Of Limb Morphogenesis in a Model System

DEVELOPMENTAL BIOLOGY 28, 113-122 (1972)

Analysis of Limb Morphogenesis in a Model System

ROBERT H. SINGER’. 2

Department of Biology, Brandeis University, Waltham, Massachusetts

Accepted January 27, 1972

A method for analysis of chicken limb morphogenesis was devised. This method consisted of grafting a limb ectodermal jacket containing dissociated and pelleted mesenchymal cellular components to the host somites. Different cellular components stuffed into the ectoderm could be mixed in varied ratios. After 7 days the grafts were analyzed for outgrowth. Stage 19 mesoblast cells alone when treated as above gave limblike outgrowths with good digits, toes, and claws in all cases. However, mesoblasts from the proximal half of older limbs (stages 24, 25 and chondrocytes) gave no outgrowths, and those from stage 23 gave outgrowths in 9% of the cases. In mixtures of 5% stage 19 cells with 95% chondrocytes, consistent morphogenesis (i.e., in 65% of grafts) occurred. The amount of morphogenesis (size of graft and perfection of digits) was directly propor- tional to the amount of stage 19 cells. However, these cells mixed with proximal cells of stages 23, 24, or 25 required higher proportions for equivalent morphogenesis. To obtain morphogenesis equivalent to the 5% mixture with chondrocytes, 10% stage 19 cells were needed when mixed with proximal stage 23, 25% with proximal stage 24 and 7% with proximal stage 25. Mixtures of stage 19 cells added to nonlimb (flank, stage- 19) mesoderm, formed large tumorous mounds of tissue with no limblike features.

INTRODUCTION ectoderm ridge serves as a specific induc- The limb bud presents a good model of tive structure initiating and directing out- a developing system. The relationship be- growth and digit formation from the early tween the mesoblast and its ectodermal bud. The mesoderm, upon which the ridge cover, and the histogenesis and cytodif- exerts its effect, in turn supports the in- ferentiation of cartilage allows an investi- tegrity of the ridge by producing a “main- gation of several levels of organization in tenance factor” (see review by Zwilling, a self-contained and easily manipulated 1961). More recently, experimental inter- body part. Beginning with the pioneering est has enlightened cellular events perti- work of Saunders (1948) and Zwilling nent to limb morphogenesis including: (1949) the reciprocal interaction between the onset of histogenesis (Searls, 1965; limb bud ectoderm and mesoderm was Medoff, 1967); the stability of differenti- uncovered and identified. The relation- ated cellular phenotypes in culture (Coon, ship served as a model for the analysis of 1966; Cahn and Cahn, 1966; Nameroff the nature and the role of different tissues and Holtzer, 1967); the cellular programs in construction of the limb. The apical being followed during morphogenesis (Fallon and Saunders, 1968); and the time- ‘This research was completed under the direc- dependent decrease in the capability of tion of Professor Zwilling in partial fulfillment of the cells to change their morphogenetic pro- requirements for the degree of Doctor of Philosophy. The years I spent with Edgar Zwilling bring to mem- grams (Crosby, 1967; Cairns and Saunders, ory his kindness, integrity, and inspiration, and I 1954; Zwilling, 1955; Zwilling, 1968; dedicate my contributions in this work to his mem- Searls and Janners, 1969, 1971; Finch and ory. Supported by Grants Tl-HD-22 and HD-03465 Zwilling, 1971; Singer, 1972). from the National Institute of Child Health and Hu- The technique of independently manipu- man Development. ‘Present address: Department of Biology, 16. lating ectoderm and mesoderm in the 839, Massachusetts Institute of Technology, Cam- study of limb morphogenesis has allowed bridge, Massachusetts 02139. an elucidation of the developmental capa- 113

