To determine whether expansion of the splanchnic mesoderm of the area vasculosa is influenced by the entodermal substratum on which it occurs, entoderm was separated from a small area of splanchnic mesoderm. The splanchnic mesoderm then contracted and thickened, decreasing to 7 % of its original area in 16 h. By then entoderm had reattached to most of it, and it expanded, reaching 11 % of its original area by 24 h. It was concluded that attachment to entoderm may be required for expansion of the splanchnic mesoderm, but the small amount of expansion obtained made this conclusion tentative.

For technical reasons subsequent investigation was done on mesodermal transplants, which attached to the host’s entoderm in 6 h, by which time they had contracted to 15% of their original area. They then expanded, reaching 30% by 16 h and 49% by 24 h. The onset of their expansion was also accompanied by the formation of connexions between their blood vessels and those of the host, and by the resumption of blood flow in them. To see whether their expansion was due to resumption of blood flow or to attachment to entoderm, other transplants were made in which the middle one-third was separated from the host’s entoderm by a piece of Millipore filter. This portion failed to expand although it became connected to the host’s blood vessels and flow of blood resumed in it, while the two lateral thirds, which regained attachment to entoderm, expanded. Transplants were also rotated so that their splanchnic mesoderm attached to ectoderm instead of entoderm. These transplants also formed connexions with the host’s vessels and blood flow resumed in them, but they expanded only slightly compared to non-rotated controls, in which the splanchnic mesoderm attached to entoderm.

It was concluded that while flow of blood undoubtedly promotes splanchnic mesodermal expansion as others have shown, attachment of the splanchnic mesoderm to entoderm is also important, and without it the promotive effect of blood flow does not occur. Evidence was also obtained that attachment to entoderm maintains the thinness of the splanchnic mesoderm, and that a vascular growth stimulus may be produced by the unvascularized entoderm distal to the mesoderm.

The yolk sac of the chicken embryo is initially avascular, consisting of two epithelial layers, the ectoderm and entoderm. Blood vessels develop in the mesoderm, which spreads into the yolk sac from the body of the embryo between the ectoderm and entoderm, separating them as it advances. The expanding region occupied by this mesoderm is the area vasculosa, and that into which it spreads is the area vitellina (Fig. 1 A). Previously attention has been given to the role which the ectoderm might play as a substratum for this mesodermal expansion, and to a group of cells at the edge of the mesoderm which adhere to the ectoderm and move outward in contact with it (Augustine, 1970, 1977 a; Mayer & Packard, 1978; Mayer, Hay & Hynes, 1981). The present report concerns the role of the entoderm in mesodermal expansion.

Fig. 1.

(A) Yolk sphere with stage-18 to -20 embryo. Dotted line indicates area of entoderm separation. Circle at lower right corner indicates hole in ectoderm through which forceps and probe are inserted. (B) Section through region of circle in (A) to show separation of entoderm from splanchnic mesoderm, bm, Basement membrane; ec, edge cells; ect, ectoderm; ent, entoderm; som, somatic mesoderm; spl, splanchnic mesoderm; st, sinus terminalis.

Fig. 1.

(A) Yolk sphere with stage-18 to -20 embryo. Dotted line indicates area of entoderm separation. Circle at lower right corner indicates hole in ectoderm through which forceps and probe are inserted. (B) Section through region of circle in (A) to show separation of entoderm from splanchnic mesoderm, bm, Basement membrane; ec, edge cells; ect, ectoderm; ent, entoderm; som, somatic mesoderm; spl, splanchnic mesoderm; st, sinus terminalis.

During the formation and early expansion of the mesoderm interactions occur between it and the entoderm which affect the development of both. Differentiation of mesenchyme into endothelium and erythrocytes is promoted by its association with the entoderm (Wilt, 1965; Mato, Aikawa & Kishi, 1964). This interaction is transmissible through a Millipore filter separating the two layers and preventing intercellular contact between them (Miura & Wilt, 1969). In the reverse direction, as the mesoderm moves outward it acts on the subjacent entoderm, promoting the formation by it of a columnar epithelium from a dispersed arrangement of cells (Bremer, 1960; Bellairs, 1963; Bennett, 1973; Milos, Zalek & Phillips, 1979) and the synthesis in it of the enzyme cysteine lyase (Bennett, 1973).

