THE ANATOMICAL RECORD 233:291-300 (1992) Studies on Wound Healing in the Neuroepithelium of the Chick Embryo AARON LAWSON AND MARJORIE A. ENGLAND Department of Anatomy, University of Ghana Medical School, Accra, Ghana (A.L.),and Department of Anatomy, The Medical School, University of Leicester, Leicester, U.K. (M.A.E.) ABSTRACT Wound healing has been studied by light microscopy, SEM, and TEM in the neuroepithelium of the early neurula (stages 6 and 8) and advanced neurula (stages 10 and 12) chick embryos. Healing involves two major events: (1) apposition of the wound edges and (2) restitution of the neuroepithelium at the wound site (i.e., restoration of the epithelial integrity of neuroepithelium). Apposition of the wound edges occurs within the first 15 minutes of re-incubation and involves the entire length of the wound. The main event during restoration is a change in the shapes of the rounded cells to elongated forms (i.e., spindle, wedge, and inverted wedge shapes). Wounds of younger embryos heal faster than those of older ones. o 1992 Wiley-Liss, Inc Wound healing is a series of processes that lead to the restoration of tissue integrity. These processes have been investigated by light microscopy, transmission, and scanning electron microscopy and also by immunofluorescence and cytohistochemical techniques in a number of embryonic and adult epithelia (Croft and Tarin, 1970; Pfister, 1975; Takeuchi, 1976; England and Cowper, 1977; Stanisstreet and Panayi, 1980; Stanisstreet et al., 1980, 1985). In studies of embryonic systems, two layers that have been investigated in detail are the ectoderm, which will later form the neuroectoderm and surface epithelium of the embryo, and the endoderm, which contributes to the gut. In the chick and amphibian embryos studied, early stages were used, i.e., stages 3-5 (Hamburger and Hamilton, 1951) in the chick and 1415 (Nieuwkoop and Faber, 1956) in Xenopus laeuis. In the early neurula stages, these embryonic layers provided a simple model for studying the process of wound healing (England and Cowper, 1977; Mareel and Vakaet, 1977; Stanisstreet et al., 1980). In these simple epithelia, the primary mode of healing for the endoderm layer is cell rearrangement and migration. A similar method of healing occurs in a variety of adult systems including the skin and cornea (Gabbiani et al., 1978; Pang et al., 1978). The ectodermal layer has a more complicated structure than the endodermal layer, as the cells are more elongated and have lots of extracellular spaces. Wounds made in this layer are closed by changes in cell shape in both chick and amphibian embryos. Additionally, some cell proliferation is apparent in the avian embryo. Increased mitosis does not appear to play a crucial role in wound healing in the Xenopus, as embryos treated with colchicine and wounded still possess the ability to heal (Stanisstreet and Panayi, 1980). Wound healing in the neuroepithelium has received relatively little attention. As naturally occurring neural tube defects are believed to be the result of failure of the initial fusion of the neural folds or of a re0 1992 WILEY-LISS, INC opening of the neural tube later in development, the ability of the neural tube t o heal is of great interest. Further, as the neuroepithelium is not fully differentiated in the early embryo, it is likely that the processes involved in both neural tube formation and wound healing may be directed by similar cellular behaviours. Neural tube wound healing may therefore offer a useful model system to study aspects of neurulation that are still not fully understood. Clark and Scothorne (1988,1990) report that in chick embryos in which the closed neural tube roof plate is experimentally incised, there is a sharp decline in healing capacity between stages 14 and 15. In younger chick embryos (stages 12 and 13) similar wounds in the roof plate heal completely. Even when treated in ovo with Streptomyces hyaluronidase immediately after incision of the roof plate, healing occurs completely (Clark, 1987) in this age group. Some early studies (Waddington and Cohen, 1936; Watterson and Fowler, 1953; Kallen, 1955) reported that wounded neural tubes are capable of healing at somite stages but not a t presomite stages. Others (Spratt, 1940; Yntema and Hammond, 1945, Wenger, 1950; Birge and Hilleman, 1953) reported that some healing can occur a t neural plate stages, but none at early