bilities of combinations of nonlimb meso- form individual tissues (Crosby, 1967; derm with limb ectoderm as well as reveal- Zwilling, 1968; Singer, 1972). ing the patterns inherent in the limb The present work is a quantitative analy- mesoderm (Zwilling, 1955, 1956a, b; sis of the regulative capacities of meso- Saunders et al., 1959). More recently blast cells using the manipulative grafting Zwilling (1964) introduced a technique of procedures reviewed above. Young meso- studying the organizational and histo- blast cells were “diluted” with cells no genetic capacities of limb mesoblast cells longer capable of morphogenetic modula- which involved removal of the mesoblast, tion (no longer having “limbness” traits). dissociating the cells, then pelleting and The results show that such quantitative reimplanting them in an epidermal jacket, alterations elicit striking changes in and finally transplanting the stuffed ecto- morphogenetic expressions. derm to an ectopic flank site. Crosby (1967) used this method to measure the MATERIALS AND METHODS response to the apical ridge of mesoblast The method whereby cells are dissoci- cells in early stages before limb bud out- ated and stuffed into an ectodermal jacket growth (stage 13 of Hamburger and Hamil- is extensively described by Finch and ton, 1951). Her results affirmed the Zwilling (1971). About 50-60 leg buds, earlier conclusion of Zwilling (1955) and stage 19, were excised from White Leg- Saunders et al. (1957) that only limb meso- horn embryos and the mesoblast was derm is capable of growth and prolifera- denuded of its ectodermal covering with a tion when associated with the ridge and is 30-min treatment in a mixture of 2% tryp- the only mesoderm capable of supporting sin (Nutritional Biochemicals Corp.) (or inducing regeneration of) the ridge by and 1% pancreatin (Nutritional Biochemi- producing a maintenance factor (Amprino cals Corp.). All manipulations were done and Camosso, 1958; Saunders et al., 1957; in calcium- and magnesium-free Tyrode’s Searls and Zwilling, 1964). Moreover, solution (CMF). The freed mesoblast was Finch and Zwilling (1971) showed that then dissociated into a cell suspension by these properties are not expressed by further treatment with 1 ml of a 1% crys- mesoderm cells later than stage 24 or by talline trypsin solution (Worthington distal mesoblast tissue later than stage 26. Biochemical Corp.) for 20 min (38” C) . The The transient limb properties (“limbness”), mixture was poured into a full test tube which are typified by the response of (8 ml) of horse serum-Tyrode’s solution limb cells to the ridge resulting in growth (1: 1) and the cells were allowed to settle and morphogenesis of a limb, has been to the bottom. All but the bottom milliliter termed by Zwilling (1968) the “morpho- of horse serum solution was removed, and genetic phase” of limb development. The the cellular conglomerate was further dis- time when the limb cells acquire the po- rupted with a Vortex mixer. In the dissoci- tential for forming a limb precedes the ation of mesoblast cells from later stages, time when they become committed to a the cells from the dense chondrogenic cytodifferentiative path, for example as regions which failed to suspend even measured by the first appearance of en- after increased trypsinization (up to an zyme activity for synthesis of sulfated hour) were removed by filtration through mucopolysaccharides and other end prod- a 20-mesh Nitex cloth (Tobler, Ernst ucts (Searls, 1965; Medoff, 1967). Finally, and Traber; Elmsford, New York). the technique of dissociation and reim- For the preparation of chondrocytes plantation was also used to analyze the from stage 30 limbs (7 days), the legs capacity of mesoblast cells to sort out and were minced and then incubated with SINGER Analysis of Limb Morphogenesis 115 crystalline trypsin for 45 min. The tissue became firm. The pellet was removed clusters were dropped through a column of with a sterile spatula and placed in a dish CMF and further disrupted in the remain- containing horse serum solution. The pellet ing 1 ml after removing the supernatant. was subdivided into fragments each of After disruption, 4 ml of CMF was added which was stuffed into an ectodermal sac and mixed and the remaining chunks were which was just removed from stage 20 leg allowed to settle. The supernatant was buds. The stuffed ectoblasts were then discarded; the tissue pieces were then incubated in horse serum solution for 45 incubated in a depression dish in CMF min at 38°C. Eggs windowed at 4 days for 30 min, and the disruption procedures of incubation were used for grafting. An were repeated. The remaining tissue was excision of epidermis about one somite mainly cartilage rods. The rods were in- square was made over the somites of a cubated in the crystalline trypsin solution stage 20 animal just dorsal to the wing for 1 hr at 38”C, and then dropped through bud. The stuffed ectoblast was grafted to a test tube of horse serum solution; the the excised area with the free mesodermal supernatant was removed, and the re- area in contact with the graft site. The mainder was disrupted by Vortex agita- window was then sealed with Scotch tion. The cellular suspension was then brand tape and the egg was returned to passed through Nitex cloth to remove the incubator. After 2 days the grafts were undissociated rods. In this way it was observed and hosts which rejected grafts possible to obtain 6 million chondrocytes were discarded. Grafts were observed at from 12 limb buds. Viability was greater the seventh day and then fixed for histo- than 85% as measured by eosin exclusion. logical analysis. For the preparation of flank mesen- thyme cells, stage 19 embryos were re- RESULTS moved and resected in the region between Mixtures of Stage I9 Mesoblast Cells the limbs. The resected fragments were and Stage 30 cartilage Cells immersed in trypsin and pancreatin solution for 20 min, the ectoderm teased away All the reconstituted limbs showed some and the flank mesoderm freed from the outgrowth by 7 days after grafting. The somites and the splanchnic mesoderm. outgrowths ranged from tiny projections The flank mesoderm was immersed in of a few millimeters to full-sized digits CMF for 15 min and then in a 1% solution with three claws. There was also examples of crystalline trypsin for 30 min. The of unorganized massive growths resembling conglomerate was then immersed in a test enlarged tumorous formations. The char- tube of horse serum solution and centri- acteristics of each graft were tabulated. fuged gently (750 rpm, 5 min). The super- Then it was excised with the surrounding natant was removed, the cellular pre- host tissue, fixed in formalin, stained in cipitate vigorously dissociated on a Vortex methylene blue, dehydrated and cleared mixer for 20 set, and the cells passed in methyl salicylate to reveal the cartilagi- through a Nitex cloth. Twenty-four em- nous skeleton (Finch and Zwilling, 1971). bryos yielded about four million cells. The morphogenesis which had occurred After counting in a hemacytometer, the could by assayed by the presence of isolated mesenchymal and cartilage cells jointed rods and other features of skeletal were mixed in selected proportions. Then structure. Cartilage nodules alone and the mixture was pelleted at 750 rpm in a rods without joints indicated that histo- conical centrifuge tube and incubated for genesis but not morphogenesis had taken 1 hr at 38”, during which time the pellet place. 116 DEVELOPMENTAL BIOLOGY VOLUME 28, 1972

