Chick neural tubes were cultured either on planar substrata of collagen-coated Falcon plastic in growth medium with serum or within a hydrated collagen lattice (HCL) in growth medium either with or without serum. Using time-lapse cinemicrography, neural crest cells were observed emigrating from neural tubes over the collagen substrata. Once separated from the neural tube, they seldom reunite with it. Though the average rate at which the neural crest cells translocate was the same in the different culture conditions, approximately 1·0 μm/min, distinct differences in morphology and mode of translocation were observed. Neural crest cells on collagen-coated culture dishes have a flattened fibroblastic morphology and mode of translocation; in an HC1 with serum, they have a bipolar shape and translocate by advancing a long, narrow leading protrusion and by periodically retracting the attenuated trailing portion of the cell; and in a serum-free HCL, they have a unipolar shape and translocate by advancing a long, narrow, branched leading protrusion and by periodically transferring the cytoplasm of the large, rounded trailing cell body forward, past a bulbous structure, and into the leading protrusion.

Some embryos are sufficiently translucent to allow observation of cell behavior in their interiors, but the majority are opaque and thus restrict our view to their surfaces. How then can we observe the translocation (i.e. locomotion from one place to another) of cells within these opaque embryos? Since we cannot observe them in vivo, we may study their movement in vitro. Culturing, however, often alters the morphology and mode of translocation, presumably because of dissimilarities in the in vivo and in vitro environments.

It seems possible that the culture environment may be modified to simulate the fibrillar nature of the living environment. In such an attempt, Bard & Hay (1975) cultured chick corneal mesenchyme cells in a hydrated collagen lattice (Elsdale & Bard, 1972) and found their locomotory behavior and morphology to be similar to that in situ, although the collagenous lattice lacks glycosaminoglycans and other intercellular components normally found in vivo. In the hope that use of this culture technique might give us some insight into the mode of locomotion of other kinds of cells within opaque embryos, neural crest cells of the chick embryo were cultured in a similar way. Neural crest cells were selected for this study because they engage in intensive locomotion as individual cells during development (Weston, 1970) and, in part, utilize a fibrillar extracellular matrix as the substratum for their emigration from the neural tube (Ebendal, 1977).

The methods were to culture neural tubes on a flat substratum in growth medium with serum and within a hydrated collagen lattice, either with or without serum in the growth medium. Serum was omitted from the medium in order to determine what effects the presence of serum may have on the form and migratory behavior of neural crest cells. For at this stage of the development the circulatory system is still rudimentary (Hamilton, 1952) and the neural tube is not yet vascularized (Feeney & Watterson, 1946), hence there may be little or no serum in the normal environment of the migrating neural crest cells.

Rat tail collagen was used to simulate the collagenous portion of the extracellular matrix utilized by translocating neural crest cells in vivo (see Noden, 1978 for review). The locomotory behavior of the cells was recorded with time-lapse cinemicrography and compared to other cells that have been observed translocating within translucent embryos.

Preparation of collagen substrata

Stock solutions of rat tail collagen were prepared (Konigsberg, 1971) and diluted with 04 M acetic acid to a final concentration of 2·1 mg/ml. Estimates of the collagen concentration were made, assuming that 10% of the dry weight of collagen is hydroxyproline. The concentration of hydroxyproline was determined colormetrically (Bergman & Loxley, 1963) by hydrolysing lyophilized samples of collagen in 6 N-HCI (azotropic) at 110 °C in a N2 atmosphere for 24 and 48 h. The hydrated collagen lattice, HCL (Elsdale & Bard, 1972) was prepared by adding 50 of the collagen stock solution to 4·0 ml of the growth medium (Dulbecco’s Modified Eagle Medium containing 100 i.u./ml penicillin, 2·5 mg/ml Fungizone, and 1·0 mg/ml streptomycin), either with or without 10% fetal serum (Gibco), and adjusting the pH to 7·4 with approximately 20 μl of 0·2 M-NaOH. Falcon tissue culture dishes were coated with collagen as described by Konigsberg (1971).