somite stages. The present study re-examines the exact process of neuroepithelial wound healing, not only as a sequence of cellular stages in time but also as a means for examining normal neural tube formation. This work incorporates studies of early neurula stages (6 and 8) and advanced neurula stages (10 and 12) in the chick embryo as studied by light microscopy, scanning electron microscopy (SEM), and transmission electron microscopy (TEM). Received July 2, 1991; accepted October 29, 1991. Address reprint requests to Dr. Aaron Lawson, Department of Anatomy, University of Ghana Medical School, P.O. Box 4236, Accra, Ghana, West Africa. A. LAWSON AND M.A. ENGLAND 292 MATERIALS AND METHODS Wound Healing Studies finally examined with a JEOL lOOCX transmission electron microscope. Fertilised eggs from White Leghorn hens were incubated at 38°C to obtain embryos at stages 6, 8, 10, and 12 (Hamburger and Hamilton, 1951). These were mounted by New Culture (New, 1955) and then prepared for wounding as follows. The well, formed by the glass ring, was flooded with saline and the cranial edge of the blastoderm was carefully detached from the vitelline membrane using jeweller’s forceps. The detached part was folded onto the caudal half of the blastoderm to expose the neural plate, groove, or tube. Special care was taken to avoid damage to the underlying vitelline membrane. A single, straight, longitudinal, or transverse wound, approximately 0.2-0.5 mm long, was made on one side of the neural fold or tube in the midbrain region with a cactus needle (England, 1981). For stage 6 and 8 embryos in which neural folds were present, the rostra1 end of the notochord served as a guide t o the midbrain region and wounds were made in the neural folds about halfway between the region of the notochord and the edge of the neural fold. This involved the full thickness of the neuroepithelium (not mesoderm and endoderm as well). In contrast, in stages 10 and 12 embryos in which the neural tube has formed in the midbrain region, the wounds were made through the surface ectoderm into the neural tube. The folded part of the blastoderm was then reflected back onto the vitelline membrane. Control embryos were similarly treated, but their neural folds or tubes were left intact. The wounded embryos were either fixed immediately in half-strength Karnovsky’s (1965) fixative or re-incubated for 15 minutes, 1 hour, 2 hours, or 5 hours (some stage 10 and 12 embryos were re-incubated for 8 hours) before fixation. Following fixation for 2-24 hours, they were washed in 0.2 M sodium cacodylate buffer (pH 7.4) (Plumel, 1948) for an equal period of time, postfixed in 2% cacodylate-buffered osmium tetroxide, and dehydrated in graded ascending series of ethanouwater up to 100%ethanol. All microdissection to expose the neural tube surfaces of control or experimental embryos was done in 70% ethanol. Specimens for light microscopy, which consisted of 50 embryos for each stage, were embedded in araldite and sectioned serially a t 2 pm along the length of the wound. They were then stained with 1%toluidine blue in 1% borax. Those for scanning electron microscopy, which were of the same number for each stage, were transferred from 100%ethanol to 100% acetone, critical point dried by acetone replacement with liquid carbon dioxide, mounted on stubs, and coated with 20 nm of gold. They were finally examined with International Scientific Instruments IS1 60 and DS 130 scanning electron microscopes at 15kv. Forty stage 6 embryos were wounded and processed for transmission electron microscopy. They were processed like those for light microscopy but were fixed with half-strength Karnovsky’s fixative to which had been added l%tannic acid to prevent the leakage of glycosaminoglycans (GAGS)from the extracellular matrix. Following postfixation and dehydration, they were embedded in epon. Ultrathin sections were cut with a diamond knife and stained with uranyl acetate and lead citrate. They were Morphometric Analysis I Slides were selected randomly from a collection of serial sections of wounds of four stage 10 embryos for each of the following re-incubation periods: 0 minute, 15 minutes, 1 hour, 2 hours, or 5 hours. They were coded and the experimenter conducting the analysis was unaware of the significance of the codes. For each embryo and for each re-incubation period, four sections were randomly selected and analyzed. Rounded