812 days on CAM. x 15.

Fig. E. Graft showing scales and feathers produced from 612-day mid-dorsal epidermis in contact with 13-day dermis from tarsometatarsus (mid-posterior region), after 8 days on CAM. xl5.

Fig. F. Graft showing scales with associated feathers, produced from 7-day mid-dorsal epidermis (White Leghorn) in contact with 13-day dermis (White Leghorn) from tarsometa-tarsus (posterior, proximal), after 812 days on CAM. x 15.

Fig. G. Graft showing normal spur scale (arrow), and surrounding feathers produced from 5-day mid-dorsal epidermis in contact with dermis from 13-day tarsometatarsus (spur and surrounding region), after 8 days on CAM. x 15.

Fig. H. Graft showing normal spur scale (arrow), and feathers produced from 8-day mid-dorsal epidermis (White Leghorn) in contact with dermis from 13-day (White Leghorn) tarsometatarsus (spur and surrounding region), after 8 days on CAM. x 15.

Fig. I. Graft showing feathers produced from 8-day mid-dorsal epidermis in contact with dermis from 13-day tarsometatarsus (posterior, proximal), after 7 days on CAM. X 15.

Fig. J. Graft showing feathers produced from 812-day mid-dorsal epidermis (White Leghorn) in contact with dermis from 13-day, White Leghorn, tarsometatarsus (mid-pos-terior), after 8 days on CAM. x 15.

Fig. K. Graft showing feathers and aberrant scale formation produced from 8-day-mid-dorsal epidermis in contact with dermis from 13-day tarsometatarsus (mid-posterior), after 9 days on CAM. x 15.

Fig. L. Graft showing feathers and aberrant scale formation produced from 8-day mid-dorsal epidermis in contact with dermis from 15-day tarsometatarsus (anterior, distal), after 7 days on CAM. Graft developed tightly pressed against egg shell. (See text.) x 15.

Fig. M. Graft showing feathers produced from 812-day mid-dorsal epidermis (White Leghorn) in contact with dermis from 13-day (White Leghorn) tarsometatarsus (Anterior, distal), after 8 days on CAM. xl5.

Fig. N. Graft showing relatively normal feathers produced from 812-day mid-dorsal epidermis in contact with dermis from 14-day tarsometatarsus (mid-posterior), after 412 days on CAM. x 15.

Fig. O. Graft showing many relatively normal feathers, produced from 812-day mid-dorsal epidermis in contact with 14-day dermis from tarsometatarsus (anterior, distal), after 712 days on CAM. x 15.

PLATE 4

Fig. A. Graft showing scales and feathers, produced from 5-day mid-dorsal epidermis in contact with 15-day dermis from tarsometatarsus (anterior, distal), after 9 days on chorioallantoic membrane (CAM), x 15.

Fig. B. Graft showing scales and associated feathers, produced from 612-day mid-dorsal epidermis in contact with 13-day dermis from tarsometatarsus (anterior, distal portion), after 8 days on CAM. x 15.

Fig. C. Graft showing scales with exceptionally few associated feathers produced from 7-day mid-dorsal epidermis in contact with 13-day dermis from tarsometatarsus (anterior, distal), after 9 days on CAM. x 15.

Fig. D. Graft showing scales and feathers produced from 6-day mid-dorsal epidermis in contact with 14-day dermis from tarsometatarsus (posterior, distal), after 812 days on CAM. x 15.

Fig. E. Graft showing scales and feathers produced from 612-day mid-dorsal epidermis in contact with 13-day dermis from tarsometatarsus (mid-posterior region), after 8 days on CAM. xl5.

Fig. F. Graft showing scales with associated feathers, produced from 7-day mid-dorsal epidermis (White Leghorn) in contact with 13-day dermis (White Leghorn) from tarsometa-tarsus (posterior, proximal), after 812 days on CAM. x 15.

Fig. G. Graft showing normal spur scale (arrow), and surrounding feathers produced from 5-day mid-dorsal epidermis in contact with dermis from 13-day tarsometatarsus (spur and surrounding region), after 8 days on CAM. x 15.

