Beginning during the late blastula stage in zebrafish, cells located beneath a surface epithelial layer of the blastoderm undergo rearrangements that accompany major changes in shape of the embryo. We describe three distinctive kinds of cell rearrangements. (1) Radial cell intercalations during epiboly mix cells located deeply in the blastoderm among more superficial ones. These rearrangements thoroughly stir the positions of deep cells, as the blastoderm thins and spreads across the yolk cell. (2) Involution at or near the blastoderm margin occurs during gastrulation. This movement folds the blastoderm into two cellular layers, the epiblast and hypoblast, within a ring (the germ ring) around its entire circumference. Involuting cells move anteriorwards in the hypoblast relative to cells that remain in the epiblast; the movement shears the positions of cells that were neighbors before gastrulation. Involuting cells eventually form endoderm and mesoderm, in an anteriorposterior sequence according to the time of involution. The epiblast is equivalent to embryonic ectoderm. (3) Mediolateral cell intercalations in both the epiblast and hypoblast mediate convergence and extension movements towards the dorsal side of the gastrula. By this rearrangement, cells that were initially neighboring one another become dispersed along the anterior-posterior axis of the embryo. Epiboly, involution and convergent extension in zebrafish involve the same kinds of cellular rearrangements as in amphibians, and they occur during comparable stages of embryogenesis.
In the zebrafish embryo, after an early developmental period of rapid cleavages, morphogenetic movements occur that rapidly produce major changes in the appearance and organization of the blastoderm. During epiboly (Trinkaus, 1984a; 19846), beginning at the late blastula stage about 4h after fertilization, the blastoderm thins and spreads to completely cover the yolk cell during the course of 6h. Gastrulation begins about an hour after epiboly is underway. The blastoderm, a single multilayer of cells, rearranges into a two-layered structure consisting of a more superficial epiblast, and an inner hypoblast (Wilson, 1891). Shortly after gastrulation begins, the embryonic axis appears and lengthens along one side of the embryo (the dorsal side), as cells accumulate and line up specifically at that location. The rearrangements that occur among the cells of the blastoderm during early morphogenesis, particularly with respect to their lineal relationships and their future fates, are not well understood.
For example, several early embryologists concluded that during gastrulation the hypoblast originates by cell involution, a streaming of cells lying at the blastoderm margin inward and underneath their neighbors (Wilson, 1891; Morgan, 1895; Pasteels, 1936). Later, Ballard (1966a,b,c) concluded that the movement was in the opposite direction: deep-lying blastoderm cells spread outward towards the margin to form the hypoblast. Ballard’s view has been generally accepted, but very recently, involution was observed directly in the small embryo of a teleost, the rosy barb (Wood and Timmermans, 1988).
During the course of cell-lineage analyses, we have followed cell movements during epiboly and gastrulation in zebrafish. We observed cell rearrangements that seemed nonsensical if considered only in terms of the eventual fates that the lineages produced. First, in the late blastula, cells scatter chaotically (Kimmel and Law, 1985b; Kimmel and Warga, 1986). Second, in the gastrula, neighboring cells at the blastoderm margin undergo anterior-posterior inversions in their positions (Kimmel and Warga, 1987a). Finally, cells in either ectodermal (Kimmel and Warga, 1986) or mesodermal (Kimmel and Warga, 1987a) lineages disperse along the anterior-posterior axis of the embryo.
We now show that each of these cellular rearrangements are understandable if they are considered in relation to the changes in form of the blastoderm that occur at the same time. Studies done mostly in Xenopus suggest that cells undergo specific rearrangements to mediate the changes in form (Keller, 1987). We find the same rearrangements occur in zebrafish at the comparable stages of development.
Materials and methods
Embryos and stages
Zebrafish embryos were obtained from natural spawnings and staged by cell number during early cleavage. They were dechorionated with watchmaker’s forceps and kept at 28.5°C in an incubation medium of 14mM NaCl, 0.6 mM KC1, 1.3 DIM CaCl2, ImM MgSO4, and 0.07mM sodium-potassium phosphate buffer (pH 7.2). In some experiments, we used embryos homozygous for the gol-1 (golden) mutation (Streisinger et al. 1981), because they are lightly pigmented relative to the wild type, and fluorescently labeled cells in their bodies can be observed more clearly in whole-mount preparations after pigment cells differentiate.