The present work has been done on embryos of days of incubation (Stage 18-20 of Hamburger & Hamilton, 1951). By this time the mesoderm has split into two layers, somatic, lining the ectoderm, and splanchnic, lining the entoderm and containing the blood vessels (Fig. 1B). The two layers remain joined at their edge and at scattered points proximal to the edge, and they continue to expand as a unit, but they are largely separated by the coelomic space. In the splanchnic mesoderm a pattern of large and small vessels is present which has developed from an earlier capillary network. Both arteries and veins are present, but their walls are relatively undifferentiated, consisting of endothelium surrounded by mesenchyme. Mesenchyme also fills the spaces between vessels as a thin layer laying on the entoderm. The interactions between entoderm and mesoderm described at earlier stages may still be occurring, since the processes affected by them are apparently continuing: in the mesoderm erythropoiesis continues, especially in distal regions (Houser, Ackerman & Knouff, 1961), and differentiation of mesenchyme into endothelium may still be occurring at stages as late as this (Sabin, 1922; Hughes, 1937). In the entoderm formation of a columnar epithelium in response to overpassage by the mesoderm continues. In addition, changes in entodermal intracellular yolk drops suggest that yolk digestion has begun and that absorption of the resulting nutrients into the overlying bloodstream in the mesoderm is occurring (Bellairs, 1963).

These reports indicate that a close relationship exists between the entoderm and splanchnic mesoderm. Interactions occurring between them may be part of the mechanism by which the mesoderm expands, as well as differentiates. To investigate this possibility changes in area of the mesoderm have been observed following its separation from the entoderm in portions of the area vasculosa. The effects of reassociating the two layers and of replacing the entoderm with other substrata have also been observed. The results indicate that the entoderm plays an important role in the mechanism of mesodermal expansion. An abstract of part of this work has been published previously (Augustine, 1977b).

White Leghorn eggs having embryos at stage 18–20 were windowed and part of the albumen was removed as described previously (Augustine, 1970). A portion of the area vasculosa between 10 and 25 mm2 in area extending from the edge proximally was photographed. Entoderm was then peeled from the splanchnic mesoderm in this portion with watchmaker’s forceps and a dental probe (Fig. 1B) as done somewhat similarly by Thomas (1934). The separation was also extended distally about 1 mm into the area vitellina separating the entoderm from the basement membrane of the ectoderm. The whole separation took 20–40 min. Upon completion this portion was again photographed, and the egg was sealed with cellophane tape and returned to the incubator. It was photographed at subsequent intervals of 6, 16 and 24 h.

In other experiments, after being photographed in situ a piece of mesoderm was taken from the distal one-third of the area vasculosa of a donor embryo and transplanted to the area vitellina of a host. To obtain this mesoderm the donor ectoderm was first removed from its basement membrane as described previously (Augustine, 1970). The entoderm was then separated from the splanchnic mesoderm of the same portion, leaving splanchnic and somatic mesoderm, with ectodermal basement membrane attached to the latter (Fig. 2). A piece of this preparation having its distal edge approximately 1 mm proximal to the sinus terminalis was then cut out with scissors and transferred to the host. The area vasculosa edge with its special cells which project distally into the area vitellina was thus absent from the transplants. A space for reception of the transplant was then made in the host’s area vitellina by separating the entoderm from the basement membrane of the ectoderm. The transplant was pushed by probe into this space and pressed firmly against the ectodermal basement membrane, where it tended to adhere. The host’s vitelline membrane was then torn away completely, resulting in an immediate expansion of the yolk sphere. This tended to flatten out the sagging entoderm of the host, hastening its attachment to the splanchnic mesoderm of the transplant. The ectodermal basement membrane was not removed from the transplant because it appeared to adhere to the ectodermal basement membrane of the host; its removal from earlier transplants frequently resulted in their failing to adhere and rolling up. The transplant was photographed immediately after completing this procedure and subsequently at 6, 16 and 24 h. In a third series a piece of Millipore filter (TH) having average pore size 0·45 μm and thickness of approximately 25 μm was inserted between the host’s entoderm and the middle one-third of a transplant prepared in the same way. The transplant had been marked with carbon particles in four places, designating three approximately equal portions, by rubbing it between two probes, one of which was coated with dried India ink. The piece of filter was positioned directly under the middle two marks (Fig. 6 A). The area of transplants and of regions of entoderm removal was measured by tracing their outline in the photographs taken at each interval and counting the number of squares the outline contained when placed on graph paper.

Fig. 2.

Transplantation of mesoderm. Donor: ectoderm has been removed from its basement membrane, and entoderm from splanchnic mesoderm, in area vasculosa. A piece of the remaining tissues indicated by dotted lines is then excised and transferred to host. Host: mesodermal transplant has been placed between entoderm and basement membrane of ectoderm in area vitellina.

Fig. 2.

Transplantation of mesoderm. Donor: ectoderm has been removed from its basement membrane, and entoderm from splanchnic mesoderm, in area vasculosa. A piece of the remaining tissues indicated by dotted lines is then excised and transferred to host. Host: mesodermal transplant has been placed between entoderm and basement membrane of ectoderm in area vitellina.