cells were counted within a 1 mm2 area on each side of the wound using an eye piece graticule a t a total magnification of 400. The criterion used for selecting rounded cells was as follows: the presence of a rounded nucleus with a well-defined nuclear membrane and nucleolus. By this method of selection, rounded cells undergoing mitosis were avoided. Other cells within the area (spindle, wedge, and inverted wedge shape; Schoenwolf and Franks, 1984) were also counted in order to obtain the total number of cells within the area. The numerical density of rounded cells for each treatment group Ke., number of rounded cells per total number of cells counted) was calculated as well as the mean density and standard error of the mean for each time of reincubation. A one-factor analysis of variance (ANOVA) was done using Scheffe F-test at a 95%significance. RESULTS The rate of healing varied from one stage to the next and even among embryos of the same stage. The sequence of events during healing, however, remained the same. Furthermore, within certain limitations the size of the wound did not affect the rate of healing. Wound healing did not alter the pattern of neurulation in the wounded embryos probably because of the rapidity of healing, but the rate at which the latter occurred was slightly slower in some wounded embryos of younger stages than in control embryos of the same stages. The detached anterior edge of the blastoderm, when reflected back onto the vitelline membrane, healed within an hour of re-incubation. Wound Healing Studies Early neurulae-stages 6 and 8 Control. In control embryos, the neural folds formed a thickened pseudostratified columnar epithelium, comprising closely packed neuroepithelial cells. These were elongated apicobasally. The TEM appearance of an unwounded neuroepithelium was similar to that de- B BL CP D EL EM L LU M NU RC Abbreviations bleb basal lamina cell protrusion cell debris elongated cell extracellular material lateral side of neural fold neural fold lumen medial side of neural fold nucleus rounded cell NEUROEPITHELIAL WOUND HEALING Fig. 1. A freshly made stage 6 wound (arrows)in transverse section. x 100. Fig.2. SEM of the apical surface of a freshly made stage 6 wound. x 293 Fig.3.(a)TEM of the edge of a freshly made stage 6 wound. x 1900. (b)Higher magnification of the rounded cell in Figure 3. Note the well-defined nuclear membrane and the intense tannic acid staining of the nuclear envelope. x 14,000. 1000. scribed by Schoenwolf and Franks (1984). Briefly, they reported the presence of wedge-shape cells in the supranotochordal region with their nuclei close to the basal lamina. In other regions, spindle-shape cells were predominant. Zero hour (fresh wounds). A freshly made wound involved the full thickness of the right neural fold and a gap was present between the wound edges (Fig. 1). The wound appeared slit-like or gaped very slightly when viewed from the basal surface of the neural folds by SEM. The basal lamina appeared as a cut sheet at its edges. On the apical surfaces, there were rounded cells a t the wound edges and numerous blebs on the edge cells (Fig. 2). Cell debris was also present. The cells a t the wound edges appeared to have lost their normal configuration in the neuroepithelial sheet and were loose at the edge. Transmission electron microscopy confirmed the presence of rounded cells in the immediate vicinity of the wound and elongated cells away from the wound (Fig. 3a). The rounded cells had welldefined and rounded nuclei with a distinct nuclear membrane and nucleolus. They also had blebs projecting into the wound (Fig. 3b). Extracellular materials were present in the wound as short fibrils and tufts of punctate deposits. The latter also coated the surfaces of the rounded cells and blebs immediately bordering the wound (Fig. 3b). Fifteen minutes. The wound edges became apposed to each other near the apical surface and a depression was present near the basal side (Fig. 4). The full thickness of the neuroepithelium in the wound was decreased when compared with the opposite side. Whereas cells a t the wound edges were rounded, those away from the wound were elongated and showed a normal orientation. By SEM, it was observed that apposition of the wound edges had occurred along the whole length of the wound. The basal surface of the wound was curled inwardly and showed that the depression observed by LM