Developmental time usually was determined from the morphological features of the embryo, and Table 1 gives a staging series for the period of development of interest, from midblastula period until somites begin to form. We use the letter h to mean hours after fertilization at 28.5°C. A previously published series, although less complete, includes useful sets of photographs (Hisaoka and Battle, 1958; Hisaoka and Firlit, 1960). In our series, names in common usage in embryology denote major periods of development (e.g. midblastula, gastrula), and the stages subdivide these periods. We name rather than number the stages, which seems to help one to remember them, and is more flexible.
Single blastomeres were injected (Kimmel and Law, 1985a), in midand late blastula embryos with the lineage tracer dye tetramethylrhodamine-isothiocyanate dextran (Molecular Probes, Eugene, OR; 10x103MT, diluted to 5 % (wt./vol.) in 0.2 M KC1). The second dye for double-label experiments was fluorescein-dextran (Sigma), dissolved the same way. Injections were made by pressure, usually over the course of a few seconds, either into a cell in the surface enveloping layer (EVL), or, in other cases, into a cell in the deep layer (DEL) of the blastoderm. To inject a DEL cell, the injection pipette was advanced through the intact EVL. It was technically more difficult to specifically inject single DEL cells than EVL cells, even under visual control. As an aid, we monitored voltage through the injection pipette. We observed that successful passage of the pipette through the EVL was accompanied by a rise in voltage of up to 40 mV; the extracellular space surrounding DEL cells is at a positive potential relative to the bath (Bennett and Trinkaus, 1970). Upon intracellular penetration of a DEL cell, we then observed the expected shift to negative potential, reflecting the membrane potential of the cell.
Observations of fluorescent cells in live embryos
For short-term viewing of labeled cells, embryos were usually positioned as desired in a gel of 3 % methyl cellulose made in the aqueous incubation medium described above and viewed without a coverglass. Alternatively, embryos in incubation medium were sandwiched between two micro cover glasses that were spaced apart with three pairs of cover glasses (each 0.13-0.17mm thick). For longer term viewing and for timelapse recordings, the embryos were held stationary in such chambers in a gel of 0.1 % agarose made in the same medium, and the chamber was then sealed with Vaseline to prevent evaporation. Observations were made using a Zeiss microscope with illumination from both a transmitted and an epilight source (Zeiss filter set 48-77-14), which permitted simultaneous imaging of labeled and unlabeled cells. The fluorescent image was amplified with a Silicon-IntensifiedTarget (SIT) video camera (Dage) to prevent light-induced damage to the labeled cells. In some experiments, the depths of fluorescent cells were determined with a digital shaft encoder fitted to the fine-focus knob of the microscope.
For time-lapse recordings, single-frame images were taken with a Gyre video recorder at 4 s intervals. The epi-light source was controlled by a shutter that illuminated the embryo for only 60 ms during each exposure, in order to minimize light-induced damage to the labeled cells. Frequent refocusing of the image was required during the recording session, which began at sphere stage (4 h) and continued for at least 6h, when epiboly is completed, and generally for a few hours longer. Afterwards, the embryo was released from the viewing chamber and reexamined at 24-30 h, when many cell types have begun to differentiate and can be distinguished by their morphologies and positions in identifiable tissues (Kimmel and Warga, 1987a). Labeled cells that were still undifferentiated were reexamined on the 2nd and/or the 3rd day of development.
We also studied a set of sectioned embryos. They were fixed at intervals between late blastula (4h) and midgastrula (7 h) periods by immersion in Bouin’s solution (Humason, 1962), dehydrated and embedded in Epon A12. Serial 5 μm sections were cut and stained with azure A, methylene blue and basic fuchsin (Humphrey and Pittman, 1974).
Deep and shallow DEL blastomeres intercalate during epiboly
In zebrafish, cleavages generate two populations of distinctive blastoderm cells; flattened epithelial cells in a surface enveloping layer (EVL), and rounded, more loosely associated deep layer (DEL) cells lying beneath the EVL. The EVL is a monolayer and the DEL a multilayer of cells. All of the movements we describe in this paper pertain to the DEL: the EVL cells behave relatively passively. Neighbór exchanges occur within the EVL (Keller and Trinkaus, 1987), but they are infrequent. We have not observed neighbor exchanges between the EVL and DEL.