All incubation was at 38 °C. Tissues were fixed in Bouin’s fluid, sectioned at 10 μm and stained with PAS and haematoxylin.

Results are given as the mean and standard error of the mean.

Group 1. Separation of entoderm from splanchnic mesoderm

Entoderm was separated from splanchnic mesoderm in nine embryos. Rapid contraction and thickening of the splanchnic mesoderm resulted (Fig. 3). Within 1 min small vessels began to constrict and shorten, and circulation of blood in them stopped. By the time entoderm had been separated from the entire 10–25 mm2 region (20–40 m) the mean area of the region had decreased to 35·5 ± 2·7% of its original size; by 6 h it had decreased to 10·2 ±0·9% (Fig. 3B, C), and by 16 h to 7·4 ± 0·8 %. Sections of tissue from embryos fixed at 6 h showed that the splanchnic mesoderm was still entoderm-free, but at 16 h entoderm had reattached to most of the splanchnic mesoderm, which was greatly thickened. Circulation had also resumed at this time. From 16 to 24 h the splanchnic mesoderm expanded to 11·0± 1·4% of its original area. In sections of tissue fixed at 24 h entoderm had reattached to all of the splanchnic mesoderm which nevertheless remained greatly thickened (Fig. 3D).

Fig. 3.

Contraction and thickening of splanchnic mesoderm after separation from entoderm. (A) Portion of area vasculosa before separation of entoderm from splanchnic mesoderm. Upper two arrows point to same positions on sinus terminalis as those in B, and lower two arrows point to same branches as those in B. (B) Same portion of area vasculosa 6 h after separation. (C) Section of area vasculosa fixed 6 h after separation of entoderm from splanchnic mesoderm. Entoderm has not yet reattached. Ectoderm and somatic mesoderm have been wrinkled by contraction of splanchnic mesoderm, which is thicker than normal (compare Fig. 5 B). (D) Section of area vasculosa fixed 24 h after separation. Entoderm has reattached, but splanchnic mesoderm is greatly thickened and architecturally rearranged, with blood vessels present throughout its thickness, ect, Ectoderm; ent, entoderm; som, somatic mesoderm; spl, splanchnic mesoderm.

Fig. 3.

Contraction and thickening of splanchnic mesoderm after separation from entoderm. (A) Portion of area vasculosa before separation of entoderm from splanchnic mesoderm. Upper two arrows point to same positions on sinus terminalis as those in B, and lower two arrows point to same branches as those in B. (B) Same portion of area vasculosa 6 h after separation. (C) Section of area vasculosa fixed 6 h after separation of entoderm from splanchnic mesoderm. Entoderm has not yet reattached. Ectoderm and somatic mesoderm have been wrinkled by contraction of splanchnic mesoderm, which is thicker than normal (compare Fig. 5 B). (D) Section of area vasculosa fixed 24 h after separation. Entoderm has reattached, but splanchnic mesoderm is greatly thickened and architecturally rearranged, with blood vessels present throughout its thickness, ect, Ectoderm; ent, entoderm; som, somatic mesoderm; spl, splanchnic mesoderm.

These results suggest that attachment of splanchnic mesoderm to entoderm is necessary for its expansion since following its separation from entoderm it stopped expanding and contracted, and as entoderm reattached to it, it stopped contracting and expanded. However, this expansion was slight, and the fact that it followed entoderm reattachment may have been fortuitous (Fig. 4). Also, the greatly thickened, architecturally rearranged tissue in which it occurred (Fig. 3D) suggests that the processes producing it may have differed from the normal ones. Attempts to make reattachment occur earlier, before the mesoderm became greatly thickened, were not successful. However, it was found that pieces of mesoderm transplanted from the area vasculosa of donor embryos to the area vitellina of hosts attached to the host’s entoderm sooner, before much thickening occurred in them, and then expanded rapidly. Such transplants were therefore used for further study of entodermal influence on mesodermal expansion.

Fig. 4.

Area changes in Groups 1 and 2. Group 1, area of mesoderm before and at intervals after separation from entoderm. Group 2, area of mesodermal transplants before separation from entoderm and at intervals after separation, excision and transplantation. Arrows in both groups indicate time by which entoderm had reattached to most or all of the splanchnic mesoderm; they also indicate change from period of contraction to one of expansion.

Fig. 4.

Area changes in Groups 1 and 2. Group 1, area of mesoderm before and at intervals after separation from entoderm. Group 2, area of mesodermal transplants before separation from entoderm and at intervals after separation, excision and transplantation. Arrows in both groups indicate time by which entoderm had reattached to most or all of the splanchnic mesoderm; they also indicate change from period of contraction to one of expansion.

