The axolotl pronephric duct rudiment is readily accessible to both SEM observation and surgical manipulation. The rudiment segregates from the dorsal part of the lateral mesoderm and then extends caudally along the ventrolateral border of the segmenting somites, eventually contacting the cloacal wall. The marked thinning of the rudiment which accompanies this migration is paralleled by a corresponding reduction in cell number across the duct’s diameter and by caudad translocation and elongation of vital dye marks applied to the duct mesoderm. Duct extension thus involves appreciable cell rearrangement. The morphology of duct mesoderm and its substratum (somite and lateral mesoderm) suggests that active locomotion of cells near its tip marshals the duct’s caudad elongation. Filopodia and small focal areas of intercellular contact may mediate the adhesions between duct cells which must be broken and reformed as the cells rearrange.

The amphibian pronephric duct during its early morphogenetic phase provides an example of directed tissue migration that is especially well suited for experimental analysis. Scanning electron microscopy of normal embryos fixed at various stages, vital dye marking and simple surgical deletions or blockages have shown the events of duct formation to be very similar in Ambystoma maculatum (Poole & Steinberg, 1977) and, in the present work, in the axolotl A. mexicanum. In these embryos the pronephric duct rudiment segregates from the mesoderm as an ovoid, solid tissue mass five to six somites long and then by cell rearrangement extends to more than twice its original length along the ventrolateral margin of the somites to join with the cloaca. Thus, the salamander duct forms by the caudal extension of a solid stream of cells along a predetermined and easily identifiable path readily accessible to scanning electron microscopic (SEM) observation and surgical manipulation.

The mode and mechanisms of outgrowth of the amphibian pronephric duct have been subjects of some controversy since the turn of the century (reviewed by Burns, 1955; Fox, 1963; Poole & Steinberg, 1977). According to one view, the duct forms by progressive recruitment of cells in situ, while another view holds that it forms by caudal extension of an anterior rudiment. The latter view, which has come to be generally accepted (exceptions: Shin-Ike, 1955; Fox & Hamilton, 1964; see Poole & Steinberg, 1977 for species differences), has been supported by experiments utilizing localized vital dye staining, surgical deletion, blockage or reorientation of the duct tip, and explantation. Since elongating Ambystoma pronephric duct rudiments do not have a higher mitotic rate than surrounding tissues (Overton, 1959), their extension seems to be due to cell migration. The migratory propensity of duct rudiment fragments has previously been demonstrated by outgrowth in plasma clots (Overton, 1959), by ablation of a major part of the duct rudiment (Nieuwkoop, 1947) and by transplantation of young duct rudiments to virgin ‘duct paths’ of older hosts (Gipouloux & Cambar, 1961; Cambar & Gipouloux, 1970).

What are the nature and specificity of the environmental factors determining the duct’s course? Holtfreter and others addressed this question by confronting the advancing duct primordium with surgically produced foreign tissue terrains. Holtfreter (1944) found that the (urodele) duct could be deviated, by a wound, ventrally onto the surface of the lateral mesoderm. From this position it was able in several exceptional cases to return to its normal path and complete its migration. In the same year, Tung & Ku (1944), working with anuran embryos, found that the duct rudiment resisted extension at right angles to this path. Bijtel (1948) observed a deviation of the duct from its normal path to a laterally implanted secondary cloaca.

Thus, although there is much suggestive evidence, the manner in which the cells of the duct rudiment migrate and the environmental factors that guide them are not yet understood. Because yolkiness of amphibian embryos during duct migration makes paraffin sectioning at this stage difficult, the results of surgical operations have usually been assessed on embryos fixed at later stages, after much of the yolk has been digested. Thus the consequences of microsurgical procedures were first observed only after the duct rudiment had completed its caudal migration, when secondary influences might have deviated the duct from its originally chosen path. We therefore chose to make our observations by scanning electron microscopy, which not only permits observations to be made at any time but also reveals the appearance of individual cells and cellular processes during elongation of the duct rudiment.

Axolotl (Ambystoma mexicanum) embryos were obtained from spawnings of our colony and that of Indiana University. Embryos were staged according to Schreckenberg & Jacobson, 1975 (S & J) and manually demembranated with fine watchmaker’s forceps in full-strength Steinberg’s solution (see Discussion with Reviewers in Poole & Steinberg, 1977).