DEL cells become motile in the midblastula, after the tenth cleavage at 3h (D. A. Kane, unpublished observations). The embryo flattens to take on a spherical shape by 4h (late blastula; Fig. 1A), and during the next hour of development, a rapid thinning of the blastoderm becomes evident, signifying epiboly is underway. The first change observable is very deep in the embryo, where the yolk cell begins to bulge or ‘dome’ towards the animal pole (Fig. IB). The blastoderm then rapidly takes on a cup-shaped appearance, and spreads to cover the yolk cell (Fig. 1C).
At the beginning of the late blastula stage, single clones descended from a progenitor cell labeled earlier are coherent groups of cells. Later, during epiboly, the DEL cells in such clones rapidly spread apart, interspersing with unlabeled cells (Kimmel and Law, 19856). In Xenopus, epiboly is known to occur by radial cell intercalations, in which cells at different depths in the blastoderm intercalate, thus producing its thinning (Keller, 1980). Such a rearrangement could produce the cell scattering we observed in zebrafish, and we have examined whether radial intercalations occur in this species.
We took advantage of the pattern of cell division during early cleavages to label, with two different colored dyes, two sibling blastomeres; one underlying the other at the 64-cell stage (Fig. 2). The deeper cell generated a clone located deep in the DEL of the midblastula, and immediately underlying the clone originating from its superficial sib, as confirmed by direct inspection (Fig. 3A). DEL cells of the two clones became thoroughly intermixed by early gastrula stage (Fig. 3B; note that the intermixing does not extend into the EVL). Subsequently, both sets of DEL cells gave rise to very similar sets of derivatives in the later embryo. In this example, both clones developed head ectodermal cell types (Fig. 3C). These results establish that blastoderm cells intercalate along radii during early epiboly, and that such movements are confined to the DEL.
The hypoblast arises by involution
At the time when the blastoderm half covers the yolk cell (5.2h; referred to as 50%-epiboly; Fig. 1C) new cell movements begin, including involution movements that form the hypoblast. These new movements mark the onset of gastrulation (see Discussion). Within about 15 minutes, the blastoderm becomes noticeably thicker in a circumferential band at its margin. The band, or germ ring, at first appears uniform in structure, and using time-lapse video microscopy, we observed in views from the animal pole (3 embryos) that it forms more- or-less simultaneously, within about 15 min, around the entire circumference of the blastoderm.
An analysis of involution is shown in Fig. 4, a case where we kept track of the depths of cells in the blastoderm as their rearrangements occurred. Here, minutes after the onset of gastrulation, a clone of 5 labeled cells was located near the margin of the blastoderm. The cells initially occupied a shallow position within the DEL, indicated by blue color-coding in Fig. 4A. The labeled cells moved towards the margin, apparently actively, since we observed blebbing and formation of filopodia, and two of them divided. Upon reaching the margin, each cell protruded processes and moved deeper within the DEL (Fig. 4B; green), away from the EVL and towards the surface of the yolk cell. Each marked cell then reversed its direction relative to the margin, and proceeded away from the margin, now located deeply in the blastoderm (Fig. 4C and D; red), within its new hypoblast layer.
These findings show that involuting DEL cells form the hypoblast. The first ones to involute are those located just at the blastoderm margin at the beginning of gastrulation (Fig. 5A), and as they migrate inwards the germ ring forms behind them (Fig. 5B and C). Analysis of sectioned embryos revealed that before germ ring formation there is no layering of cells within the DEL itself (Fig. 6A). However, after the germ ring forms, the DEL appears folded inwards at the margin, and is split, within the germ ring specifically, into the epiblast and hypoblast (Fig. 6B). As epiboly and gastrulation continue, cells that initially were located distantly from the margin move towards it within the epiblast, and involute. Consequently the hypoblast increases in area and extent beneath the epiblast, eventually spreading all the way from the margin to the animal pole.
Fates of epiblast and hypoblast cells
Single cells present during gastrulation generate clones restricted to single types of tissues (Kimmel and Warga, 1986). Now we have shown that some of these cells, but not others, involute during gastrulation, and we can ask whether involuting and noninvoluting cells have different fates. We kept track of cells in labeled clones during gastrulation and determined the fates of their descendants at a later stage, as illustrated in Fig. 7A. From such records, we reconstructed the cell lineage (Fig. 7B) and the pathways of movement of the cells during gastrulation (Fig. 7C). In this example, all of the labeled cells involuted, inverting their relative positions as they did so. The progeny of one of the cells originally present (Fig. 7, black lineage) all formed somitic mesoderm, including differentiated muscle fibers. The progeny of the other cell (stippled) formed derivatives of two different germ layers; gut epithelium (endoderm) and somitic muscle (mesoderm).