Fig. H. Graft showing normal spur scale (arrow), and feathers produced from 8-day mid-dorsal epidermis (White Leghorn) in contact with dermis from 13-day (White Leghorn) tarsometatarsus (spur and surrounding region), after 8 days on CAM. x 15.

Fig. I. Graft showing feathers produced from 8-day mid-dorsal epidermis in contact with dermis from 13-day tarsometatarsus (posterior, proximal), after 7 days on CAM. X 15.

Fig. J. Graft showing feathers produced from 812-day mid-dorsal epidermis (White Leghorn) in contact with dermis from 13-day, White Leghorn, tarsometatarsus (mid-pos-terior), after 8 days on CAM. x 15.

Fig. K. Graft showing feathers and aberrant scale formation produced from 8-day-mid-dorsal epidermis in contact with dermis from 13-day tarsometatarsus (mid-posterior), after 9 days on CAM. x 15.

Fig. L. Graft showing feathers and aberrant scale formation produced from 8-day mid-dorsal epidermis in contact with dermis from 15-day tarsometatarsus (anterior, distal), after 7 days on CAM. Graft developed tightly pressed against egg shell. (See text.) x 15.

Fig. M. Graft showing feathers produced from 812-day mid-dorsal epidermis (White Leghorn) in contact with dermis from 13-day (White Leghorn) tarsometatarsus (Anterior, distal), after 8 days on CAM. xl5.

Fig. N. Graft showing relatively normal feathers produced from 812-day mid-dorsal epidermis in contact with dermis from 14-day tarsometatarsus (mid-posterior), after 412 days on CAM. x 15.

Fig. O. Graft showing many relatively normal feathers, produced from 812-day mid-dorsal epidermis in contact with 14-day dermis from tarsometatarsus (anterior, distal), after 712 days on CAM. x 15.

PLATE 5

Fig. A. Graft showing normal down feathers produced from epidermis of 11-day tarso-metatarsus (anterior, distal) in contact with 7-day mid-dorsal dermis, after 812 days on chorio-allantoic membrane (CAM). xl5.

Fig. B. Graft showing normal down feathers produced from epidermis of 11-day tarso-metatarsus (posterior, distal) in contact with 8-day mid-dorsal dermis, after 712 days on CAM. xl5.

Fig. C. Graft showing normal down feathers, produced from epidermis of 10-day tarso-metatarsus (posterior, proximal) in contact with 5512-day mid-dorsal dermis, after 8 days on CAM. x 15.

Fig. D. Graft showing normal down feathers, produced from epidermis of 101012-day tarsometatarsus (anterior, distal) in contact with 612-day mid-dorsal dermis, after 7 days on CAM. Feathers forced apart by expansion of underlying vesicle, x 15.

Fig. E. Graft showing normal down feathers produced from epidermis of 11-day tarso-metatarsus (mid-anterior), in contact with 6-day mid-dorsal dermis, after 812 days on CAM. x15.

Fig. F. Graft showing normal down feathers produced from epidermis of 10-day tarso-metatarsus (anterior, distal), in contact with 6-day mid-dorsal dermis, after 7 days on CAM. x 15.

Fig. G. Graft showing feathers produced from epidermis of 12-day tarsometatarsus (anterior, distal), in contact with 8-day mid-dorsal dermis, after 8 days on CAM. Graft not kept completely ‘submerged’ . Effect of drying is seen in the two short, thick feathers on upper, right part of graft. xl5.

Fig. H. Graft showing normal down feathers, in early stage of formation, produced from epidermis Of 12-day tarsometatarsus (anterior, distal), in contact with 712-day mid-dorsal dermis, after 4 days on CAM. Epidermis distended by vesicle-formation beneath. (See text.) xl5.

Fig. 1. Graft showing aberrant feathers produced from epidermis of 13-day (White Leg-horn) tarsometatarsus (mid-anterior), in contact with 8-day mid-dorsal epidermis (White Leghorn), after 7days on CAM. x 15.