Experimental manipulations were carried out under aseptic conditions in full-strength Steinberg’s solution using standard microsurgical procedures (Jacobson, 1967). Embryos were vitally stained with Nile blue sulphate-dyed agar slivers by a procedure similar to that described by Keller (1975).

Embryos were generally fixed at room temperature with modified Karnovsky’s (1965) fixative (2·5% glutaraldehyde, 2·5% paraformaldehyde and 5 HIM calcium chloride in 0·1 M-sodium cacodylate buffer, pH 7·4). After h fixation, the ectoderm was manually peeled off with fine watchmaker’s forceps and tungsten needles under adissecting microscope. Peeled embryos were transferred to fresh fixative and usually left at 4°C overnight. Samples were then rinsed in several changes of 0·15 M sodium cacodylate buffer and postfixed in sodium cacodylate-buffered 1% osmium tetroxide for 1–3 h at 4°C.

Embryos for scanning electron microscopy were dehydrated in ethanol and critical-point dried from liquid CO2. Dried embryos were affixed to stubs with a low-resistance contact cement (Fullam) or with silver paint and sputter coated with gold-palladium (60:40). Specimens were examined at 15–25 kV in a JEOL JSM-35 scanning electron microscope. For transmission electron microscopy, dehydrated samples were embedded in Epon 812. Transverse sections, 1–2 μm thick, were cut with glass knives, mounted on slides and stained with methylene blue and azure IL Ultrathin sections (50–70 nm) were then cut from selected regions, mounted on grids, double stained with 2% uranyl acetate and lead citrate and examined with a JEOL 100C electron microscope operated at 80 kV.

The dimensions of embryos were measured directly from SEM negatives with a Zeiss MOP-3 image analyzer.

Segregation and elongation of the duct rudiment

The axolotl pronephric duct segregates between the levels of trunk somites 2 and 7 as a solid, ovoid or tear-shaped body of cells at S. & J. stages 22 and 23 (Fig. 1a). With development it extends along the ventrolateral border of the somites while narrowing markedly (Figs. 1b-d). The level of origin and extent of migration have been confirmed by vital dye marking. A mark placed to include the caudal limit of the rudiment and adjacent somite mesoderm at stage 23 (Fig. 2b) shows a pronounced translocation and spreading of stained duct cells over 24 h at 24°C. Marks made at older stages when rearrangement is already in progress and placed further caudally and cranially show reduced spreading and caudal translocation (Fig. 2b). Marked cells behind the duct tip are not incorporated into the advancing duct. These results are consistent with expectations based upon the morphology observed in low-power SEM micrographs (Fig. 1) and suggest that duct rudiment elements tend to remain near their original neighbours during the rearrangement accompanying extension. Finally, the level of origin and propensity for extension of the duct rudiment are clearly shown by surgical intervention. Most simply, a deep transection of the axial tissues caudal to the duct rudiment’s tip (i.e. posterior to trunk somite 7, see Fig. 2a), ip all 18 cases halted duct progression at the level of the incision (Fig. 3a). This confirms that the rudiment extends over 5 somite widths at stage 22. Following a more shallow incision, the duct rudiment has, in eight cases, detoured ventrolaterally a short distance across lateral mesoderm and returned to its normal path Caudal to the incision (Fig. 3b).

Fig. 1

Scanning electron micrographs of Ambystoma mexicanum embryos fixed before peeling of ectoderm from the right side. Arrows indicate pronephric duct’s caudal tip. Duct rudiment’s extension is accompanied by the segmentation of additional somites and straightening of the embryonic axis, (a) Stage 22, (b) Stage 24, (c) Stage 28, (d) Stage 32.

Fig. 1

Scanning electron micrographs of Ambystoma mexicanum embryos fixed before peeling of ectoderm from the right side. Arrows indicate pronephric duct’s caudal tip. Duct rudiment’s extension is accompanied by the segmentation of additional somites and straightening of the embryonic axis, (a) Stage 22, (b) Stage 24, (c) Stage 28, (d) Stage 32.

Fig. 2

Camera-lucida tracings of vitally stained embryos, (a) Distal segment of pronephric duct stained with Nile blue sulfate at stage 22 has moved caudad and elongated markedly by stage 32. (b) Proximal segment of duct stained at stage 26 has moved caudad and elongated to a lesser extent by stage 32. A stained section of the duct’s path is obscured as the duct passes over it.