was present along the length of the wound. This made it difficult to identify the cut edges of the basal lamina. The apical surface showed the wound plugged with cell debris, rounded cells and blebs (Fig. 5a,b). Some of the rounded cells were large and projected from the surface. Also, some were wedge shaped with their apices pointing in the direction of the wound. Cell protrusions were seen by TEM at the apical surface. Some of these comprised mainly cell debris with disintegrating cytoplasm and cell membrane, whereas oth- 294 A. LAWSON AND M.A. ENGLAND Fig. 4. A stage 6 wound in transverse section, re-incubated for 15 minutes. Note triangular-shape depression (*) a t the basal end of the wound. x 200. Fig. 5. (a) SEM of the apical surface of a stage 6 wound re-incubated for 15 minutes. Most of the cell debris has been removed to show apposed wound edges. x 1000.(b)Higher magnification of the wound in (a) showing wedge-shape cells (*). x 2500. Fig. 6. TEM of the apical end of a stage 6 wound re-incubated for 15 minutes. Intercellular spaces are absent. x 2900. Fig. 7. TEM of a stage 6 wound re-incubated for 15 minutes showing a rounded cell adjacent to a depression (*). x 7200. ers had contact with the wound and it appeared they were being extruded from it (Fig. 6). The rounded cells in the region where the wound edges were apposed to each other and those adjacent to the depression had rounded nuclei with clearly defined nuclear membranes and nucleoli and the cytoplasm around them was uniform (Figs. 6,7). There was a marked reduction of intercellular spaces around the rounded cells. Cells undergoing mitosis were seen very rarely at the wound NEUROEPITHELIAL WOUND HEALING 295 Fig. 8. TEM of a stage 6 wound re-incubated for 15 minutes showing cell processes in contact with each other at discrete points (thick arrow) within a depression. X 5800. Fig. 9. TEM of a stage 6 wound re-incubated for 15 minutes showing interlocking cell processes (long arrows). x 14,000. Fig. 10. A stage 6 wound in transverse section re-incubated for 1 hour. Arrows indicate wound site. x 200. site. The depression on the basal margin was filled with extracellular materials in the form of short fibrils and granular material. In some regions of the wound, the edges of the depression ran almost parallel with each other and in other regions, cell processes approached each other across the depression (Figs. 7, 8). The cell processes were especially numerous in regions devoid of a basal lamina. When those from the opposite sides came into contact with each other, fusion of the wound occurred across the depression. This process occurred first at discrete points with intervening oval spaces filled with extracellular materials (Fig. 8). The cell processes often seemed to interlock with each other (Fig. 9). One hour. It became more difficult to identify the wound by light microscopy after 1 hour of re-incubation. Either a shallow depression present on its basal surface or cell debris on its apical surface often helped to identify it. In these sections, most of the cells in the wound had become elongated and displayed an apicobasal orientation (Fig. 10). The basal surface of the wound, by SEM, showed ones where the depression previously seen had disappeared (Fig. 11)and these were interspersed among areas where the depression was still present. In the former, there was a uniform basal lamina over the basal surface of the wound implying that healing was completed in these areas. It was normal in appearance when compared with that in control embryos. At the apical surface, the wound edges were more apposed to each other and appeared to be fusing together. Blebs and cell debris were fewer in number and rounded cells were sometimes seen in some areas (Fig. 12a,b). There were other parts of the apical surface that showed complete healing with a restoration of the normal architecture of the neuroepithelium. Sections examined by TEM showed that although a basal lamina being formed over the wound comprised laminae densa and rara, these could not be clearly distinguished from each other. Two hours. Two hours after re-incubation, the wound examined by LM showed the features of that a t 1 hour, but in addition to these the neural folds had become 296 A. LAWSON AND M.A. ENGLAND Fig. 11. SEM of the basal surface of a partially healed stage 6 wound, re-incubated for 1 hour. Arrowheads indicate healed areas