Primary cultures of neural crest cells

Neural tubes were isolated from stage-14 (Hamilton, 1952) White Leghorn chick embryos (Spafas, Norwich, Conn.) by digesting with 0·1% trypsin (Trypsin 1:250, Gibco) in calcium- and magnesium-free phosphate-buffered saline, pH 7·4 for 45 min at 4 °C (Cohen, 1972). Though the neural tube could often be freed from the seventh to the most posterior somite with tungsten needle after digestion, any somite tissue remaining attached to the neural tube after treatment was excised and discarded. The neural tubes were then cultured directly on collagen-coated surfaces of Falcon plastic in growth medium with serum and incubated at 37 °C in a 5 % CO2, 95 % air atmosphere, or were suspended within an HCL, either with or without serum in the growth medium, sealed in coverglass culture chambers (Bellco Glass), and incubated at 37 °C. Cultures were incubated overnight before filming to allow time for the neural crest cells to begin their emigration from the neural tube.

Time-lapse cinemicrography

Time-lapse films of neural crest cells migrating from the neural tubes were made with a Bolex Camera and Sage Intervalometer. The camera was mounted on Nikon Model M and Zeiss inverted microscopes. Phase contrast and Nomarski interference optics were used throughout. The interval between frames was 8 or 12 sec. The film used was Kodak 16 mm Plus X Reversal Movie Film, type 7276, and was developed commercially. Films were analysed with the aid of a Photo-Optical Data Analyzer, model 224 A, L-W Photo, Inc. Van Nuys, California.

Collagen and neural crest cell migration

As previously observed, neural crest cells emigrate readily from explanted neural tubes on to the surface of collagen-coated plastic tissue culture dishes (Maxwell, 1976). An HCL also serves as a substratum for translocating neural crest cells. Single cells and streams of cells emigrate from and seldom reunite with neural tubes suspended in HCL cultures (Fig. 1). Thus, outward migration is encouraged but backward migration is somehow inhibited.

Fig. 1.

A frame from a time-lapse film of neural crest cells migrating from a neural tube (at the bottom of the field) suspended in an HCL with serum and cultured in sealed coverglass culture chambers at 37 °C. Note that some cells are in focus while others are not. This results from cells emigrating from the dorsal portion of the tube at various depths within the three-dimensional substratum of collagen. The culture was incubated over night before filming. Nomarski interference optics, × 230.

Fig. 1.

A frame from a time-lapse film of neural crest cells migrating from a neural tube (at the bottom of the field) suspended in an HCL with serum and cultured in sealed coverglass culture chambers at 37 °C. Note that some cells are in focus while others are not. This results from cells emigrating from the dorsal portion of the tube at various depths within the three-dimensional substratum of collagen. The culture was incubated over night before filming. Nomarski interference optics, × 230.

Identification of the cells emigrating from trypsin-treated neural tubes

A scanning electron microscopic examination of neural tubes cleaned with trypsin and freed of somites with a tungsten needle, as in these experiments, has shown them to be free of somitic tissue (Davis, 1979). Though small quantities of contaminating fibroblasts may remain undetected, it is unlikely they affect the analysis presented below, since the emigration of cells near the dorsal part of the neural tube is so extensive. Two distinct kinds of cells were observed at the dorsal surface of the neural tube: (1) neural crest cells, which are similar to those shown by Bancroft & Bellairs (1975 and 1976), Ebendal (1977) and Tosney (1978), and (2) an occasional small patch of ectoderm. The small amount of ectoderm presents no problem to the analysis of neural crest cell translocation for it grows as discrete patches of tissue in HCL cultures that are easily distinguished from neural crest cells.