The ratio of stage 19 mesenchyme cells proximal mesoblast has only recently lost to cartilage cells in a mixture determined its “limbness” properties, but has not yet the extent of limb outgrowth. The series begun to exhibit the cartilage tissue of photographs in Fig. 1 depict the direct found in the stage 25 proximal mesoblast. dependence of outgrowth on the ratio of Since a population of cells in this stage of stage 19 cells when mixed with chondro- the limb are not homogeneous with respect cytes of stage 30. Mixtures which gave to cytodifferentiation as are chondrocytes “no outgrowth” were apparently resorbed from stage 30, or morphogenetic expres- by 7 days, although 2 days after grafting sion, as are stage 19 cells, they must con- they appeared well vascularized and tain either some components of both cell healthy. Small growths (less than 1 mm in types, or an intermediate cell type with length) or an unorganized tumorlike mass the properties of neither. If this popula- with no identifiable limb features were tion contained a subpopulation of stage also not classified as outgrowths. 19 cells, then morphogenesis should re- Stage 19 cells, called by Zwilling (1968) sult from the addition of less stage 19 “competent” or “uncommitted” cells cells than was needed to initiate develop- (meaning that they have not become sta- ment with the chondrocytes (i.e., 5%). bilized in their cytodifferentiated prop- This is because the stage 19 cells would erties and retain “limbness” properties) be added to a substantial population of were mixed with chondrocytes in per- competent cells; a population which pre- centage variations of 0 to 30 (see Fig. 1). sumably did not exist in the chondrocyte Table 1 indicates the percentage of grafts population of stage 30. giving outgrowths and morphogenesis To test this, dissociated cells from the from each one of these mixtures. The stage 19 mesoblast were mixed with cells salient aspect of the results is that the from the proximal region of various limb addition of only a few percent of the stages. The proximal half or third of limbs younger cells was sufficient to effect in stages 23, 24, and 25 were removed and morphogenesis from a population of dissociated after removal of the ectoderm. chondrocytes which has long lost its limb The stage 19 cells were then mixed in vari- properties. Thus these older cells, having ous proportions as already described. In elaborated cytodifferentiated properties, this way, the percent threshold of stage 19 would never give an outgrowth when sub- cells required for morphogenesis of cell jected to the grafting procedure mentioned mixtures with stages 23, 24, and 25 meso- above. However, morphogenesis occurred blasts could be determined. once the competent cells surpassed a Tables 2-4 illustrate the results of these threshold ratio of somewhere between 3 experiments. The tabulation shows that and 5% of the total mixture. proximal cells from each of the stages 23, 24, and 25 yield outgrowths and morpho- Mixtures of Stage 19 Cells and Stages genesis when mixed with stage 19 cells. 23, 24, and 25 Proximal Cells However, the amount of stage 19 cells At stage 24, chondrification is only just needed to give limblike responses differs beginning in the central areas (Searls, for each of the three stages. Unexpectedly, 1965) and the peripheral areas still show the proximal cells of stage 24 require a high mitotic activity (Searls and Janners, significantly more younger cells than 1971); yet morphogenesis is no longer either stage 25 or stage 23 in order to give feasible upon disruption and grafting of the same number of outgrowths. Stage 25 the proximal mesoblast although the cells give a response with only 3% stage 19 median mesoblast still has morphogenetic cells added, almost the same results as properties (Finch and Zwilling, 1971). The done earlier with chondrocytes. Stage 23 SINGER Analysis of Limb Morphogenesis