Fig. 2

Camera-lucida tracings of vitally stained embryos, (a) Distal segment of pronephric duct stained with Nile blue sulfate at stage 22 has moved caudad and elongated markedly by stage 32. (b) Proximal segment of duct stained at stage 26 has moved caudad and elongated to a lesser extent by stage 32. A stained section of the duct’s path is obscured as the duct passes over it.

Fig. 3

Two axolotl embryos were split at stage 22 by a dorsal incision and peeled after fixation at stage 32. (a) A deep cut has blocked duct migration, (b) A more shallow cut has permitted some extension of the duct rudiment below the wound.

Fig. 3

Two axolotl embryos were split at stage 22 by a dorsal incision and peeled after fixation at stage 32. (a) A deep cut has blocked duct migration, (b) A more shallow cut has permitted some extension of the duct rudiment below the wound.

Duct rudiment elongation occurs by cell rearrangement

How do the pronephric duct shape changes come about? The cellular basis of the reduction in the duct’s diameter can be appreciated by comparing SEM micrographs at a given level (beneath trunk somite 6) at various stages of development. In the sequence shown in Fig. 4, the duct narrows from about eight cell widths at stage 23 to two cell widths at stage 32. The duct cells themselves Change little if at all in size or shape. The marked thinning of the rudiment, accompanied by a decrease in the number of cells across its diameter, is also seen in cross-sectional views. Figure 5 shows two views produced by fracturing transversely through trunk somite 6 of critical-point-dried embryos. Thinning of the rudiment reduces the number of cells spanning the duct’s width from six to eight at stage 24 (Fig. 5b) to two to three at stage 28 (Fig. 5b). The same reduction can be seen in 2 μm Epon sections. All of these observations indicate that cell rearrangements, and not proliferation or cell-shape changes, are prinjarily responsible for Ambystoma pronephric duct extension.

Fig. 4

The duct rudiment, seen here below trunk somite 6, thins markedly by tell rearrangement as it elongates, (a) Stage 22, six to eight cells wide; (b) Stage 26, five to six cells wide; (c) Stage 28, about four cells wide; (d) Stage 32, two to three cells wide.

Fig. 4

The duct rudiment, seen here below trunk somite 6, thins markedly by tell rearrangement as it elongates, (a) Stage 22, six to eight cells wide; (b) Stage 26, five to six cells wide; (c) Stage 28, about four cells wide; (d) Stage 32, two to three cells wide.

Fig. 5

Decrease in cell number in transverse sections of the pronephric duct (arrows) is apparent in critical-point-dried embryos fractured through the level of trunk somite 6. (a) A. maculatum (essentially like A. mexicanum), Stage 24; (b) A. mexicanum, Stage 28.

Fig. 5

Decrease in cell number in transverse sections of the pronephric duct (arrows) is apparent in critical-point-dried embryos fractured through the level of trunk somite 6. (a) A. maculatum (essentially like A. mexicanum), Stage 24; (b) A. mexicanum, Stage 28.

In Fig. 6 the developmental changes in several parameters of duct outgrowth are summarized graphically. This illustrates several significant points. Despite the increase in total embryo length (straight line head to tail), the length of the duct path surprisingly remains nearly constant. Inspection of the tracings in Fig. 7 reveals that the embryo’s elongation between stages 24 and 32 results from the gradual straightening of the embryonic axis and the lifting and extension of the head. The boundary between presumptive somite and lateral mesoderm which defines the duct’s path is quite curved at stage 22 and merely straightens out as the embryo ‘elongates’. Duct extension is closely correlated with somite segmentation; during elongation, the caudal tip of the duct rudiment maintains a position two somite widths behind the most caudally developing somite fissure. Finally, both the increase in duct length and the decrease in duct diameter are linear with time and have similar slopes (Fig. 6).

Fig. 6

Dimensional changes during axolotl pronephric duct extension. Embryo length measurements (EL; ●)are recorded as the straight-line distance from tip of head to tip of tail. Total path length (PL; ◯) and duct length (DL; ●) measurements are curvilinear. Duct diameter (DD; ◼) is the linear distance from somite to lateral mesoderm across the duct at the level of trunk somite 6 as seen in Fig. 4. Dimensions taken from scanning electron micrographs.