Group 2. Mesodermal transplants

Fourteen transplants were made. By the time their transplantation to the host was completed (40–60 m) they had contracted to a mean area 22·8 ± 1·1 % of their original size, and by 6 h to 15·1 ± 1·4 % (Figs 4, 5 A). Sections of transplants fixed at 6 h showed that the host’s entoderm had attached to the splanchnic mesoderm of the transplants, the thickness of which was only slightly greater than normal (Fig. 5B). By 16 h the advancing edge of the host’s area vasculosa had reached the transplants, connexions between host and transplant vessels were present, circulation of blood had resumed in the transplant vessels, and the transplants had expanded to 30·0 ±2·7% of their original size (Fig. 5C). Most of their original vascular pattern had been replaced by a capillary plexus. By 24 h they had expanded to 49·0 ± 3·6 % of their original size (Fig. 5D).

Fig. 5.

Mesodermal transplant. (A) 6 h after transplantation. (B) Section of a similar transplant fixed 6 h after transplantation. Entoderm of host has attached to splanchnic mesoderm of transplant, which is slightly thicker than normal. (C) 16 h after transplantation. (D) 24 h after transplantation.

Fig. 5.

Mesodermal transplant. (A) 6 h after transplantation. (B) Section of a similar transplant fixed 6 h after transplantation. Entoderm of host has attached to splanchnic mesoderm of transplant, which is slightly thicker than normal. (C) 16 h after transplantation. (D) 24 h after transplantation.

In this experiment attachment of entoderm to splanchnic mesoderm occurred earlier than in Group 1, resulting in an earlier resumption of expansion (Fig. 4) and cessation of thickening. Expansion also progressed much further, and the mesoderm in which it occurred was approximately normal in thickness and architecture. These results therefore further indicate an entodermal promotive effect on mesodermal expansion.

On the other hand, by the time expansion of the transplants was observed (16 h) they had become incorporated into the expanding area vasculosa of the host, and circulation of blood had resumed in them. Their expansion may have been caused by these events rather than by reattachment of entoderm to them, since expansion of the area vasculosa has been shown to be markedly promoted by the presence of the circulation in it (Chapman, 1918; Grodzinski, 1934; Patterson, 1909). To test this possibility transplants were made with a separator inserted between their middle one-third and the host’s entoderm to prevent reattachment of entoderm to splanchnic mesoderm in this portion while permitting its reattachment in the two lateral thirds.

Group 3. Separation of entoderm from mesoderm by a filter

After a number of materials had been tried as separators, pieces of Millipore filter proved satisfactory and were used in a group of ten transplants. The result of one of these, typical of the group, is shown in Fig. 6. As indicated by the change in distance between the carbon marks, the middle third of the transplant, separated from the entoderm by the filter, contracted slightly in 24 h while the two lateral thirds expanded. In both middle and lateral thirds connexions with the host’s area vasculosa were present and circulation of blood had resumed at 16 h as in Group 2, and circulation in a new pattern of vessels was present at 24 h in all three thirds. In sections of the transplants fixed at 24 h the splanchnic mesoderm in the two lateral thirds had reattached to entoderm but that in the middle third remained separated from entoderm by the filter. These results thus indicate that attachment of the splanchnic mesoderm to entoderm is important for its expansion, and the promotive effect of the circulation does not occur without it.

In addition to contracting the splanchnic mesoderm in the middle third also thickened (Fig. 6D). However, it did not contract or thicken as much as that in Group 1 (Fig. 3D). This might indicate a partial transmission through the filter of an entodermal effect; however, extensive regions were present in which either the entoderm or mesoderm was not closely applied to the filter surface but was separated from it by several microns. It seems unlikely that an interaction between the two tissues could be transmitted across such gaps. An alternative possibility is that the mesoderm got caught on the corners or edges of the filter or adhered to part of its surface and was thereby prevented from contracting and thickening more fully.

Fig. 6.

Mesodermal transplant with its middle one-third separated from host’s entoderm by a piece of Millipore filter. (A) Immediately after transplantation. (B) 6 h after transplantation. (C) 24 h after transplantation. Mean distance between middle two marks for the group of ten transplants was 84·7 + 4·4% of the distance at 0 h; between left middle and left lateral marks 145·2 ± 12·6%; between right middle and right lateral 166·3 + 10·7%. (D) Section through middle portion of transplant fixed 24 h after transplantation, ent, Host’s entoderm ; fit, filter.

Fig. 6.

Mesodermal transplant with its middle one-third separated from host’s entoderm by a piece of Millipore filter. (A) Immediately after transplantation. (B) 6 h after transplantation. (C) 24 h after transplantation. Mean distance between middle two marks for the group of ten transplants was 84·7 + 4·4% of the distance at 0 h; between left middle and left lateral marks 145·2 ± 12·6%; between right middle and right lateral 166·3 + 10·7%. (D) Section through middle portion of transplant fixed 24 h after transplantation, ent, Host’s entoderm ; fit, filter.