FIG. 1. Photographs of representative outgrowths fixed 7 days after grafting. (A-E) indicate dose-e ffect of stage 19 cells mixed with stage 30 chondrocytes. (F) reveals the tumorous outgrowth characteristic of a mixture of flank cells. All photographs enlarge material ten times. (A) 3% stage 19 cells mixed with 97% chondrocytes. (B) 5% stage 19 cells mixed with 95% chondrocytes. (C) 7% stage 19 cells mixed with 93% chondrocytes. (D) 10% stage 19 cells mixed with 90% chondrocytes. (E) 20% stage 19 cells mixed with 80% chondrocytes. (F) 50% stage 19 cells mixed with 50% flank cells. 118 DEVELOPMENTAL BIOLOG:Y VOLUME 28. 1972 proximal cells alone are capable of pro- TABLE 3 ducing an outgrowth and some morpho- RESULTANT OUWROWTHS AND MORPH~CENESIS (PRES- genesis, perhaps indicating presence of a ENCE OF JOINTED CARTILAGE RODS) FROM VARIOUS MIXTURES OF DISSOCIATED STAGE 19 MESODERM still existent stage 19 cell population. How- AND STAGE 24 PROXIMAL MESODERM CELLS= ever, when stage 19 cells are added, the Percent Percent Percent number of grafts giving limblike responses stage 19 Sample giving giving cells in size outgrowth morphogenesis increases considerably. But the stage 24 mixture proximal mixture with stage 19 shows few 0 20 0 0 cases of morphogenesis at the lower per- 10 19 38 15 centage. This implies that the stage lQ- 15 16 55 30 like population of cells is no longer present. 20 24 62 60 25 9 62 62 Mixtures of Stage 19 Limb and Flunk 30 13 85 70 Mesoderm a Procedure as in Table 1. Although the limb and flank arise from TABLE 4 the same general area of the lateral so- RFXILTANT OUTGROWTHS AND MORPHOGENESIS(PRES- matic mesoderm (see Crosby, 1967), they ENCE OF JOINTED CARTILAGE RODS) FROM VARIOUS MIXTURES OF DISSOCIATED STAGE 19 AND STAGE 25 become developmentally distinct areas PROXIMAL MESODERMAL CELL@ by stage 14. The flank and limb meso- Percent dermal cells taken from stage 19 em- stage 19 Sa”lple Percent Percent cells in size giving giving mixture outgrowth morphogenesis