Summary lineage analyses for cells in other embryos analyzed the same way are shown in Fig. 8. Without exception (and also for 5 additional clones not illustrated), cells that involuted (arrows) later formed mesoderm and endoderm, and cells in lineages where involution did not occur (no arrows) formed ectoderm. The lineage shown in Fig. 8B is noteworthy, for in this case two sibling subclones, separated at the first division shown in the diagram, developed differently from one another but still followed the general rule: None of the cells in one of the subclones involuted and they subsequently developed as ectoderm. The cells in the second subclone all involuted and then formed mesoderm. We conclude from these results that the epiblast is the rudiment of ectoderm and the hypoblast is the rudiment of both mesoderm and endoderm.
The time of involution is related to fate
Cells that involute early during gastrulation usually form endoderm, and cells that involute later formed mesoderm. This can be seen most clearly from single clones that contributed to both germ layers (e.g. arrows in Figs. 7B and 8D). Except for the early involution of a cell that later formed heart tissue, a mesodermal derivative (Fig. 8D), this rule was generally followed (including those clones that contributed cells to only a single germ layer; compare the times of involution of cells in Fig. 8C and E).
Cells involuting at different times during gastrulation also had different pathways of movement within the hypoblast, and later occupied different positions in the embryo. Cells that involuted soon after the beginning of gastrulation sharply reversed their direction of movement as they involuted, turning towards the animal pole (Fig. 7C; stippled cell ppp). Cells that involuted somewhat later turned less sharply (stippled cell aaa), and cells that involuted much later during epiboly (black cell aaa) did not turn at all, but continued to move towards the vegetal pole after entering the hypoblast. No matter whether a turn occurred or not, all involuting cells recede from the margin of the blastoderm, which continues to rapidly advance by epiboly towards the vegetal pole. We measured the rates that the involuting DEL cells moved (18 cells from 3 embryos), but no significant differences were found in the speed of movement that correlated with whether a turn occurred, or with later fate (data not shown).
The animal pole of the gastrula develops into the anterior-most structures of the later embryo (Kimmel and Warga, 19876), and accordingly, cells that turn towards the animal pole during gastrulation develop more anterior structures. It follows that the order in which cells of a clone involute corresponds to their subsequent order along the anterior-posterior axis of the embryo, as can be seen from the positions of the black and stippled cells in Fig. 7. This relationship was consistent, as shown in Fig. 9 where data is collected from a set of 5 embryos in which the labeled clones were positioned at approximately the same lateral blastoderm location at the beginning of gastrulation.
Convergent extension movements are mediated by mediolateral cell intercalations
Convergence is a third morphogenetic cell movement occurring simultaneously with epiboly and involution. DEL cells move towards the dorsal side of the gastrula (Ballard, 1973; Kimmel and Warga 19876), ‘converging’ there from their original locations in the blastoderm. The formation of the embryonic shield, a local dorsal thickening of the germ ring (Oppenheimer, 1937; Hisaoka and Battle, 1958), is a prominent effect of early convergence movements. In Xenopus, convergence cannot be separated from extension, the elongation of the embryonic axis (Keller and Danilchik, 1988). Cells move from lateral positions dorsalwards by intercalating between neighboring cells that lie more medially (i.e. towards the axis). Such ‘mediolateral intercalations’ (Keller and Tibbetts, 1989) produce both narrowing and elongation of the axis.
Convergent extension occurs both in the epiblast (Kimmel and Warga, 1986; and see below) and in the hypoblast. Convergence of hypoblast cells is illustrated in Fig. 7C as a drift in the pathways of the cells towards the right side of the figure, the direction towards the embryonic shield in this example. That intercalations occur during this movement is revealed by the separation and spreading apart of the clonally-related cells along the anterior-posterior axis (Fig. 7A). During this dispersion, the labeled cells are intercalating among more medial unlabeled neighbors.
The intercalations are more completely illustrated in Fig. 10, an example following cells that remained in the epiblast. This case is instructive because intercalations occurred not only between labeled and unlabeled cells, but also between different labeled cells, and it is clear when they occurred. Four DEL cells (from a single clone) were present at the beginning of gastrulation. Their descendants eventually formed a dispersed series of clusters in the hindbrain, distributed along the axis, and on both sides of the midline. The cells are shaded to indicate their lineal relationships. Eventually black and hatched cells were positioned between different white cells. The intercalations effecting this intermixing occurred in the gastrula, beginning at about 6.8 h (Fig. 10). Shortly after both the black and hatched cells divided, one of the two daughter cells from each of the divisions inserted between the pair of white sister cells.