Fig. 6

Dimensional changes during axolotl pronephric duct extension. Embryo length measurements (EL; ●)are recorded as the straight-line distance from tip of head to tip of tail. Total path length (PL; ◯) and duct length (DL; ●) measurements are curvilinear. Duct diameter (DD; ◼) is the linear distance from somite to lateral mesoderm across the duct at the level of trunk somite 6 as seen in Fig. 4. Dimensions taken from scanning electron micrographs.

Fig. 7

Tracings of scanning electron micrographs of partially peeled axolotl embryos, showing the caudad progression of the pronephric duct, (a) Stage 22, (b) Stage 25, (c) Stage 32.

Fig. 7

Tracings of scanning electron micrographs of partially peeled axolotl embryos, showing the caudad progression of the pronephric duct, (a) Stage 22, (b) Stage 25, (c) Stage 32.

The substratum for duct migration

As seen in the transverse fractures (Fig. 5), the duct rudiment is bordered medially by somite mesoderm, ventrally by lateral mesoderm and dorso-laterally by ectoderm. As reported previously (Poole & Steinberg, 1977), the cells of exposed Ambystoma duct rudiments (mesoderm viewed en face) are attached via lobopodia, lamellipodia and many fine filopodia both to each other at the surface of the rudiment and to adjacent somite and lateral mesoderm cells at its edge (see Fig. 8 and Fig. 9). On the surface of the somites with which the duct makes attachments are localized webs of 50 – 100 nm fibres (apparently extracellular collagen fibres) as well as numerous fine, interdigitating cell extensions. The inner ectodermal surface is seen in the SEM to be partially covered by & basal lamina which obscures cell boundaries. Some 50–100 nm fibres are also seen, but there are no apparent features which might guide the duct’s migration. The duct and adjacent mesoderm adhere weakly if at all to the inner surface of the ectoderm. This is evident when the ectoderm is removed. After fixation, it can usually be easily peeled from the mesoderm with little evidence of damage to the latter. It can also be peeled from living embryos with little sign of firm adhesions to, distortion of, or damage to duct or adjacent mesoderm. Finally, chance fractures of dried embryos in which the duct mesoderm remained next to the ectoderm showed few and tenuous associations of duct cells with the inner surface of the ectoderm. It thus appears that the ventral edge of the somites and the subjacent lateral mesoderm comprise the substratum for the duct’s migration.

Fig. 8

Higher magnification view of posterior portion of stage-27 axolotl pronephric duct rudiment (PD), somites (S) and lateral mesoderm (L).

Fig. 8

Higher magnification view of posterior portion of stage-27 axolotl pronephric duct rudiment (PD), somites (S) and lateral mesoderm (L).

Fig. 9

Region of tip of stage-31 axolotl pronephric duct rudiment, (a) Posterior third of duct rudiment, (b) Cells near the tip overlap in the manner of fish scales, (c) Enlargement of the area indicated in Fig. 9b. Overlapping cells extend filopodia which contact underlying cells within the duct rudiment, (d) Meshwork of fibers approximately 0·2 μm in diameter, seen as occasional small patches on cell surfaces at this stage.

Fig. 9

Region of tip of stage-31 axolotl pronephric duct rudiment, (a) Posterior third of duct rudiment, (b) Cells near the tip overlap in the manner of fish scales, (c) Enlargement of the area indicated in Fig. 9b. Overlapping cells extend filopodia which contact underlying cells within the duct rudiment, (d) Meshwork of fibers approximately 0·2 μm in diameter, seen as occasional small patches on cell surfaces at this stage.

The morphology of duct cells and their contacts

Cells near the duct’s tip show some anteroposterior elongation (Fig. 8) and tend to overlap in the manner of fish scales (see also Fig. 9b). Back from the tip (as in Fig. 4), the duct’s cells are in a more ‘relaxed’ configuration. Cell-to-cell adhesion near the duct’s tip occurs by flat, overlapping cell processes from which arise numerous adherent filopodia roughly 200 nm in diameter and averaging about 10 μm in length (Fig. 9c, Fig. 10,b-d). The large lobopodial processes extending out toward somite and lateral mesoderm are much rarer back from the tip, especially at later stages of migration. Vital dye marking (Fig. 2) and the thinning visible in Figs. 4c-d show that these cells are still rearranging. Whether they are all engaged in active locomotion like a stream of Fundulus deep cells (Trinkaus, 1973) or whether the force causing their rearrangement arises from the locomotory activity of cells at the leading edge remains to be determined.