TABLE 1 3 24 41 8 RESULTANT OUTGROWTHS AND MORPHOGENESIS(PRES- 5 17 59 36 ENCE OF JOINTED CARTILAGE RODS) FROM VARIOUS 7 26 70 60 MIXTURES OF DISSOCIATED STAGE 19 MESODERM 15 27 72 60 AND STAGE 30 CHONDROCYTESSTUFFED INTO ECTO- 20 5 100 80 BLASTSAND GROWN AS SOMITE GRAFTS~ o Procedure as in Table 1. Percent Percent Percent TABLE 5 stage 19 Sample giving giving cells in size outgrowth morphogenesis RESULTANT OUTGROWTHSAND MORPHOGENESIS(PRES- mixture ENCE OF JOINTED CARTILAGE RODS) FROM VARIOUS 0% 8 0 0 MIXTURES OF DISSOCIATED STAGE 19 MESODERM AND 3 35 26 26 FLANK STAGE 19 CELLSO 5 22 90 65 Percent 7 24 100 65 stage 19 limb Sample Percent Percent cells in size giving giving morphogenesis 10 22 100 67 mixture outgrowth 20 10 100 70 30 9 100 70 3 6 0 0 R Fixed after 7 days; see Materials and Methods. 5 13 0 0 7 7 0 0 TABLE 2 10 5 0 0 RESULTANT OUTGROWTH AND MORPH~CENESIS (PRES- 20 14 0 0 ENCE OF JOINTED CARTILAGE RODS) FROM VARIOUS 25 37 18 0 MIXTURES OF DISSOCIATED STAGE 19 MESODERM 35 11 31 0 AND STAGE 23 PROXIMAL MESODERM CELLS~ 50 20 65 12 75 5 80 40 Percent Percent Sample Percent stage 19 giving cells in size giving a Procedure as in Table 1. mixture outgrowth morphogenesis bryos were mixed in order to determine 0 38 24 9 10 20 63 63 the effect of nonlimb mesoderm on limb 20 18 66 45 mesoderm expression. The results from 30 17 65 65 these mixtures, done as in all previous D Procedure as in Table 1. mixtures, are reported in Table 5 and SINGER Analysis of Limb Morphogenesis 119

Fig. 1F shows a typical outgrowth from the “morphogenetic phase.” The present such a mixture. The mixture often pro- study indicates that not only are the cells duced a very large cellular mound with competent to express this morphogenetic no limb features although tiny distal out- phase even when greatly diluted with non- growths were observed and cartilage competent cells, but that this expression nodules were occasionally evident. Stained can be enhanced or suppressed by the whole mounds of 50 : 50 mixtures showed particular second component in the no limblike internal structure. Thus the mixture. Therefore, the cellular environ- flank cells inhibit the expression of limb ment surrounding the cells from the properties of the stage 19 cells. This in- “morphogenetic phase” determines the hibition further emphasizes the impor- ultimate outcome of their expression. tance of the cellular environment for ex- These quantitative and qualitative obser- pression of the developmental capacities vations are expanded in the following of the stage 19 cells, an aspect which will discourse. be elaborated further. In the case of a mixture of stage 19 cells with cells of stage 25 or beyond, the cellu- DISCUSSION lar environment results in a very tentative A system of manipulation of mesodermal balance between the two cell types such limb bud cells is reported for purposes of that morphogenesis is elicited by only a quantitative analysis of developmental 2% shift in the population of the compe- capacities of limb mesoderm. The system tent cell type (see Fig. lA, B). One might is artificial in that two cell populations imagine that still fewer stage 19 cells drawn from embryos of different develop- would be required to sustain morphogene- mental stages are mixed which have no sis in a mixture of stage 24 cells. Contrarily, juxtaposition in uiuo due to their differ- results showed that more young cells ences in developmental age. The experi- were needed in order for morphogenesis mental approach is to vary the ratio of two to result. Thus, the stage 24 cells sup- cell populations in a mixture, only one of press the morphogenetic and growth ex- which is derived from a mesoblast stage pressions of the stage 19 cells. capable of growth and subsequent morpho- The means of this suppression of mor- genesis when subjected to the dissociation phogenesis might be conjectured at this and reassociation techniques of Zwilling point. Searls (1965) and Medoff (1967) (1964). We found threshold quantitative showed that at the time of stage 24 the ratios of the two populations above which condroblasts in the limb have begun to morphogenesis usually occurred and be- differentiate by augmenting chondroitin low which it was absent. The threshold sulfate production. Also at this stage, the was most striking when the two cell popu- results of Zwilling (1968) and Finch and lations were greatly separated in stage Zwilling (1971) showed that the cells are (for example, stages 19 and 30). Only a beyond their morphogenetic phase. Stage small population of the younger cells was 24 cells by themselves, unlike those of required to establish the threshold ratio. stage 23, are unable to express morpho- Therefore, quantitative alterations in the genesis in these experimental circum- cellular interrelations elicit dramatic stances; nor are they able to express fully qualitative changes in the subsequent a cartilage-producing function, as are the development of the organ. stage 25 cells. This suggests that transient The ability of suspended limb cells to cellular events occur at this stage pro- evoke morphogenesis declines with the ducing a population of cells which is just age of the cells. The time preceding this losing morphogenetic abilities, and is loss has been termed by Zwilling (1968) just gaining cytodifferentiated traits. 120 DEVELOPMENTALBIOLOG ;Y VOLUME28, 1972