In contrast to the radial intercalations that we considered above, the intercalations occurring during convergent extension do not scatter cells indiscriminately. For example, we have not observed DEL cells in the hypoblast and epiblast to mix with one another during gastrulation, although mixing within each of these layers is extensive.
This work has revealed three distinctive cell movements that accompany, and appear to produce, the early changes in shape of the zebrafish embryo. These movements are epiboly, involution and convergent extension. They are diagrammed respectively in Figs 11-13.
Radial intercalations (Keller, 1980) among DEL blastomeres occur first, in the late blastula (Fig. 11), and along with an expansion and change in shape of the yolk cell that occur simultaneously, these cell movements appear to mediate the thinning of the blastoderm that occurs rapidly during this period of development. Intercalations thoroughly scatter DEL cells and are responsible for marked dispersion of clonally related cells that we have described elsewhere (Kimmel and Law, 19856). It is interesting, however, that DEL cells do not intercalate outward into the EVL. The EVL, by this stage, has acquired the form of a highly flattened squamous epithelium. In Fundulus at a comparable stage of development, junctional complexes are present between cells of the EVL (Betachaku and Trinkaus, 1978), and it may be that such junctions mediate high adhesivity among the EVL cell, such that the underlying DEL cells are unable to penetrate this layer.
Involution movements of DEL cells located near the blastoderm margin produce the hypoblast, an inner layer of the blastoderm (Fig. 12). Involution (or in some animals its counterpart invagination; see Trinkaus, 1984b) is the singular morphogenetic movement that characterizes gastrulation in many different types of animals; hence we consider the beginning of gastrulation in zebrafish as the time when involution begins. This is the stage when the blastoderm has advanced, by epiboly, to cover just one-half of the yolk cell. The cells move first towards the blastoderm margin, in the general direction of the vegetal pole. When they reach the margin they involute to take up a new, deeper, position. Afterwards they either reverse their direction of movement and move towards the animal pole, or, in the case of cells that involute later during gastrulation, they continue moving towards the vegetal pole. In either case, once cells are in the hypoblast, they are left behind the leading edge of the blastoderm, which continues to advance across the yolk cell by epiboly during the gastrula period. We have obtained no evidence that cells can enter the hypoblast by any other movement than involution, although we have not yet carefully examined cell movements within the embryonic shield (at the dorsal side of the embryo).
Wood and Timmermans (1988) recently also observed involution in the rosy barb, another teleost in the same family as zebrafish. However, Ballard could not find involution in his careful and extensive studies of a variety of other (and larger) teleost embryos (1966a,b,c; 1973; 1981; 1982). It may be that gastrulation is dramatically different in large and small teleost embryos, but we think it more likely that all teleosts gastrulate as the zebrafish and rosy barb do, and that the cell marking procedures available to Ballard were inadequate to reveal all the cell movements that occur in the DEL.
Involution is special in teleost fish, as compared to other types of vertebrates, in that the EVL is not involved (see below). Moreover, involution doesn’t seem to be initiated first at the dorsal side of the embryo (as it does for example in amphibians). As judged from the time-course of appearance of the germ ring, involution in the zebrafish begins more- or-less simultaneously around the circumference of the blastoderm.
During gastrulation DEL cells also undergo mediolateral intercalations (Fig. 13), producing a general dorsalwards drift of the cells that has been described previously in other teleosts (e.g. Ballard, 1973; 1982). The cells accumulate dorsally to form the embryonic shield, and the subsequent narrowing (convergence) and lengthening (extension) of the shield produces a welldefined embryonic axis within about two hours after the shield first forms.
We have shown that cells in both the hypoblast and epiblast undergo convergent extension. The intercalations appear to be regulated such that extensive mixing occurs among the cells within both of these layers, but not between the layers. Moreover, the fact that most gastrula lineages are tissue-restricted (Kimmel and Warga, 1986) shows that mixing among cells must occur within, but not between, the primordia of different tissues. However, the boundaries of the primordia are invisible in the gastrula, such that we could not hope to observe distinctive cellular behaviors in their vicinities. Later in embryogenesis the boundaries become recognizable, and no mixing occurs across at least one of them the boundary separating the axial (prospective notochord) and paraxial (prospective somite) mesoderm - as recently shown for Xenopus (Wilson et al. 1989) and rosy barb (Thorogood and Wood, 1987).