Fig. 10

Cell contacts visible in transverse fractures behind trunk somite 7 of a stage32 axolotl pronephric duct rudiment, (a) Low magnification overview showing neural tube, notochord, somites, epidermis, endoderm, pronephric duct and lateral mesoderm. (b) At higher magnification the ectoderm is seen to be bilaminar (bracket) and the cells of the pronephric duct rudiment (arrows) are seen to be in the process of adopting a radial arrangement. A fibrous network resembling collagen covers the exposed intersomitic surface (S). (c) The wedge shape of duct cells at this stage is apparent here. The cell depicted has a broad base at the duct’s outer surface (arrows) and an apex (asterisk) centrally where the duct’s lumen will form. Several blunt processes (small arrows) extend between cells, (d) An area near the center of the duct rudiment (triangle). Filopodia extend along the cell surface (small arrows). The adjoining cells are also connected by shorter, blunt processes (large arrow).

Fig. 10

Cell contacts visible in transverse fractures behind trunk somite 7 of a stage32 axolotl pronephric duct rudiment, (a) Low magnification overview showing neural tube, notochord, somites, epidermis, endoderm, pronephric duct and lateral mesoderm. (b) At higher magnification the ectoderm is seen to be bilaminar (bracket) and the cells of the pronephric duct rudiment (arrows) are seen to be in the process of adopting a radial arrangement. A fibrous network resembling collagen covers the exposed intersomitic surface (S). (c) The wedge shape of duct cells at this stage is apparent here. The cell depicted has a broad base at the duct’s outer surface (arrows) and an apex (asterisk) centrally where the duct’s lumen will form. Several blunt processes (small arrows) extend between cells, (d) An area near the center of the duct rudiment (triangle). Filopodia extend along the cell surface (small arrows). The adjoining cells are also connected by shorter, blunt processes (large arrow).

Transmission electron microscope studies of cell shapes and junctions provide a structural basis for the rearrangements and give clues to the type of locomotory activity involved. Figure 11 shows electron micrographs of sections taken through a stage-26 axolotl embryo several somite widths anterior to the caudal tip of the duct primordium. Electron-dense yolk platelets (YO), lipid-filled droplets or vesicles (LD), nuclei (Nu) and mitochondria (Mi) are visible. There are large gaps between cells, close apposition of cell membranes occurring in discrete areas (Fig. 11; several examples circled in Fig. 11,a). Frequently, close apposition occurs where a process of one cell touches the body of another (arrows in Fig. 11,b). Further caudally there is even more intercellular space and duct cell surfaces show fewer complex processes. In addition, long lamellipodial processes are seen at the ventromedial edge of the duct near its tip (Fig. 11c). Such structures may be important in caudal translocation of these cells during duct extension. The morphology and contacts observed as well as the scarcity of specialized junctions call to mind the observations of Nakatsuji (1975, 1976) on the motile cells of urodele and anuran gastrulae, and those of Hogan and Trinkaus (1977) and Trinkaus and Lentz (1967) on the migratory deep cells of the Fundulus gastrula.

Fig. 11

Transmission electron micrographs of sections through the pronephric duct rudiment of a stage-26 axolotl embryo, (a) An area of contact between two duct rudiment cells several somites anterior to the duct’s caudal tip. (b) Interdigitating filopodia (arrows) of duct cells show close contacts with opposing cell surfaces, suggesting that they mediate cell-cell adhesions, (c) Cells near the advancing tip of the duct rudiment possess long, flattened lamellipodia and are separated by more extracellular space. Yolk platelets (YO), lipid droplets (LD), nuclei (Nu) and mitochondria (Mi) are indicated.

Fig. 11

Transmission electron micrographs of sections through the pronephric duct rudiment of a stage-26 axolotl embryo, (a) An area of contact between two duct rudiment cells several somites anterior to the duct’s caudal tip. (b) Interdigitating filopodia (arrows) of duct cells show close contacts with opposing cell surfaces, suggesting that they mediate cell-cell adhesions, (c) Cells near the advancing tip of the duct rudiment possess long, flattened lamellipodia and are separated by more extracellular space. Yolk platelets (YO), lipid droplets (LD), nuclei (Nu) and mitochondria (Mi) are indicated.