and long arrows, unhealed areas. x 1000. Inset shows whole wound. x 300. Fig. 12. (a) SEM of the apical surface of a stage 6 wound re-incubated for 1 hour. x 1000. (b) shows a fracture through the wound. Note the large rounded cells in the wound and the elongated cells adjacent to them. x 4000. NEUROEPITHELIAL WOUND HEALING 297 Fig. 13. TEM of the basal end of a healed stage 6 wound re-incubated for 2 hours. Wound site (arrowheads). x 19000. more elevated, thus deepening the neural groove further, and there was convergence of the folds in the midbrain region. The shallow depression could sometimes be seen on the basal surface. Scanning electron microscopy of the whole extent of the wound at this time revealed parts of the basal surface that still had the depression (i.e., parts that were not completely healed). On the apical surface, however, all parts of the wound had healed and the integrity of this surface of the neuroepithelium had been restored. Transmission electron microscopy also revealed that the elongated cells in the wound had characteristics similar to those away from the wound. Cell debris was scanty a t this time and in some parts of the wound a very shallow depression was present. A uniform basal lamina covered the basal surface of the wound and this was similar to that in adjacent parts of the neuroepithelium. The lamina rara was interposed between the cell membrane and the lamina densa and extracellular materials were present outside the latter (Fig. 13). Five hours. The wound healed completely within 5 hours of re-incubation, and no depression was seen on its basal surface by SEM (Figs. 14, 15). Advanced neurulae-stages 10 and 12 Wound healing. This followed a pattern similar to that of the early neurula stages, but the rate of healing was slower in the advanced embryos than in the younger ones (Fig. 16a-d). Nevertheless, wounds in these embryos often healed by eight hours of re-incubation. Statistical analysis The bar chart (Fig. 17) illustrates the numerical densities of rounded cells in the wound per time of reincubation. By our criteria, no rounded cells were present in the control unwounded neuroepithelium. The densities of these cells increased sharply to about 72% immediately following injury (i.e., in a freshly Fig. 14. SEM of the basal surface of a completely healed stage 6 wound. &-incubation for 5 hours. Arrowheads indicate wound site. x 2000. Fig. 15. SEM of the apical surface of a healed stage 6 wound reincubated for 5 hours. x 1000. made wound) and this level was maintained with some variability by 15mins of re-incubation. By 1 hour the density had significantly decreased to about 26%.Densities thereafter fluctuated insignificantly with respect to that a t 1 hour, although levels were always significantly lower than that in the freshly made wound. DISCUSSION The present study demonstrates that wound healing in the neuroepithelium of the chick embryo involves two events. The first comprises apposition and fusion of the wound edges, and this is followed by a second event during which there is restitution of the neuroepithelium in the wound. Our study shows that apposition of the wound edges is: (1) a rapid process that occurs within the first 15 minutes following wounding, and (2) involves the whole length of the wound simultaneously. Healing therefore commences along the whole length of the wound. The wound at this time is "plugged" by rounded cells, and this results in a decrease in the thickness of the neuroepithelium in the wound area. The depression on the basal surface of the wound may be a reflection of 298 A. LAWSON AND M.A. ENGLAND Fig. 16. LM of stage 10 wounds in transverse section. (a)Freshly made wound; (b)re-incubated for 15 minutes; ( c ) re-incubated for 1 hour; (d) re-incubated for 5 hours. x 150. this. This pattern of healing clearly differs from that in the endoderm or ectoderm of chick and Xenopus embryos where the wound edges are not apposed to each other at the start of healing (England and Cowper, 1977;Stanisstreet et al., 1980).Wound healing in these tissues, therefore, starts from the ends of the wound, and the cells migrate t o close it. Adult epidermal wounds similarly heal by cell migration (Odland and Ross, 1968; Croft and Tarin, 1970; Krawczyk, 1971; Winter, 1972; Repesh and Oberpriller, 1980). This study also demonstrates that wound fusion immediately follows apposition of its edges. At the start of healing when the wound edges first meet at its apical end, the cell processes from the rounded cells become interlocked with each other. This stage of the healing process simulates a stage in the morphogenesis of the neural tube when there is an apposition followed by a fusion of the neural folds (Gouda, 1974; Bancroft and Bellairs, 1975; Waterman, 1975: Santander and Cuadradro, 1976; Silver and Kerns, 1978). We suggest here, as has been suggested for neurulation (Waterman, 1975, 19761, that the cell Processes might guide the edges of the wound together for fusion to occur. 