Group 4. Rotated mesodermal transplants

The effect of attachment to entoderm on splanchnic mesodermal expansion was tested in another way by rotating the transplant 180° around its radial axis (i.e. a line from the embryo body to the sinus terminalis) so that its splanchnic side faced the host’s ectoderm instead of entoderm, while its proximal and distal edges remained proximal and distal. In such a transplant the splanchnic mesoderm is separated from the host’s entoderm by the transplant’s coelomic space, somatic mesoderm and ectodermal basement membrane, and it attaches to the basement membrane of the host’s ectoderm (Fig. 7 A). Ten such transplants were made, and each was accompanied by a control transplant which was not rotated. Experimental and control transplants were taken from the same donor and placed side by side in the same host. After 24 h the mean area of the rotated transplants was 130·9±9·8% of their mean area at Oh, while that of the controls was 339·5 ± 62·0 % (Fig. 8). Both rotated and control transplants had become incorporated into the host’s area vasculosa and circulation was present in both. Since expansion was much greater in the splanchnic mesoderm of the controls attached to entoderm, than in that of the rotated transplants attached to ectoderm, the results again indicate the importance of entoderm to splanchnic mesodermal expansion and to the promotive effect which circulation has on expansion.

Fig. 7.

Diagrams of mesodermal transplants rotated 180° around their radial axis. (A) 0 h: transplant has been inserted between ectodermal basement membrane and entoderm of host; vessels of the transplant face the host’s ectoderm. (B) 24 h after transplantation. Rotated transplant is incorporated into host’s area vasculosa. Vessels of the transplant line the host’s ectoderm and have grown a short distance inward on the host’s entoderm from both edges of the transplant. (C) 48 h after transplantation. Host’s ectoderm and entoderm are both completely vascularized. Small outgrowth from distal edge of transplant is solid tissue having vessels but no coelom; this was seen in 8 of the 10 rotated transplants at 48 h.

Fig. 7.

Diagrams of mesodermal transplants rotated 180° around their radial axis. (A) 0 h: transplant has been inserted between ectodermal basement membrane and entoderm of host; vessels of the transplant face the host’s ectoderm. (B) 24 h after transplantation. Rotated transplant is incorporated into host’s area vasculosa. Vessels of the transplant line the host’s ectoderm and have grown a short distance inward on the host’s entoderm from both edges of the transplant. (C) 48 h after transplantation. Host’s ectoderm and entoderm are both completely vascularized. Small outgrowth from distal edge of transplant is solid tissue having vessels but no coelom; this was seen in 8 of the 10 rotated transplants at 48 h.

Fig. 8.

Rotated and control transplants. Edges have been marked with carbon in several places. (A) Control transplant immediately after transplantation (0 h). (B) Control transplant 24 h after transplantation. (C) Rotated transplant immediately after transplantation. (D) Rotated transplant 24 h after transplantation.

Fig. 8.

Rotated and control transplants. Edges have been marked with carbon in several places. (A) Control transplant immediately after transplantation (0 h). (B) Control transplant 24 h after transplantation. (C) Rotated transplant immediately after transplantation. (D) Rotated transplant 24 h after transplantation.

In sections of the rotated transplants fixed at 24 h the thickness of the splanchnic mesoderm attached to the host’s ectoderm was somewhat greater than normal in some cases, but normal in others. This was also true of the splanchnic mesoderm in the controls, attached to the host’s entoderm. Thus attachment to ectoderm maintains the thinness of the splanchnic mesoderm as effectively as attachment to entoderm even though it promotes expansion much less effectively.

In the same sections it was also noted that blood vessels were present in the proximal and distal portions of the somatic mesoderm of the rotated transplants, attached to the host’s entoderm (Fig. 7B). The vessels had apparently grown a short distance inward from the edges of the overlying splanchnic mesoderm (Fig. 9). In some cases vessels also extended across the coelom from splanchnic mesoderm somewhat inward from the edges of the transplant. To see if more of the entodermal side of rotated transplants would become vascularized if given more time, and if the transplants would expand more, ten more rotated transplants were made and incubated for 48 h. Seven of them were accompanied by non-rotated controls as before. At 48 h the mean area of these seven rotated transplants was 134·6 ±27·5% of their size at 0 h, while that of the controls was 745·7 ± 182·4%. In sections of the rotated transplants both somatic and splanchnic sides were completely vascularized in nine of the ten cases (Fig. 7C); in one case a portion of the somatic side near the centre of the transplant lacked vessels. This spreading of vessels in the rotated transplants from the side attached to ectoderm to cover the entire entodermal side is another indication of the promotive effect of the entoderm: vessels did not spread in the opposite direction -from entodermal to ectodermal side-in the control transplants.