Additional evidence supporting the contention that the primary tissue of emigration from trypsin-cleaned neural tubes is neural crest comes from three sources: (1) Neural tubes were isolated from chick embryos utilizing techniques previously developed and employed by others in their study of neural crest cell growth and differentiation in culture (Cohen, 1972 and Cohen & Konigsberg, 1975). (2) The morphology of the cells emigrating from neural tubes explanted onto a planar substratum of collagen-coated glass (Fig. 2) is very similar to that of cells cultured under similar condition and shown to be neural crest cells by other criteria (Cohen & Kongisberg, 1975). (3) No cells were observed to emigrate from the ventral portion of cornally bisected neural tubes suspended in HCL cultures during a 36 h culturing period, whereas many cells emigrated from their dorsal counter part (Davis, unpublished results).

Fig. 2.

A frame from a time-lapse film of neural crest cells cultured on a planar substratum consisting of collagen-coated Falcon plastic at 37 °C. The culture was incubated over night before filming. The neural tube is outside the field at the right. Phase contrast optics, × 460.

Fig. 2.

A frame from a time-lapse film of neural crest cells cultured on a planar substratum consisting of collagen-coated Falcon plastic at 37 °C. The culture was incubated over night before filming. The neural tube is outside the field at the right. Phase contrast optics, × 460.

Morphology of chick neural crest cells in culture

Single neural crest cells have a morphology similar to fibroblasts (Fig. 2), when cultured on a plane surface of collagen-coated Falcon plastic. In contrast, neural crest cells in HCL cultures have a variety of shapes, but generally maintain a bipolar symmetry. In an HCL with serum, single neural crest cells have a spindle shape with a long, narrow leading protrusion, a constriction segment composed of a pair of small constrictions near the center of the cell, and long trailing protrusion (Figs. 1 and 3). Morphologically, single neural crest cells in an HCL free of serum differ. They have a long, narrow leading protrusion with a highly branched leading edge, a bulbous structure composed of a bulging area between a pair of constrictions located just posterior to the leading protrusion, and a large, round trailing cell body (Figs. 4 and 5).

Fig. 3.

A sequence from time-lapse films of a neural crest cell translocating in an HCL with serum. Chick neural tubes were suspended in an HCL with serum and cultured in sealed coverglass culture chambers at 37 °C over night before filming to allow time for the neural crest cells to emigrate from the neural tube. Time is indicated in minutes and seconds. The predominant morphology of locomoting cells in this environment is shown in this sequence. Both leading and trailing edges advance during cellular translocation. As the cell elongates, constrictions near the center of the cell develop (arrows). These constrictions divide the cell into three segments: a leading protrusion (LP), a constriction segment (CS), and a trailing protrusion (TP). At the onset of the retraction phase, the trailing protrusion retracts, forms a sphere, and is incorporated into the leading protrusion. Progress in the translocation of the cell toward the lower left can be measured by reference to the stationary particle in the gel labeled by the triangle at the lower right starting with frame 57:05. Note that the orientation of the cell corresponds to that of the collagen fibers. Phase contrast optics, × 580.

Fig. 3.

A sequence from time-lapse films of a neural crest cell translocating in an HCL with serum. Chick neural tubes were suspended in an HCL with serum and cultured in sealed coverglass culture chambers at 37 °C over night before filming to allow time for the neural crest cells to emigrate from the neural tube. Time is indicated in minutes and seconds. The predominant morphology of locomoting cells in this environment is shown in this sequence. Both leading and trailing edges advance during cellular translocation. As the cell elongates, constrictions near the center of the cell develop (arrows). These constrictions divide the cell into three segments: a leading protrusion (LP), a constriction segment (CS), and a trailing protrusion (TP). At the onset of the retraction phase, the trailing protrusion retracts, forms a sphere, and is incorporated into the leading protrusion. Progress in the translocation of the cell toward the lower left can be measured by reference to the stationary particle in the gel labeled by the triangle at the lower right starting with frame 57:05. Note that the orientation of the cell corresponds to that of the collagen fibers. Phase contrast optics, × 580.

Fig. 4.