Thrown into such a hybrid pool, the expressions of stage 19 cell capacities may no longer be possible-perhaps as a result of intercellular interferences. Flank cells from the stage 19 embryo totally inhibit any expression of the stage 19 limb cells, until the limb cells constitute almost 75% of the mixture. This indicates that the matrix tissue around the stage 19 cells must be of limb derivation in order to effect morphogenesis. Thus, despite the similar histogenic fate of flank and FIG. 2. Graph summarizing the relationship be- limb cells (they both make muscle and tween the mixtures. cartilage in situ), the inhibition of morphogenetic control must be based on their with the stage 19 cells. Although the form morphogenetic incompatability. Crosby of the graphs are only hypothetical, it (1967) established the time of morpho- seems evident that each mixture has a genetic distinction of the flank cells to be characteristic curve; and all the curves approximately at stage 14. In this stage seem to plateau at about 70% morpho- they become irreversibly distinct from genesis. These curves indicate that those in the limb field. Although stage 19 morphogenesis is determined by several cells cannot express their capacities when factors in addition to the simple addition mixed with cells competent to produce of cells (except for that of stage 23 which nonlimb tissue, they can express such is almost linear). Perhaps an important potentialities when mixed with a totally occurrence necessary for the induction of noncompetent and highly differentiated morphogenesis is the coalescence into population of limb cells (chondrocytes from competent foci of the competent stage 19 stage 30). This would support Zwilling’s cells. The time of focal formations may be concept of “limbness” (1968), on a quanti- critical for outgrowth because a succession tative cellular level. It appears that of steps are involved in morphogenesis. morphogenesis requires all cells to be Some of the steps already known are stamped with a common property (i.e., reciprocal interactions of the ridge and “limbness,” or “flankness”). This phe- the mesodermal cells, and the chronologi- nomenon has also been noted by Moscona cal history of the cells. and Moscona (1965) in the suppression of In addition, morphogenesis is the ulti- feather morphogenesis by cells from other mate result of an interaction of the ecto- tissues, although there was no attempt to dermal ridge and the limb mesoderm cells control the mixtures quantitatively. The (Zwilling, 1949). Essential to this inter- flank cells do not inhibit the limb cells action may be the juxtaposition of the from growth because the entire graft limb cells to the ridge. This proximity undergoes a tremendous tumorous ex- requirement may be met by cellular move- pansion (Fig. 1F). This uncontrolled ments such as attraction or migration of growth is almost neoplastic. The mixing the stage 19 cells, and could explain the of the cell types may have released them complex functions exhibited in Fig. 2. from morphogenetic growth-control con- These histological possibilities will be straints, for morphogenesis ordinarily examined in a further study. parallels limitation in growth. These data further extend the work The graph in Fig. 2 summarizes the done by Zwilling (1964) on the use of limb relationships of the different mixtures mesoderm as an experimental system for SINGER Analysis of Limb Morphogenesis 121