Fig. 9.

Section through distal edge of rotated transplant fixed 24 h after transplantation. Vessels line the ectodermal side of the coelom and have grown a short distance proximally from the sinus terminalis on the entodermal side. Thickening of basement membrane above sinus terminalis resembles that normally present at the area vasculosa edge and, together with normal appearing edge cells, was found at the distal edge of both rotated and control transplants fixed 24 h after transplantation. Since all transplants were taken from regions proximal to edge in donors, sinus terminalis and associated edge features apparently regenerate, bm, Basement membrane; ect, ectoderm; ent, entoderm; st, sinus terminalis.

Fig. 9.

Section through distal edge of rotated transplant fixed 24 h after transplantation. Vessels line the ectodermal side of the coelom and have grown a short distance proximally from the sinus terminalis on the entodermal side. Thickening of basement membrane above sinus terminalis resembles that normally present at the area vasculosa edge and, together with normal appearing edge cells, was found at the distal edge of both rotated and control transplants fixed 24 h after transplantation. Since all transplants were taken from regions proximal to edge in donors, sinus terminalis and associated edge features apparently regenerate, bm, Basement membrane; ect, ectoderm; ent, entoderm; st, sinus terminalis.

The aim of the present work was to investigate the role of the entoderm in the expansion of the mesoderm in the area vasculosa. The following results indicate that the association of the splanchnic mesoderm with entoderm is important for its expansion : (1) expansion stopped and contraction and thickening occurred in splanchnic mesoderm separated from entoderm in Group 1, while contraction stopped and expansion resumed when attachment to the entoderm was regained. (2) In the mesodermal transplants of Group 2, in which splanchnic mesoderm regained attachment to entoderm earlier than in Group 1, expansion also resumed earlier. (3) In Group 3 portions of splanchnic mesoderm separated from entoderm by a Millipore filter contracted, while adjacent portions attached to entoderm expanded. (4) In the rotated transplants of Group 4, in which the splanchnic mesoderm was separated from entoderm by the coelom, somatic mesoderm and ectodermal basement membrane of the transplant, much less expansion occurred than in the control transplants, in which the splanchnic mesoderm was attached to entoderm.

The promotive effect of the entoderm on splanchnic mesodermal expansion may be due to a number of factors. One of these may be the effect of the entoderm on the flow of blood in the splanchnic mesoderm, since cessation of blood flow occurred when entoderm was removed and resumed following entodermal reattachment. The absence of blood flow has been shown to retard expansion of the area vasculosa in embryos without a circulation (Patterson, 1909; Chapman, 1918; Grodzinski, 1934). However, in the present work expansion was not merely retarded but stopped completely and replaced by contraction. Therefore, other factors in addition to blood flow must be involved. This is also evident from the results of Groups 3 and 4, in which portions of mesodermal transplants separated from entoderm contracted, or expanded less than those attached to entoderm even though flow of blood was present in them.

Flow of materials through the walls of vessels to or from their lumen has also been proposed as a factor causing their growth (Clark & Clark, 1939). Passage of nutrients from the yolk into the blood stream involves such a flow and is probably dependent on the entoderm for absorption from the yolk. Such flow would then be absent in splanchnic mesoderm separated from entoderm. This might account for the curtailment of expansion in the mesodermal transplants separated from entoderm in Groups 3 and 4, even though flow of blood was present in the lumina of their vessels.

Some of the material passing from the entoderm into the splanchnic mesoderm may remain there instead of entering the blood stream, causing an increase in the volume of the splanchnic mesoderm partly expressed as outward expansion. Bremer (1960) has reported a decrease in the volume of entoderm cells when they are overpassed by the mesoderm, and an almost equal increase in the volume of the overlying mesoderm. He considers that these changes are due to a transfer of entodermal contents to the mesoderm in a process involving active transport of ions by the mesoderm, and he reports finding evidence of an electric potential difference in support of this possibility. He suggests that the transferred material is derived from entodermal yolk inclusions and is used by the mesoderm in the production of new vasculature and colom (coelomic mesoderm?). While his measurements were done on earlier embryos (5–19 somites) than those used in the present work, Bremer notes that these changes increase with the age of the embryo. Interference with such a process by separation of entoderm from splanchnic mesoderm may be another factor contributing to the loss of mesodermal expansion in the present work.