A sequence from a time-lapse film of a neural crest cell translocating within an HCL without serum in the growth medium. Chick neural tubes were suspended in an HCL without serum and cultured in sealed coverglass culture chambers at 37 °C over night before filming to allow time for the neural crest cells to emigrate from the neural tube. This sequence demonstrates the changes that occur in the cell shape during translocation. As the cell translocates towards the right, the leading edge is extended and periodically the cytoplasmic content of the trailing cell body (CB) is transferred past the bulbous structure (BS) and into the leading protrusion (LP). A new cell body and bulbous structure develop anteriorly, and the translocatory process is repeated. Time is indicated in minutes and seconds. Phase contrast optics, × 1660.

Fig. 4.

A sequence from a time-lapse film of a neural crest cell translocating within an HCL without serum in the growth medium. Chick neural tubes were suspended in an HCL without serum and cultured in sealed coverglass culture chambers at 37 °C over night before filming to allow time for the neural crest cells to emigrate from the neural tube. This sequence demonstrates the changes that occur in the cell shape during translocation. As the cell translocates towards the right, the leading edge is extended and periodically the cytoplasmic content of the trailing cell body (CB) is transferred past the bulbous structure (BS) and into the leading protrusion (LP). A new cell body and bulbous structure develop anteriorly, and the translocatory process is repeated. Time is indicated in minutes and seconds. Phase contrast optics, × 1660.

Fig. 5.

A sequence from a time-lapse film of a neural crest cell translocating in an HCL without serum in the growth medium. Chick neural tubes were suspended in an HCL without serum and cultured in sealed coverglass culture chambers at 37 °C overnight before filming to allow time for the neural crest cells to emigrate from the neural tube. Time is indicated in minutes and seconds. Typical morphology of single neural crest cells translocating within this environment is shown in this sequence. Note the rounded cell body and the branched leading protrusion. Both leading and trailing edges advance continuously during cellular translocation. The trailing edge advances in part by cytoplasmic flow, as in frames 12:20 through 17:00. Phase contrast optics, × 460.

Fig. 5.

A sequence from a time-lapse film of a neural crest cell translocating in an HCL without serum in the growth medium. Chick neural tubes were suspended in an HCL without serum and cultured in sealed coverglass culture chambers at 37 °C overnight before filming to allow time for the neural crest cells to emigrate from the neural tube. Time is indicated in minutes and seconds. Typical morphology of single neural crest cells translocating within this environment is shown in this sequence. Note the rounded cell body and the branched leading protrusion. Both leading and trailing edges advance continuously during cellular translocation. The trailing edge advances in part by cytoplasmic flow, as in frames 12:20 through 17:00. Phase contrast optics, × 460.

Rate of translocation

In contrast to the morphology of neural crest cells, their average rates of translocation were not greatly altered by the different culture conditions.

Single neural crest cells on flat collagen-coated substratum and in HCL cultures with and without serum translocate at nearly the same average rate of l·0 μm/min (Table 1).

Table 1.

The rates of translocation of chick neural crest cells*

The rates of translocation of chick neural crest cells*
The rates of translocation of chick neural crest cells*

Mode of translocation

The mode of translocation of neural crest cells is dependent on the culture conditions. On a flat collagen-coated substratum, neural crest cells translocate like fibroblasts. In an HCL with serum, however, they translocate by advancing a narrow, leading protrusion and by periodically retracting a long, attenuated trailing edge (Fig. 3). Upon retraction, the trailing edge apparently loses its adhesions to the collagen substratum and within 10 sec the cell length shortens and a rounded trailing cell body is formed (Fig. 3, frames 64:20 through 64:30). The protoplasmic extension of the trailing protrusion which had been part of the extended process in contact with the substratum can be seen as a phase-dark tail and is quickly incorporated into the rounded cell body (Fig. 3, frames 64:30 through 65:00), and retraction fibers form at the trailing edge of the rounded cell body (Fig. 3, frames 64:40 through 65:00). The rounded cell body slowly combines with the leading protrusion so that a uniform bipolar spindleshaped cell is formed (Fig. 3, frame 75:00). Thus, by the continuous extension of the leading edge and the periodic retraction of the trailing edge, the translocatory process continues.