80 y1 - 5 60 - 4 5 40 - 20 - ~9 0- I * Time of healing Fig. 17. A bar chart illustrating densities (mean + standard error) of rounded cells per time of re-incubation. ANOVA: F (4, 15) = 43.5; p = 0.0001. * rounded cells significantly different from 0 min (p < 0.05). 299 NEUROEPITHELIAL WOUND HEALING Our study further demonstrates that restitution of the neuroepithelium in the wound (i.e., restoration of the full thickness of the neuroepithelium in the wound area) involves a change in the shapes of the rounded cells to elongated forms. We have shown that rounded cells with markedly reduced intercellular spaces are present a t the edges of freshly made wounds and also during the initial stages of healing. Although these cells were closely apposed to each other, no junctional complexes were established a t this time as these could hinder a change in their shapes (Stanisstreet et al., 1980). Schoenwolf (1982) has described four classical cell types that characterize a normal pseudostratified neuroepithelium; spindle, wedge, inverted wedge, and globular (rounded). The globular cells are rarely seen and only when they are undergoing mitosis. Moreover, they are only seen at the apical surface. Our study has identified another variety of globular (rounded) cells present during neuroepithelial wound healing. These were evidently not undergoing mitosis as they possessed definite and rounded nuclei with well-defined nuclear envelope. Further, these cells lacked mitotic figures and they were present both near the apical and basal sides of the wound. The appearance of rounded cells during neuroepithelial wound healing, however, is only a transient phenomenon. Indeed this study reports a very high percentage (about 72% of the total cell population near the wound) immediately following wounding. The density remains around this level through the first 15 minutes of healing, but by about 1 hour, there is a significant decrease to about 26%. This latter event coincides with the time of appearance of more elongated cells in the wound with a concomitant re-establishment of intercellular spaces. It is at this time that the full thickness of the neuroepithelium is established in the wound. That the neurulating chick embryo is capable of undergoing changes in the shapes of the neuroepithelial cells has been demonstrated during neural plate shaping and bending in previous studies (Schoenwolf and Franks, 1984), and it is, therefore, not surprising that it utilizes one of the same mechanisms during wound healing. The cytoskeleton has long been known to be associated with changes that occur in cell shape in the neuroepithelium and in other embryonic organ systems during morphogenesis. Previous experimental studies that have used colchicine and cytochalasin B to disrupt microtubules and microfilaments, respectively, in the neuroepithelium have suggested that these cytoskeletal elements may direct the observed cell shape changes during neurulation (Handel and Roth, 1971; Karfunkel, 1971,1972; Fernandez et al., 1987; Schoenwolf and Powers, 1987; Schoenwolf et al., 1988). Studies are currently being undertaken to determine the role microtubules and microfilaments play in wound healing in the neuroepithelium. Our study revealed a close association between extracellular materials (specifically Glycosaminoglycans-GAGS) and the healing process. Although this may suggest their active involvement in healing, it was not clear what precise role they played. However, in view of their known ability to modulate cell shape (Van Hoof et al., 1986), we suggest that they might act in concert with the cytoskeleton to perform this function during neuroepithelial wound healing. Also, they are likely t o provide part of the “building blocks” for a new basal lamina over the basal surface of the wound. Further experiments are needed to