morphogenesis. In the present study his series of normal stages in the development of the experimental system was modified by chick embryo. J. Morphol. 88, 49-92. MEDOFF, J. (1967). Enzymatic events during carti- combining populations of different de- lage differentiation in the chick embryo limb bud. velopmental ages to yield a method which Deuelop. Biol. 16, 118-143. serves to assay the potentiality of cells to MOSCONA, M. H., and MOSCONA, A. A. (1965). Con- control morphogenesis. It is now possible trol of differentiation in aggregates of embryonic to investigate with this technique the skin cells: suppression of feather morphogenesis by cells from other tissues. Develop. Biol. 11, interactions between cells needed to 401-423. elicit morphogenesis. Histological investi- NAMEROFF, M., and HOLTZER, H. (1967). The loss of gations could reveal the extent to which phenotypic traits by differentiated cells. IV. the stage 19 cells participate in the forma- Changes in polysaccharides produced by dividing tion of the limb and the meaning of the chondrocytes. Develop. Biol. 16, 250-281. SAUNDERS, J. W., JR. (1948). The proximo-distal “threshold” effect. Cell divisions of the sequence of origin of parts of the chick wing and two cell types as well as their migrations the role of the ectoderm. J. Exp. Zool. 108, 363- could be elucidated. Hopefully, an histo- 404. logical elaboration of the mechanism SAUNDERS, J. W., JR., CAIRNS, J. M., and GASSELING, whereby suppression and inhibition of cell J. T. (1957). The role of the apical ridge of ectoderm in the differentiation of the morphological expression occurs might quantify and structure and inductive specificity of limb parts expose some morphogenetic mechanisms. in the chick. J. Morphol. 101, 57-87. Some of these and other problems are SAUNDERS, J. W., JR., GASSELING, J. T., and CAIRNS, analyzed in a work now in preparation in J. M. (1959). The differentiation of prospective which the experimental system em- thigh mesoderm grafted beneath the apical ectodermal ridge of the wing bud in the chick em- ployed in the present paper is further bryo. Deuelop. Biol. 1, 281-301. exploited. SEARLS, R. L. (1965). An autoradiographic study of the uptake of S3%ulfate during the differentia- REFERENCES tion of limb cartilage. Deuelop. Biol. 11, 155-168. SEARLS, R. L., and JANNERS, M. Y. (1969). The AMPRINO, R., and CAMOSSO, M. (1958). Analisi speri- stabilization of cartilage properties in the carti- mentale dello sviluppo dell’ala nell ‘embrione di lage-forming mesenchyme of the embryonic chick poll0 Wilhelm Roux Arch. Entwicklungmech. limb. J. Exp. Zool. 170, 365-376. Organismen 150, 509-541. SEARLS, R. L., and JANNERS, M. Y. (1971). The ini- CAHN, R. D., and CAHN, M. (1966). Heritability of tiation of limb bud outgrowth in the embryonic cellular differentiation: Clonal growth and expres- chick. Develop. Biol. 24, 198-213. sion of differentiation in retinal pigment cells in SEARLS, R. L., and ZWILLING, E. (1964). Regeneration vitro. Proc. Nat. Acad. Sci. U.S.A. 55, 106-114. of the apical ectodermal ridge of the chick limb CAIRNS, J. M., and SAUNDERS, J. W., JR. (1954). The bud. Deuelop. Biol. 9, 38-55. influence of embryonic mesoderm on the regional SINGER, R. H. (1972). Cellular interactions in a specification of epidermal derivatives in the chick. morphogenetic system. Submitted to Develop. J. Enp. Zool. 127, 221-248. Biol. COON, H. G. (1966). Clonal stability and phenotypic SINGER, R. H., and ZWILLING, E. (1971). Analysis of expression of chick cartilage cells in uitro. Proc. limb morphogenesis by mixing of cellular compo- Nat. Acad. Sci. U.S.A. 55, 66-73. nents. J. Cell Biol., Suppl., Nov. 1971 (Abstract). CROSBY, G. M. (1967). Developmental capabilities of ZWILLING, E. (1949). The role of epithelial compo- the lateral somatic mesoderm of early chick em- nents in the developmental origin of the “wing- bryos. Ph.D. thesis, Brandeis University, Walt- less” syndrome of chick embryos. J. Exp. Zool. ham, Massachusetts. 111, 175-187. FALLON, J. F., and SAUNDERS, J. W., JR. (1968). In vitro analysis of the control of cell death in a zone ZWILLING, E. (1955). Ectoderm-mesoderm relation- of prospective necrosis from the chick wing bud. ship in the development of the chick embryo limb Develop. Biol. 18, 553-570. bud. J. Enp. Zool. 128, 423-441. FINCH, R. A., and ZWILLING, E. (1971). Culture sta- ZWILLING, E. (1956a). Interaction between limb bud bility of morphogenetic properties of chick limb- ectoderm and mesoderm in the chick embryo. II. bud mesoderm. J. Exp. Zool. 176, 397-408. Experimental limb duplication. J. Exp. Zool. HAMBURGER, B., and HAMILTON, H. L. (1951). A 132, 173-187. 122 DEVELOPMENTAL BIOLOGY VOLUME 28. 1972

ZWILLING, E. (1956b). Interaction between limb bud ZWILLING, E. (1964). Development of fragmented ectoderm and mesoderm in the chick embryo. IV. and dissociated limb bud mesoderm. Deuelop. Experiments with a wingless mutant. J. Exp. Zool. Biol. 9, 20-37. 132, 241-253. ZWILLING, E. (1968). Morphogenetic phases in de- ZWILLING, E. (1961). Limb morphogenesis. Aduan. velopment. Sympo. Sot. Deuelop. Biol. 27th, Morphol. 1, 301-330. pp. 184-207.