Fig. J. Graft showing aberrant feathers produced from epidermis of 13-day tarsometa-tarsus (mid-posterior), in contact with 7-day mid-dorsal dermis, after 712 days on CAM. Dark spots on surface are remnants of outer, nonliving, portion of original epidermis, x 15.

Fig. K. Graft showing feathers produced from epidermis of 12-day (White Leghorn) tarsometatarsus (mid-anterior), in contact with 7-day mid-dorsal dermis, after 8 days on CAM. Feathers developed under tension from expansion of enormous vesicle beneath. Crinkling of thin epidermis due to shrinkage of vesicle during fixation. (See text.) x 12.

PLATE 5

Fig. A. Graft showing normal down feathers produced from epidermis of 11-day tarso-metatarsus (anterior, distal) in contact with 7-day mid-dorsal dermis, after 812 days on chorio-allantoic membrane (CAM). xl5.

Fig. B. Graft showing normal down feathers produced from epidermis of 11-day tarso-metatarsus (posterior, distal) in contact with 8-day mid-dorsal dermis, after 712 days on CAM. xl5.

Fig. C. Graft showing normal down feathers, produced from epidermis of 10-day tarso-metatarsus (posterior, proximal) in contact with 5512-day mid-dorsal dermis, after 8 days on CAM. x 15.

Fig. D. Graft showing normal down feathers, produced from epidermis of 101012-day tarsometatarsus (anterior, distal) in contact with 612-day mid-dorsal dermis, after 7 days on CAM. Feathers forced apart by expansion of underlying vesicle, x 15.

Fig. E. Graft showing normal down feathers produced from epidermis of 11-day tarso-metatarsus (mid-anterior), in contact with 6-day mid-dorsal dermis, after 812 days on CAM. x15.

Fig. F. Graft showing normal down feathers produced from epidermis of 10-day tarso-metatarsus (anterior, distal), in contact with 6-day mid-dorsal dermis, after 7 days on CAM. x 15.

Fig. G. Graft showing feathers produced from epidermis of 12-day tarsometatarsus (anterior, distal), in contact with 8-day mid-dorsal dermis, after 8 days on CAM. Graft not kept completely ‘submerged’ . Effect of drying is seen in the two short, thick feathers on upper, right part of graft. xl5.

Fig. H. Graft showing normal down feathers, in early stage of formation, produced from epidermis Of 12-day tarsometatarsus (anterior, distal), in contact with 712-day mid-dorsal dermis, after 4 days on CAM. Epidermis distended by vesicle-formation beneath. (See text.) xl5.

Fig. 1. Graft showing aberrant feathers produced from epidermis of 13-day (White Leg-horn) tarsometatarsus (mid-anterior), in contact with 8-day mid-dorsal epidermis (White Leghorn), after 7days on CAM. x 15.

Fig. J. Graft showing aberrant feathers produced from epidermis of 13-day tarsometa-tarsus (mid-posterior), in contact with 7-day mid-dorsal dermis, after 712 days on CAM. Dark spots on surface are remnants of outer, nonliving, portion of original epidermis, x 15.

Fig. K. Graft showing feathers produced from epidermis of 12-day (White Leghorn) tarsometatarsus (mid-anterior), in contact with 7-day mid-dorsal dermis, after 8 days on CAM. Feathers developed under tension from expansion of enormous vesicle beneath. Crinkling of thin epidermis due to shrinkage of vesicle during fixation. (See text.) x 12.

PLATE 6

Fig. A. Section through graft showing structurally normal down feathers produced from epidermis of 9-day tarsometatarsus (mid-anterior), in contact with 5-day mid-dorsal dermis, after 5 days on chorioallantoic membrane (CAM). Note barb ridges (br), feather pulp (p), and feather sheath (fs), x 60.

Fig. B. Transverse section through normal down feather of graft produced from epi-dermis of 12-day (White Leghorn) tarsometatarsus (mid-anterior), in contact with 612-day mid-dorsal dermis (White Leghorn), after 812days on CAM. Note fully differentiated barbs (b) and barbules (bb), enclosed by sheath (fs). Central pulp (p) contains blood vessels. x210.