The enveloping layer
All of the movements we have described appear to closely resemble their counterparts that have been thoroughly described in Xenopus (Keller, 1986). Radial intercalations, involution movements and mediolateral intercalations occur at the equivalent stages, relative to gastrulation onset, in zebrafish and Xenopus and they produce equivalent changes in shape and organization of the embryo. There is a single important difference, however; the outside layer of cells in the teleost blastoderm does not participate in any of them. DEL cells do not enter the EVL during their radial intercalations, as we have shown in this study. We also confirmed that EVL cells do not undergo involution, as was first convincingly shown by Ballard for the trout (1966a). This finding was expected in zebrafish since the exclusive fate of the EVL is the periderm - an outermost epithelial cell layer covering the embryo (Kimmel and Warga, 1986; Kimmel et al. 1990). We showed earlier (Kimmel and Warga, 19876) that the EVL cells do not undergo convergence, at least in the sense used here to mean a specific dorsalwards movement.
The EVL may be a relatively passive participant in blastoderm epiboly; as revealed by studies in Fundulus, it seems to be pulled and stretched across the yolk cell by the yolk syncytial layer of the yolk cell itself (Betchaku and Trinkaus, 1978; Trinkaus, 1984a). However, perhaps active rearrangements among EVL cells have recently been shown to occur both in Fundulus (Keller and Trinkaus, 1987) and the medaka (Kageyama, 1982), where they serve to continuously decrease the diameter of the EVL as epiboly is completed and the marginal ring of EVL cells closes at the vegetal pole of the yolk cell. It is likely that this rearrangement also occurs in the zebrafish, for EVL cells in single clones do sometimes become dispersed from one another, rather than being present in a single coherent patch (e.g. Kimmel and Warga, 19876). The dispersion is, however, markedly less than that occurring in the DEL.
Control and patterning of cell movements
Our studies are descriptive, and do not reveal the mechanisms that underlie these morphogenetic movements. However, the rearrangements appear to be active ones, for DEL cells constantly change in shape and they move relative both to neighboring cells and to a fixed point on the yolk cell. The yolk cell and EVL cells both participate in epiboly and, as we have shown here, so do DEL cells. The gastrulation movements of involution and convergence may also depend upon interactions among cells of all three classes. Recently Symes and Smith (1987) suggested that activation of gastrulation movements in amphibians is an early consequence of mesodermal induction. This might involve the yolk cell; Long (1983) obtained evidence from transplantation experiments in the trout that the yolk cell can induce dorsoventral polarity of the blastoderm.
Progress in understanding how such specific movements are produced may come through mutational analysis in zebrafish. We have recently described a mutation, spt-1, that appears to selectively disrupt convergence of laterally positioned mesodermal cells during gastrulation (Kimmel et al. 1989). Convergence of ectoderm is not disturbed, suggesting that different genes control dorsalwards movements of cells that occupy different germ layers. Furthermore, mosaic analysis suggests that the wild-type gene is required in the mesoderm specifically (Ho et al. 1989). The gene could code for, or regulate the expression of, a receptor or adhesion molecule required for the convergence movements of a subset of mesodermal cells.
An important finding from our study is that cells that involute to enter the hypoblast then give rise to endoderm or mesoderm and, conversely, that the epiblast is the equivalent of ectoderm in other vertebrates. This observation is in accord with the interpretations of early investigators of teleost embryology (Wilson, 1891; Morgan, 1895; Pasteels, 1936). We also show that whether a cell in the hypoblast will form endoderm or mesoderm, and where its clonal descendants will come to lie along the anterior-posterior axis of the embryo is directly correlated with when it entered the hypoblast. Furthermore, we detected no differences in direction or rate of movements of DEL cells as they approached the margin, before involution, that correlated with their future fates. Together, these observations lead to the suggestion that whether and when a cell involutes is a direct function of how far from the margin it was positioned prior to the onset of gastrulation. We address this issue in the accompanying paper (Kimmel et al. 1990), examining in more detail how cell fate is related to cell position in the early gastrula.
We thank D. A. Kane, A. Felsenfeld and D. Frost for their critical comments on early versions of this paper, and for stimulating discussion throughout the course of the study. C. Cogswell, P. Myers, H. Howard, and R. Kimmel provided technical assistance. The research was supported by NSF grant BNS-8708638, NIH grant HD22486, and a grant from the Murdock Foundation.