Scanning electron microscopic observations of the outer mesodermal surface of normal and surgically modified embryos have provided new insights into the mechanisms directing the caudad extension of the amphibian pronephric duct. Even at low magnifications in the SEM, the Ambystoma pronephric rudiment can be seen to segregate out as a solid mass from the dorsal portion of the hoinogeneous lateral plate mesoderm ventral to trunk somites 2 through 7, as previously inferred by O’Connor (1938) and Holtfreter (1944) from studies utilizing vital staining and transplantation. Our own staining and surgical procedures confirm their results. The Ambystoma duct clearly forms by the caudal extension of an anterior rudiment.

Although a variety of cellular mechanisms (such as cell shape change, cell reorientation, individual cell movement, cell proliferation) might in principle cause the observed transformation from a short, thick cord to a long thin One, our observations have shown that this solid mesodermal cylinder of nearconstant volume extends itself by cell rearrangement. The rudiment’s marked thinning during elongation is accompanied by the redistribution of a nearly Constant number of constituent cells most of which remain in ‘relaxed’, polygonal shapes throughout the process. The force guiding this cellular rearrangemeht is not made obvious by ultrastructural observations. Although the morphology and distribution of cell processes, contacts and junctions suggest that the caudal-most cells are actively pulling out the duct rudiment, it is difficult to reconstruct with certainty the process of duct extension from the static images obtainable with electron microscopy. We have been able to approach this problem experimentally by surgical rearrangements of the salamander rrjesoderm. The results will be discussed in subsequent papers.

Cell rearrangement as a morphogenetic mechanism

Morphogenetic movements can be classified according to whether the cells migrate as individuals or as part of a cell group. Translocations of individual cells such as germ cells, neural crest cells and processes of neurons toward specific destinations require guidance by specific environmental clues. SEM observations by a number of authors have implicated cellular and extracellular fibers in such guidance (Bancroft & Bellairs, 1976; Ebendal, 1976, 1977; Lôfberg & Ahlfors, 1978; Tosney, 1978; Wylie, Heasman, Swan & Anderton 1979). The morphogenesis of cell groups, however, may be guided by other control mechanisms mediated by cell interactions such as adhesion, contact inhibition or changes in the shapes of firmly associated, individual cells (Phillips Steinberg & Lipton, 1977). One of us has recently divided tissue movements into two broad categories. When cells remain firmly affixed to their neighbours, individual cell shape changes (such as apical constriction or cell elongation) can summate to produce tensions which result in the expansion, contraction or folding of the cell sheet, which behaves as a deformable solid. Tissues may also flow in the manner of a viscous liquid, a process that has sometimes been overlooked. In such cases cells may retain their original shapes but move past one another, changing cell neighbours (Phillips et al. 1977; Phillips & Steinberg, 1978). These movements of cells relative to one another have been termed ‘cell shear’ (Jacobson & Gordon, 1976), ‘cell slippage’ (Phillips et al. 1977; Phillips & Steinberg, 1978) and ‘cell rearrangement’ (Fristrom, 1976). Such fluid rearrangements of cells occur during pronephric duct rudiment extension. Cell slippage might result either from active cell movements or from passive relaxation of tensions externally imposed on a tissue mass during morphogenesis. Recent detailed studies of cell movements in embryos, made possible at least in part by advances in SEM techniques (see review by Poole & Steinberg, 1977), have made it increasingly clear that movement of cells in groups or streams is more common than previously realized (Trinkaus, 1976). Unfortunately, the mechanisms and forces mediating and directing such morphogenetic movements in vivo have remained obscure. Elsewhere we present evidence that the same thermodynamic principles which govern adhesion-mediated cell sorting and tissue spreading in vitro (reviewed by Steinberg, 1978 a, b) also operate within embryos to direct cell migrations and stabilize anatomically ‘correct’ cell associations (Poole & Steinberg, 1978, 1981).