When serum is omitted from the HCL cultures, neural crest cells translocate with much cytoplasmic flow. In an HCL without serum, neural crest cells extend a highly branched leading protrusion and periodically transfer the cytoplasm of the large, rounded trailing cell body through the bulbous structure and into the leading protrusion. As the cytoplasm flows, the diameter of the trailing cell body decreases and the diameter of the leading protrusion increases (Figs. 4 and 5, frames 12:20 through 17:00). Once the transfer of cytoplasm is complete, the old trailing cell body and bulbous structure are incorporated into the new trailing cell body and a new bulbous structure develops anteriorly (Fig. 4, frame 3:20). Occasionally, the leading protrusion shortens (Fig. 5, frames 16:10 through 17:00); however, this does not alter the shape of the trailing cell body or the advancement of the branched leading edge.

The rate of advancement of the leading edge of neural crest cells in HCL cultures with and without serum is nearly uniform, albeit it fluctuates in a series of small extensions and retractions (Fig. 6). In contrast, the trailing edge advances at two velocities, one slower and the other greater than the rate of advancement of the leading edge. The slow advancement of the trailing edge may result from distortions elicited in the collagen fibrils by the translocating cells. The rapid advancement of the trailing edge of neural crest cells in an HCL with serum occurs as the trailing edge loses its adhesions to the collagen substratum and springs forward. In an HCL without serum, the fast rate of advancement of the trailing edge of neural crest cells occurs as the cytoplasm of the cell body pours forward through the bulbous structure and into the leading protrusion.

Fig. 6.

Abscissa: time (min) cells were observed in locomotion; ordinate: distance traversed (μm) by the leading edge (○) and the trailing edge (Δ). The translocating activity of neural crest cells in HCL cultures with and without serum in the growth medium was recorded with time-lapse cinemicrography and the relationship between the rates of advancement of the leading and trailing edges during locomotion was compared. (A) Neural crest cells in an HCL with serum. This figure represents the translocatory behavior of one neural crest cells. The leading edge advances at nearly a constant rate (approximately 1·0 μm/min) which equals the average rate of cellular translocation, whereas the trailing edge advances at two different velocities that are either slower or faster than the average rate of cellular translocation. Though many cells were observed for shorter periods of time, only two cells were observed to translocate for nearly 60 min and in each case, the leading edge advanced at a nearly constant rate, whereas the advancement of the trailing edge was biphasic. The longer periods of observation were necessary to assure that the advancement of the trailing edge would repeat the slow and fast rates of advancement. In both cases, the advancement of the leading edge was nearly constant, whereas the advancement of the trailing edge was biphasic. (B) Neural crest cells in an HCL without serum. This figure represents the translocatory behavior of one neural crest cell. The rates of advancement of the leading and trailing edges on neural crest cells in an HCL without serum do not differ from the advancement rates of cells in serum. The leading edge advances at nearly a constant rate, whereas the trailing edge advances at two different rates. Three cells were observed to translocate for nearly 60 min and in each case, the leading edge advanced at nearly a constant rate, whereas the advancement of the trailing edge was biphasic.

Fig. 6.