The extended state and thinness of the splanchnic mesoderm under normal conditions may be a response of its cells to surface properties of the entoderm on which it spreads. The rapidity with which it contracts and thickens upon separation from entoderm resembles that of cells separated from their substratum in tissue culture, where such properties play an important role in determining the shape and arrangement of attached cells (Martin & Rubin, 1974; Bard & Hay, 1975). That the splanchnic mesoderm is responsive in this way to surface properties was suggested by the possibility that its decreased contraction and thickening on the surface of a Millipore filter in Group 3 may have occurred in response to the surface properties of the filter rather than to something transmitted through the filter from the entoderm on the other side. Miura & Wilt (1969) reported a tendency of area vasculosa ectomesoderm to move and flatten out on the surface of a Millipore filter when no entoderm was present on the other side. An abstract of some experiments testing this possibility has appeared previously (Augustine, 1978). The thinness of the splanchnic mesoderm is important for its rapid expansion because it maximizes the area covered by an increase in its volume, and maintenance of this thinness by its attachment to entoderm is therefore a component of the entodermal promotive effect on splanchnic mesodermal expansion. That other components are important also is indicated by the fact that thinness was maintained equally well in splanchnic mesoderm associated with ectoderm in Group 4, yet it expanded much less than that associated with ectoderm.

The splanchnic mesoderm of the area vasculosa spreads by growth of all of its parts, from center to edge, rather than by sprouting of vessels at its edge (Grodzinski, 1934). A stimulus for its growth has therefore been sought in the entoderm subjacent to it rather than in the unvascularized entoderm of the area vitellina distal to it. However, in the rotated transplants of Group 4 vessels spread from the edges of the transplant, and from across the coelom, to the unvascularized entoderm beneath them. This raises the possibility that a growth stimulus from the unvascularized entoderm may be still another component of the entodermal promotive effect, and one not shared by the ectoderm, which did not become vascularized in this way in the controls. Such a stimulus has been proposed in vascularization of other embryonic tissues and organs (Evans, 1909; Sims, 1964).

Although not of direct relevance to the present subject, the lability of the patterns formed by area vasculosa blood vessels was strikingly demonstrated in the transplants, in which, following the resumption of circulation the original pattern of large and small vessels rapidly changed to a capillary plexus (Fig. 5 A, C), in which a new pattern of large and small vessels subsequently formed. In contrast, Ausprunk, Knighton & Folkman (1975) have reported that in pieces of chick chorioallantoic membrane transplanted to the chorioallantois of hosts -these changes apparently do not occur, the original vascular pattern of the transplant remaining intact except in restricted loci.

Three other instances of mesodermal contraction resulting from separation from an epithelium or other membrane in the chick may be of interest in relation to the present work: (1) When ectoderm is removed from the apex or dorsal surface of the limb bud the subjacent mesoblast immediately contracts (Saunders, 1948). (2) When ectoderm is removed from its basement membrane in the area vasculosa the subjacent mesoderm undergoes a gradual contraction during the subsequent 16 h (Augustine, 1970). (3) When the splanchnic mesoderm of the allantois is prevented from attaching to the chorion in the formation of the chorioallantoic membrane by removal of the chorion its expansion slows and stops and is followed by contraction during the subsequent three days (Schnitzer, 1970; Symons, 1974). Whether the mechanisms in these three instances and that involved in the present work have anything in common is not known.

Part of this work was supported by NSF Grant GU 3683 to Hahnemann Medical College and Hospital.