We thank Edward Kennedy, Doris White, Pam Knab-Mclntyre and Dorothy Spero for their technical assistance. This study was supported by research grants PCM76-84588 from the National Science Foundation and CA13605 from the National Cancer Institute, and by PHS training grant CA9167 from the National Cancer Institute. The electron microscopy was carried out in Department of Biology facilities supported by the Whitehall Foundation. From a dissertation submitted by T.J.P. to the Department of Biology, Princeton University, in partial fulfillment of the requirements for the Ph.D. degree.

Bancroft
,
M.
&
Bellairs
,
R.
(
1976
).
The development of the notochord in the chick embryo studied by scanning and transmission electron microscopy
.
J. Embryol. exp. Morph
.
35
,
383
401
.
Bijtel
,
J. H.
(
1948
).
Déviation experimental de l’uretère primaire par la greffe de materiel cloacal chez le Triton
.
Experientia
4
,
1
5
.
Burns
,
R. K.
(
1955
).
Urogenital system
.
In Analysis of Development
, (ed.
B. H.
Willier
,
P.
Weiss
&
V.
Hamburger
), pp.
462
468
.
Philadelphia
:
W. B. Saunders and Co
.
Camber
,
R.
&
Gipouloux
,
J. D.
(
1970
).
Les facteurs determinant la migration des cellules de Furètere primaire sont spécifiques. Démonstration expérimentale chez les amphibiçns anoures
.
C. r. hebd Séanc Acad. Sci
.,
Paris
270
,
1359
1361
.
Ebendal
,
T.
(
1976
).
The relative roles of contact inhibition and contact guidance in orientation of axons extending on aligned collagen fibrils in vitro
.
Expl Cell Res
.
98
,
159
169
.
Ebendal
,
T.
(
1977
).
Extracellular matrix fibrils and cell contacts in the chick embryo. Possible roles in orientation of cell migration and axon extension
.
Cell Tissue Res
.
175
,
439
458
.
Fox
,
H.
(
1963
).
The amphibian pronephros
.
Q. Rev. Biol
.
38
,
1
25
.
Fox
,
H.
&
Hamilton
,
L.
(
1964
).
Origin of the pronephric duct in Xenopus laevis
.
Archs Biol. (Liege)
75
,
245
251
.
Fristrom
,
D.
(
1976
).
The mechanism of évagination of imaginai discs of Drosophila melanogaster. III. Evidence for cell rearrangement
.
Devi Biol
.
54
,
163
171
.
Gipouloux
,
J. D.
&
Cambar
,
R.
(
1961
).
Recherches sur les facteurs déterminant la migration des cellules de l’uretère primaire (Amphibien Anoures
).
C.r. hebd. Séanc Acad. Sci
.,
Paris
252
,
3643
3645
.
Hogan
,
J. C.
, Jr
. &
Trinkaus
,
J. P.
(
1977
).
Intercellular junctions, intramembranous particles and cytoskeletal elements of deep cells of the Fundidas gastrula
.
J. Embryol. áxp. Morph
.
40
,
125
141
.
Holtfreter
,
J.
(
1944
).
Experimental studies on the development of the pronephros
.
Rev. Can. Biol
.
3
,
220
250
.
Jacobson
,
A. G.
(
1967
).
Amphibian cell culture, organ culture and tissue dissociation
.
In Methods in Developmental Biology
, (ed.
F. H.
Wilt
&
N. K.
Wessells
), pp.
531
542
.
New York
:
T. Y. Crowell Co
.
Jacobson
,
A. G.
&
Gordon
,
R.
(
1976
).
Changes in the shape of the developing vertebrate nervous system analyzed experimentally, mathematically and by computer simulation
.
J. exp. Zool
.
197
,
191
246
.
Karnovsky
,
M. J.
(
1965
).
A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy
.
J. Cell Biol
.
27
,
137a
.
Keller
,
R. E.
(
1975
).
Dye mapping of the gastrula and neurula of Xenopus laevis. Prospective areas and morphogenetic movements of the superficial layer
.
Devi Biol
.
42
,
222
241
.
Lôfberg
,
J.
&
Ahlfors
,
K.
(
1978
).
Extracellular matrix organization and early neural Crest cell migration in the axolotl embryo
.
Zoon
6
,
87
101
.