Figs. C-G. Sections through aberrant feathers produced from 13-day tarsometatarsal epidermis in contact with mid-dorsal dermis from embryos of 6 days (C), 7 days (D and E), and 8 days (F and G). b = barbs; bb = barbules; br = barb ridges; fs = feather sheath; p = feather pulp; dp = dermal papilla. x210.

Fig. H. Graft showing normal scales and three associated, rudimentary feathers (arrow), produced from epidermis of 6-day beak (White Leghorn), in contact with dermis of 13-day White Leghorn tarsometatarsus (posterior, proximal), after 8 days on CAM. x 15.

Fig. I. Graft showing normal scales and several associated rudimentary feathers (arrows), produced from epidermis of 634-day lateral apterium in contact with dermis of 13-day tarsometatarsus (anterior, distal), after 8 days on CAM. x 15.

Fig. J. Longitudinal section through beak of graft shown in adjacent (lower) Fig. L. Note typical beak epidermis composed of numerous compressed layers of keratin, x 60.

Fig. K. Graft showing perfect beak produced from epidermis of 13-day (White Leghorn) tarsometatarsus (anterior, distal) in contact with 6-day beak dermis (White Leghorn), after 8 days on CAM. x 12.

Fig. L. Graft showing normal beak produced from epidermis of 11-day tarsometatarsus (anterior, distal) in contact with 5-day beak dermis, after 8 days on CAM. x 12.

Fig. M. Graft showing normal down feathers produced from 6-day mid-dorsal epidermis in contact with 8-day mid-dorsal dermis, after 8 days on CAM. x 15.

PLATE 6

Fig. A. Section through graft showing structurally normal down feathers produced from epidermis of 9-day tarsometatarsus (mid-anterior), in contact with 5-day mid-dorsal dermis, after 5 days on chorioallantoic membrane (CAM). Note barb ridges (br), feather pulp (p), and feather sheath (fs), x 60.

Fig. B. Transverse section through normal down feather of graft produced from epi-dermis of 12-day (White Leghorn) tarsometatarsus (mid-anterior), in contact with 612-day mid-dorsal dermis (White Leghorn), after 812days on CAM. Note fully differentiated barbs (b) and barbules (bb), enclosed by sheath (fs). Central pulp (p) contains blood vessels. x210.

Figs. C-G. Sections through aberrant feathers produced from 13-day tarsometatarsal epidermis in contact with mid-dorsal dermis from embryos of 6 days (C), 7 days (D and E), and 8 days (F and G). b = barbs; bb = barbules; br = barb ridges; fs = feather sheath; p = feather pulp; dp = dermal papilla. x210.

Fig. H. Graft showing normal scales and three associated, rudimentary feathers (arrow), produced from epidermis of 6-day beak (White Leghorn), in contact with dermis of 13-day White Leghorn tarsometatarsus (posterior, proximal), after 8 days on CAM. x 15.

Fig. I. Graft showing normal scales and several associated rudimentary feathers (arrows), produced from epidermis of 634-day lateral apterium in contact with dermis of 13-day tarsometatarsus (anterior, distal), after 8 days on CAM. x 15.

Fig. J. Longitudinal section through beak of graft shown in adjacent (lower) Fig. L. Note typical beak epidermis composed of numerous compressed layers of keratin, x 60.

Fig. K. Graft showing perfect beak produced from epidermis of 13-day (White Leghorn) tarsometatarsus (anterior, distal) in contact with 6-day beak dermis (White Leghorn), after 8 days on CAM. x 12.

Fig. L. Graft showing normal beak produced from epidermis of 11-day tarsometatarsus (anterior, distal) in contact with 5-day beak dermis, after 8 days on CAM. x 12.

Fig. M. Graft showing normal down feathers produced from 6-day mid-dorsal epidermis in contact with 8-day mid-dorsal dermis, after 8 days on CAM. x 15.