Abscissa: time (min) cells were observed in locomotion; ordinate: distance traversed (μm) by the leading edge (○) and the trailing edge (Δ). The translocating activity of neural crest cells in HCL cultures with and without serum in the growth medium was recorded with time-lapse cinemicrography and the relationship between the rates of advancement of the leading and trailing edges during locomotion was compared. (A) Neural crest cells in an HCL with serum. This figure represents the translocatory behavior of one neural crest cells. The leading edge advances at nearly a constant rate (approximately 1·0 μm/min) which equals the average rate of cellular translocation, whereas the trailing edge advances at two different velocities that are either slower or faster than the average rate of cellular translocation. Though many cells were observed for shorter periods of time, only two cells were observed to translocate for nearly 60 min and in each case, the leading edge advanced at a nearly constant rate, whereas the advancement of the trailing edge was biphasic. The longer periods of observation were necessary to assure that the advancement of the trailing edge would repeat the slow and fast rates of advancement. In both cases, the advancement of the leading edge was nearly constant, whereas the advancement of the trailing edge was biphasic. (B) Neural crest cells in an HCL without serum. This figure represents the translocatory behavior of one neural crest cell. The rates of advancement of the leading and trailing edges on neural crest cells in an HCL without serum do not differ from the advancement rates of cells in serum. The leading edge advances at nearly a constant rate, whereas the trailing edge advances at two different rates. Three cells were observed to translocate for nearly 60 min and in each case, the leading edge advanced at nearly a constant rate, whereas the advancement of the trailing edge was biphasic.

Advancement of the leading edge during translocation

The advancement of the leading edge of neural crest cells in an HCL is not restricted by the rate of advancement of the trailing edge. In Fig. 6 A, the extension of the leading edge is not constant but fluctuates from the average rate of advancement which is represented by the straight line. This variability in rate, however, does not coincide with the rate of advancement of the trailing edge. Note that during periods just prior to the onset of the rapid advancement of the trailing edge (Fig. 6 A, 10 and 40 min), when the cell is near its maximum length, the rate of advancement of the leading edge is slightly greater than its average rate. This observation suggests that even at times when the cell is approaching its maximum length and is presumably under the greatest tension, the advancement of the leading edge is not impeded.

Cell length changes during translocation

The mode of translocation of neural crest cells in an HCL produces continuous variations in their cell length. These changes in cell length are produced by the different rates of advancement of the leading and trailing edges. Thus as neural crest cells translocate in an HCL, the cell length increases when the trailing edge advances at a slow rate and shortens when the trailing edge advances at a fast rate.

Behavior of neural crest cells in culture

Collagen clearly promotes rapid outgrowth of cells from the neural tube (Maxwell, 1976), whether coated on flat substrata of glass or plastic or in the form of an HCL. On a flat collagen-coated substratum, neither the morphology nor mode of translocation of neural crest cells appear to differ from that of fibroblast-type cells cultured in similar conditions (for summary see Trinkaus, 1976). As on flat substrata, there are many similarities between neural crest cells and fibroblasts of different tissue origins in HCL cultures with serum in the growth medium. Human embryonic lung fibroblasts (Elsdale & Bard, 1972), corneal mesenchyme (Bard & Hay, 1975), and neural crest cells all possess a bipolar, spindle shape, as they utilize the collagen fibrils as substrata for translocation. In contrast to the attenuated trailing process of fibroblasts on flat substrata and in an HCL with serum, however, neural crest cells in an HCL without serum maintain a rounded trailing cell body containing much of the cytoplasm of the cell, even when the cell is fully extended, and an attenuated branched leading protrusion. In the absence of serum, the cells translocate in part by pouring the cytoplasm of the large, rounded trailing cell body into the leading protrusion.

Although the morphology and mode of translocation of neural crest cells on planar and three-dimensional collagen substrata with and without serum in the growth medium differ greatly, their average rates of translocation are nearly equal and about the same as for other fibroblast-like cells cultured under similar conditions (Trinkaus, 1976), about 1 μm per min. Why neural crest cells in the three different cultures translocate at nearly the same average rate is unclear. In particular, it is puzzling that neural crest cells in an HCL without serum translocate at the same low rate, for their mode of translocation has been associated with rates of cellular translocation eight to ten times greater than that of fibroblasts. For example, Fundulus deep cells in vivo (Trinkaus, 1973) and polymorphonuclear leukocytes in vitro (de Bruyn, 1946; Ramsey, 1972) both show much cytoplasmic flow during locomotion and both translocate at an average rate of 8–10 μm/min. The significance of a rounded cell body and cytoplasmic flow for the rate of locomotion of neural crest cells is, therefore, unclear. Apparently, the association between rapid rates of translocation and cytoplasmic fluidity previously observed with other cells (cf. Trinkaus, 1976) cannot be generalized.