determine precisely the role GAGS play in neuroepithelial wound healing in the chick embryo. It is intriguing that different parts of the neural tube of the chick embryo display varied mechanisms of wound healing. In contrast to our results, Clark and Scothorne (1990) reported that cell migration was the main mechanism for healing of the roof plates of spinal cords of stages 12 to 18 chick embryos and also that there was a 95% failure of healing at stages 17 and 18. It is yet to be determined, however, whether the failure of healing in the older age group is due purely to mechanical factors in the neural tube or to a loss of healing capabilities (i.e., loss of regulation). If the latter situation is true, it will be interesting to find out whether the loss of regulation is localized to the spinal cord alone or generalized, involving the entire CNS, and also whether there is a craniocaudal timing of onset. ACKNOWLEDGMENTS We thank Darko Farms, Ghana, for donating eggs for part of the study. We are grateful to Drs. Gary C. Schoenwolf, C.N.B. Tagoe, and R.S.K. Apatu for their useful comments on the manuscript. We are also grateful to Mr. G.L.C. McTurk for operating the IS1 60 and DS 130 scanning electron microscopes. Dr. I-Li Chen helped with the TEM. Miss M. Reeve did an excellent and rapid job of typing the manuscript. A.L. was supported by the British Council and the Association of Commonwealth Universities as a Medical Fellow while at Leicester University, Great Britain. The DS 130 scanning electron microscope was obtained through a Medical Research Council grant to M.A.E. LITERATURE CITED Bancroft, M., and R. Bellairs 1975 Differentiation of the neural plate and neural tube in the young chick embryo. Anat. Embryol., 147: 309-335. Birge, W.J., and H.H. Hilleman 1953 Metencephalic development and differentiation following experimental lesions in the early chick embryo. J. Exp. Zool., 124545-569. Clark, B.J. 1987 On the response of chick neural tube to experimental incision of the roof plate and the effect of hyaluronidase. J. Anat., 152:224-225. Clark, B., and R.J. Scothorne 1988 On the incidence and extent of open neural tube defects resulting from experimental incision of the roof plate in chick embryos of various ages. J . Anat., 158:225 (Abstract). Clark, B.J., and R.J. Scothorne 1990 Variation in the response of chick embryos to incision of the roof plate of the neural tube at different developmental ages. J . Anat., 168:167-184. Croft, C.B., and D. Tarin 1970 Ultrastructural studies of wound healing in mouse skin. 1: Epidermal behaviour. J . Anat., 106:63-77. England, M.A. 1981 Applications of the scanning electron microscope to the analysis of morphogenetic events. J . Micros., 123:133-146. England, M.A., and S.V. Cowper 1977 Wound healing in the chick embryo studied by scanning electron microscopy. Anat. Embryol., 152:l-14. Fernandez, J.G., P. de Paz, and C.A. Chamorra 1987 Effects of colchicine on the shape of neuroepithelial cells during neurulation. Anat. Rec., 219:296-303. Gabbiani, G., C. Chapponier, and I. Huttner 1978 Cytoplasmic filaments and gap junctions in epithelial cells and myofibroblasts during wound healing. J . Cell Biol., 76561-568. Gouda, J.G. 1974 Closure of the neural tube in relation to the developing somites in the chick embryo (Galus galus domesticus). J. Anat., 118:360-361. Hamburger, V., and H.L. Hamilton 1951 A series of normal stages in the development of the chick embryo. J . Morphol., 88:49-92. 300 A. LAWSON AND M.A. ENGLAND Handel, M.A., and L.E. Roth 1971 Cell shape and morphology of the neural tube: Implications for microtubule function. Dev. Biol., 25:78-95. Kallen, B. 1955 Regeneration in the hind-brain of neural tube stages of chick embryos. Anat. Rec., 123:169-182. Karfunkel, P. 1971 The role of microtubules and microfilaments in neurulation in Xenopus. Developmental Biology, 25:30-56. Karfunkel, P. 1972 The activity of microtubules and microfilaments in neurulation in the chick. J . Exp. Zool., 181:289-302. Karnovsky, M.J. 1965 A formaldehyde-glutaraldehydefixative of high osmolality for use in electron microscopy. J . Cell Biol., 27: 137a. Krawczyk, W.S. 1971 A pattern of epidermal cell migration during wound healing. J . Cell Biol., 49:247-263. Mareel, M., and L.C. Vakaet 1977 Wound healing in the primitive deep layer of the young chick blastoderm. Virch. Arch. Biol. Cell Path., 26:147-157. New, D.A.T. 1955 A new technique for the cultivation of the chick embryo in vitro. J. Embryol. Exp. Morphol., 7:146-164. New, D.A.T. 1959 The adhesive properties and expansion of the chick blastoderm. J. Embryol. Exp. Morphol., 7:146-164. Nieuwkoop, P.D., and J. Faber 1956 A Normal Table of Xenopus laevis Daud. North Holland, Amsterdam. Odland, G., and R. Ross 1968 Human wound repair. 