Augustine
,
J. M.
(
1970
).
Expansion of the area vasculosa of the chick after removal of the ectoderm
.
J. Embryol. exp. Morph
.
24
,
95
108
.
Augustine
,
J. M.
(
1977a
).
Mesodermal expansion after arrest of the edge in the area vasculosa of the chick
.
J. Embryol. exp. Morph
.
41
,
175
188
.
Augustine
,
J. M.
(
1977b
).
Effects of entoderm removal on mesodermal expansion in the area vasculosa of the chick
.
Anat. Rec
.
187
,
527
.
Augustine
,
J. M.
(
1978
).
Changes in area and thickness of mesodermal expiants from the area vasculosa of the chick on three substrata
.
Anat. Rec
.
190
,
329
.
Ausprunk
,
D. H.
,
Knighton
,
D. R.
&
Folkman
,
J.
(
1975
).
Vascularization of normal and neoplastic tissues grafted to the chick chorioallantois
.
Am. J. Pathol
.
79
,
597
618
,
Bard
,
J. L.
&
Hay
,
E. D.
(
1975
).
The behaviour of fibroblasts from the developing avian cornea. Morphology and movement in situ and in vitro
.
J. Cell Biol
.
67
,
400
418
.
Bellairs
,
R.
(
1963
).
Differentiation of the yolk sac of the chick studied by electron microscopy
.
J. Embryol. exp. Morph
.
11
,
201
225
.
Bennett
,
N.
(
1973
).
Study of yolk-sac entoderm organogenesis in the chick using a specific enzyme (cysteine lyase) as a marker of cell differentiation
.
J. Embryol. exp. Morph
.
29
,
159
174
.
Bremer
,
H.
(
1960
).
Untersuchungen fiber die Entwicklung des Dotterentoderms der Hfihnerkeimscheibe
.
Wilhelm Roux Arch. EntwMech. Org
.
152
,
166
182
.
Chapman
,
W. B.
(
1918
).
The effect of the heart-beat upon the development of the vascular system in the chick
.
Am. J. Anat
.
23
,
175
204
.
Clark
,
E. R.
&
Clark
,
E. L.
(
1939
).
Microscopic observations on the growth of blood capillaries in the living mammal
.
Am. J. Anat
.
64
,
251
302
.
Evans
,
H. M.
(
1909
).
On the development of the aortal, cardinal and umbilical veins and other blood vessels of vertebrate embryos from capillaries
.
Anat. Rec
.
3
,
498
518
.
Grodzinski
,
Z.
(
1934
).
Zur Kenntnis der Wachstumvorgange der Area vasculosa beim Hiihnchen
.
Bull. Int. Akad. Pol. Sci. Lett. B
,
415
427
.
Hamburger
,
V.
&
Hamilton
,
H. L.
(
1951
).
A series of normal stages in the development of the chick embryo
.
J. Morph
.
88
,
49
92
.
Houser
,
J. W.
,
Ackerman
,
G. A.
&
Knouff
,
R. A.
(
1961
).
Vasculogenesis and erythropoiesis in the living yolk sac of the chick embryo
.
Anat. Rec
.
140
,
29
44
.
Hughes
,
A. F. W.
(
1937
).
Studies on the area vasculosa of the embryo chick. II. The influence of the circulation on the diameter of the vessels
.
J. Anat
.
72
,
1
17
.
Martin
,
G. R.
&
Rubin
,
H.
(
1974
).
Effects of cell adhesion to the substratum on the growth of chick embryo fibroblasts
.
Expl Cell Res
.
85
,
319
333
.
Mato
,
M.
,
Aikawa
,
E.
&
Kishi
,
K.
(
1964
).
Some observations on interstice between mesoderm and endoderm in the area vasculosa of chick blastoderm
.
Expl Cell Res
.
35
,
426
428
.
Mayer
,
B. W.
, Jr.
,
Hay
,
E. D.
&
Hynes
,
R. O.
(
1981
).
Immunocytochemical localization of fibronectin in embryonic chick trunk and area vasculosa
.
Devl Biol
.
82
,
267
286
.
Mayer
,
B. W.
, Jr.
&
Packard
,
D. S.
, Jr.
(
1978
).
A study of the expansion of the chick area vasculosa
.
Devl Biol
.
63
,
335
351
.
Milos
,
N.
,
Zalik
,
S. E.
&
Phillips
,
R.
(
1979
).
An analysis of the aggregation and morphogenesis of area opaca endoderm cells from the primitive-streak chick embryo
.
J. Embryol. exp. Morph
.
51
,
121
135
.
Miura
,
Y.
&
Wilt
,
F. H.
(
1969
).
Tissue interaction and formation of the first erythroblasts of the chick embryo
.
Devl Biol
.
19
,
201
211
.
Patterson
,
J. T.
(
1909
).
An experimental study of the development of the area vasculosa of the chick blastoderm
.
Biol. Bull
.
16
,
83
90
.
Sabin
,
F. R.
(
1922
).
Direct growth of veins by sprouting
.
Contr. Embryol. Carneg. Instn
.
14
,
1
10
.
Saunders
,
J. W.
(
1948
).
The proximo-distal sequence of origin of the parts of the chick wing and the role of the ectoderm
.
J. exp. Zool
.
108
,
363
403
.
Schnitzer
,
D. L.
(
1970
).
The effects of chorion removal on the growth of the allantois in the chick. Thesis, Hahnemann Medical College
.
Sims
,
R. T.
(
1964
).
The relationship between size and vascularity in the spinal cord of developing Xenopus laevis
.
J. Embryol. exp. Morph
.
12
,
491
499
.
Symons
,
A. R.
(
1974
).
Investigations into the mechanism of growth of the chorioallantoic membrane in the chick. Thesis, Hahnemann Medical College
.
Thomas
,
J. A.
(
1934
).
La culture prolongée de la paroi de la vésicule ombilicale chez l’embryon du poulet. La culture pure du syncytium entodermo-vitellin
.
Arch, exptl. Zeltforsch. Gewebezucht
.
15
,
131
148
.
Wilt
,
F. H.
(
1965
).
Erythropoiesis in the chick embryo: the role of endoderm
.
Science
147
.
1588
1590
.