A second similarity between neural crest cells in HCL cultures with and without serum, is the rates at which neural crest cells advance their leading and trailing edges. In both cases, the leading edge advances at a nearly uniform rate and the trailing edge retracts at two different rates, one greater and the other slower than the rate of advancement of the leading edge. I have no explanation at present for this similarity.

Similarities between neural crest cells in HCL cultures and embryonic cells in vivo

Although this study has been concerned with the behavior of neural crest cells in vitro, there are some remarkable similarities between neural crest cells in HCL cultures and other types of embryonic cells observed directly in vivo. In an HCL with serum, neural crest cells appear and translocate like corneal mesenchyme cells migrating within an HCL with serum or through the collagenous stroma between the endothelium and epithelium of the developing eye of a chick embryo in situ (Bard & Hay, 1975). Both neural crest cells and corneal mesenchyme cells have spindle-shaped bodies and long, branched filopodia. In contrast, neural crest cells in an HCL without serum appear much like the primary mesenchyme of echinoderms (Gustafson & Wolpert, 1961) and Fundulus deep cells translocating over the enveloping layer and yolk syncytial layer of gastrula embryos (Trinkaus, 1973) in that they possess a rounded cell body and a long, extended leading protrusion. In addition, the mode of transloction of neural crest cells and Fundulus deep cells is similar, with each translocating, in part, by pouring the cytoplasm of the large, rounded trailing cell body into the leading protrusion.

The purpose of studying neural crest cells in HCL cultures was to simulate the collagenous component of their fibrillar, in vivo environment in order to gain some insight into the possible appearance and mode of translocation of neural crest cells in vivo. It was found, interestingly, that the morphology of the trailing end and mode of translocation of neural crest cells in HCL cultures are both dependent on the presence or absence of serum in the growth medium. It is, of course, possible that serum in the growth medium selects for different cell types. This seems unlikely, however, since cells in the presence or absence of serum in HCL cultures are similar in many ways, including the morphology of their leading protrusion, constriction near the center of the cell, and the rate at which they translocate. It seems likely, therefore, that the same types of cells are being observed in the presence and absence of serum.

This study has shown that the mode of translocation of neural crest cells is related to their shape. Since their translocation in vivo cannot be observed because of the opacity of the chick embryo, it may be possible to deduce how neural crest cells translocate in vivo by observing their form. One can observe their form by fixing embryos during times when neural crest cells are known to be translocating and examining the neural crest cells with SEM. This has been done recently in three independent studies (Bancroft & Bellairs, 1976; Ebendal, 1977; Tosney, 1978). The works of Bancroft & Bellairs (1976) and Tosney (1978) show that neural crest cells near the edge of the outgrowth are elongated. These cells appear similar to that of neural crest cells in HCL cultures with serum. In contrast, however, no one using SEM to observe neural crest form has reported observing neural crest cells with a rounded trailing cell body, such as found in an HCL lacking serum. In consequence, it seems reasonable to predict that at least some of the neural crest cells in vivo translocate like cells in HCL cultures with serum, that is without massive cytoplasmic flow. If this is so, an important matter in future investigation of the mechanism of the translocation of neural crest cells will be to determine which components of the extracellular matrix or serum are responsible.

I am indebted to Professor J. P. Trinkaus for his assistance in preparing this manuscript and to Dr Ray Keller and Mr Wen-Tien Chen for their discussion and assistance. This research has been supported by grants from the NSF (BMS 70-00610) and the NIH (USPHS-HD-07137) to J. P. Trinkaus and by postdoctoral fellowships from the American Cancer Society (PF-1330) to E. M. Davis and a NIH Cell and Developmental Biology Training Grant (USPHS-HD-00032-13) to the Department of Biology of Yale University.