1: Epidermal regeneration. J. Cell Biol., 39:135-151. Pang, S.C., Wh.H. Daniels, and R.C. Buck 1978 Epidermal migration during the healing of suction blisters in rat skin: a scanning and transmission electron microscopic study. Am. J . Anat., 153:177191. Pfister, R.W. 1975 The healing of corneal epithelial abrasions in the rabbit: A scanning electron microscope study. Invest. Ophthal., 14:648-661. Plumel, M. 1948 Tampon a u cacodylate de sodium. Bull. SOC.Chim. Biol. (Paris),30:129. Repesh, L.A., and J.C. Oberpriller 1980 Ultrastructural studies on migrating epidermal cells during the wound healing stage of regeneration in the adult newt, Notophthalmus uiridescence. Am. J. Anat., 159:187-208. Santander, R.G., and G.M. Cuadradro 1976 Ultrastructure of the neural canal closure in the chick embryo. Acta Anat., 95:368-383. Schoenwolf, G.C. 1982 On the morphogenesis of the early rudiments of the developing central nervous system. Scan. Elect. Microsc., 1:289-308. Schoenwolf, G.C., D. Folsom, and A. Moe 1988 A re-examination of the role of microfilaments in neurulation in the chick embryo. Anat. Rec., 220:87-102. Schoenwolf, G.C., and M.V. Franks 1984 Quantitative analyses of changes in cell shapes during bending of the avian neural plate. Dev. Biol., 105:257-272. Schoenwolf, G.C., and M.L. Powers 1987 Shaping of the chick neu- roepithelium during primary and secondary neurulation: Role of cell elongation. Anat. Rec., 218:182-195. Schoenwolf, G.C., and J.L. Smith 1990 Mechanisms of neurulation: traditional viewpoints and recent advances. Development, 109: 243-270. Silver, M.H., and J.M. Kerns 1978 Ultrastructure of neural fold fusion in chick embryos. Scan. Elect. Microsc., 11:209-215. Singley, C.T., and M. Solursh 1980 The use of Tannic Acid for the Ultrastructural visualisation of hyaluronic acid. Histochemistry, 65:93-102. Smith, J.L., and G.C. Schoenwolf 1987 Cell cycle and neuroepithelial cell shape during bending of the neural plate. Anat. Rec., 218: 196-206. Spratt, N.T. 1940 An in vitro analysis of the organisation of the eye forming area in the early chick blastoderm. J . Exp. Zool., 85: 171-209. Stanisstreet, M., and M. Panayi 1980 Effects of colchicine, cytochalasin B and papaverine on wound healing in Xenopus early embryos. Experientia, 36:lllO-1111. Stanisstreet, M., M.J. Smedley, and R.M. Snow 1985 Tension and wound healing in amphibian early embryos. Acta Embryol. Morphol. Exp., new series, 6~177-191. Stanisstreet, M., J. Wakely, and M.A. England 1980 Scanning electron microscopy of wound healing in Xenopus and chick embryos. J . Embryol. Exp. Morphol., 59:341-353. Takeuchi, S. 1976 Wound healing in the cornea of the chick embryo. 111: The influence of pore size of millipore filters on the migration of isolated epithelial sheets in culture. Devel. Biol., 51:49-62. Van Hoof, J., F. Harrison, L.Andries, and L. Vakaet 1986 Microinjection of glycosaminoglycan-degrading enzymes in the chicken blastoderm. An ultrastructural study. Differentiation, 31 :14-19. Waddington, C.H., and A. Cohen 1936 Experiments on the development of the head of the chick embryo. J . Exp. Biol., 13:219-236. Waterman, R.E. 1975 SEM observations of surface alterations associated with neurulation in the hamster and mouse. Am. J. Anat., 146~151-172. Waterman, R.E. 1976 Topographical changes along the neural fold associated with neurulation in the hamster and mouse. Am. J . Anat., 146:151-172. Watterson, R.L., and I. Fowler 1953 Regulative development in the lateral halves of chick neural tubes. Anat. Rec., 117~773-803. Wenger, E.L. 1950 An experimental analysis of relations between parts of the brachial spinal cord of the embryonic chick. J . Exp. Zool., 11451-85. Winter, G.D. 1972 Epidermal regeneration studied in the domestic pig. In: Epidermal Wound Healing. H.I. Maibach, H.I. and D.T. Rovee, eds. Year Book Medical Publishing, Chicago. Yntema, C.L., and W.S. Hammond 1945 Depletions and abnormalities in the cervical sympathetic system of the chick following extirpations of the neural crests. J. Exp. Zool., 100:237-263.