Nakatsuji
,
N.
(
1975
).
Studies on the gastrulation of amphibian embryos: light and electron microscopic observation of a urodele Cynops pyrrhogaster
.
J. Embryol. exp. Morphy
34
,
669
685
.
Nakatsuji
,
N.
(
1976
).
Studies on the gastrulation of amphibian embryos: ultrastructure of the migrating cells of anurans
.
Wilhelm Roux’ Arch, devl Biol
.
180
,
229
240
.
Nieuwkoop
,
P. D.
(
1947
).
Experimental investigations on the origin and determination of germ cells and on the development of lateral plates and germ ridges in Urodeles
.
Arch, néerl. Zool
.
8
,
1
205
.
O’connor
,
R. J.
(
1938
).
Experiments on the development of the pronephric duct
.
J. Anat
.
13
,
145
154
.
Overton
,
J.
(
1959
).
Studies on the mode of outgrowth of the amphibian pronephric duct
.
J. Embryol. exp. Morph
.
7
,
86
93
.
Phillips
,
H. M.
&
Steinberg
,
M. S.
(
1978
).
Embryonic tissues as elesticoviscous liquids. I. Rapid and slow shape changes in centrifuged cell aggregates
.
J. Cell Sci
.
30
,
1
20
.
Phillips
,
H. M.
,
Steinberg
,
M. S.
and
Lipton
,
B. H.
(
1977
).
Embryonic tissues as elasticoviscous liquids. II. Direct evidence for cell slippage in centrifuged aggregates
.
Devi Biol
.
59
,
124
134
.
Poole
,
T. J.
and
Steinberg
,
M. S.
(
1977
).
SEM-aided analysis of morphogenetic movements: Development of the amphibian pronephric duct
.
In Scanning Electron Microscopy/1977 Vol. II
(ed
O.
Johari
), pp.
43
52
.
Chicago
:
ITT Research Institute
.
Poole
,
T. J.
&
Steinberg
,
M. S.
(
1978
).
Amphibian pronephric duct development: In situ analysis of directed tissue migrations
.
J. Cell Biol
.
79
,
337a
.
Poole
,
T. J.
&
Steinberg
,
M. S.
(
1981
).
Evidence for the guidance of pronephric duct migration by a craniocaudally travelling adhesion gradient
.
Devi Biol, (submitted for publication)
.
Shin-Ike
,
T.
(
1955
).
Further experiments on the development of the pronephric duct in amphibia
.
Annotnes zool.jap
.
28
,
215
226
.
Schreckenberg
,
G. M.
&
Jacobson
,
A. G.
(
1975
).
Normal stages of development of the axolotl, Ambystoma mexicanum
.
Devi Biol
.
42
,
391
399
.
Steinberg
,
M. S.
(
1978a
).
Cell-cell recognition in multicellular assembly: Levels of specificity
.
In Cell Cell Recognition
, (ed.
A. S. G.
Curtis
), pp.
25
49
.
New York
:
Cambridge University Press
.
Steinberg
,
M. S.
(
1978b
).
Specific cell ligands and the differential adhesion hypothesis: How do they fit together?
In Specificity of Embryological Interactions
, (ed.
D. R.
Garrod
), pp.
97
130
.
New York
:
John Wiley and Sons
.
Tosney
,
K. T.
(
1978
).
The early migration of neural crest cells in the trunk region of the avian embryo
.
Devi Biol
.
62
,
317
333
.
Trinkaus
,
J. P.
(
1973
).
Surface activity and locomotion of Fundulus deep cells during blastula and gastrula stages
.
Devi Biol
.
30
,
68
103
.
Trinkaus
,
J. P.
(
1976
).
On the mechanism of metazoan cell movements. In ‘The Cell Surface in Animal Embryogenesis and Development
, (ed.
G.
Poste
&
G. L.
Nicolson
).
Cell Surface Reviews
1
,
225
329
.
Trinkaus
,
J. P.
and
Lentz
,
T. L.
(
1967
).
Surface specialization of Fundulus cells and their relation to cell movements during gastrulation
.
J. Cell Biol
.
32
,
121
138
.
Tung
,
T. C.
and
Ku
,
S. H.
(
1944
).
Experimental studies on the development of the pronephric duct in anuran embryos
.
J. Anat
.
78
,
52
57
.
Wylie
,
C. C.
,
Heasman
,
J.
,
Swan
,
A. P.
and
Anderton
,
B. H.
(
1979
).
Evidence for substrate guidance of primordial germ cells
.
Expl Cell Res
.
121
,
315
324
.