Bancroft
,
M.
&
Bellairs
,
R.
(
1975
).
Differentiation of the neural plate and neural tube in young chick embryo: A study by scanning and transmission electron microscopy
.
Anat. Embryol
.
147
,
309
335
.
Bancroft
,
M.
&
Bellairs
,
R.
(
1976
).
The neural crest cells of the truck region of chick embryo studied by SEM and TEM
.
Zoon
4
,
73
85
.
Bard
,
J. B. L.
&
Hay
,
E. D.
(
1975
).
The behavior of fibroblasts from the developing avian cornea. Morphology and movement in situ and in vitro
.
J. Cell Biol
.
67
,
400
418
.
Bergman
,
I.
&
Loxley
,
R.
(
1963
).
Two improved and simplified methods for the specto-photometric determination of hydroxyproline
.
Anal. Chem
.
35
,
1961
1965
.
Bruyn
,
P. P. H. DE
(
1946
).
The amoeboid movement of the mammalian leukocyte in tissue culture
.
Anat. Rec
.
95
,
177
192
.
Cohen
,
A. M.
(
1972
).
Factors directing the expression of sympathetic nerve traits in cells of neural crest origin
.
J. exp. Zool
.
179
,
167
182
.
Cohen
,
A. M.
&
Konigsberg
,
I. R.
(
1975
).
A clonal approach to the problem of neural crest cell determination
.
Devi Biol
.
46
,
262
280
.
Davis
,
E. M.
(
1979
).
Contact relations of neural crest cells during locomotion within a hydrated collagen lattice
.
(In preparation.)
Ebendal
,
T.
(
1977
).
Extracellular matrix fibrils and cell contacts in chick embryo: Possible roles in orientation of cell migration and axon extension
.
Cell Tissue Res
.
175
,
439
458
.
Elsdale
,
T.
&
Bard
,
J.
(
1972
).
Collagen substrata for studies on cell behavior
.
J. Cell Biol
.
54
,
626
637
.
Feeney
,
J. F.
&
Watterson
,
R. L.
(
1946
).
The development of the vascular pattern within the walls of the central nervous system of the chick embryo
.
J. Morph
.
78
,
231
303
.
Gustafson
,
T.
&
Wolpert
,
L.
(
1961
).
Studies in the cellular basis of morphogenesis in the sea urchin embryo. Directed movements of primary mesenchyme cells in normal and vegetalized larvae
.
Expl Cell Res
.
24
,
64
79
.
Hamilton
,
H. L.
(
1952
).
Lillie’s Development of the Chick
(ed.
B. H.
Wilier
), p.
443
.
New York
:
Holt, Rinehart, & Winston Inc
.
Konigsberg
,
I. R.
(
1971
).
Diffusion-mediated control of myoblast fusion
.
Devi Biol
.
26
,
133
152
.
Maxwell
,
G. D.
(
1976
).
Substrate dependence of cell migration from explanted neural tubes in vitro
.
Cell Tissue Res
.
172
,
325
330
.
Noden
,
D. W.
(
1978
).
Interaction directing the migration and cyto-differentiation of avian neural crest cells
.
In The Specificity of Embryological Interactions
(ed.
D.
Garrod
).
Chapman and Hall
.
Ramsey
,
W. S.
(
1972
).
Locomotion of human polymorphonuclear leukocytes
.
Expl Cell Res
.
72
,
489
501
.
Remington
,
R. D.
&
Schork
,
M. A.
(
1970
).
Statistics with Application to the Biological and Health Sciences
.
Englewood Cliffs, New Jersey
:
Prentice-Hall Inc
.
Tosney
,
K. W.
(
1978
).
The early migration of neural crest cells in the trunk region of the avian embryo: an electron microscopic study
.
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
).
Amsterdam, New York and Oxford
:
North Holland Publishing Company
.
Weston
,
J. A.
(
1970
).
The migration and differentiation of neural crest cells
.
Adv. Morph
.
8
,
41
114
.