In the avian embryo ectomesenchyme cells, derived from the mesencephalic level of the cranial neural crest, migrate into the presumptive maxillary region and subsequently differentiate into the membrane bones and associated secondary cartilage of the upper jaw skeleton. The cartilage arises secondarily within the periosteum at points of articulation between membrane bones and provides an embryonic articulating surface. The stimulus for the differentiation of secondary cartilage is believed to be intermittent pressure and shear created at the developing embryonic movement. The development of one such system - the quadratojugal, has been analysed using organ and explant culture techniques and studied with particular reference to the differentiation of periosteal cells into secondary cartilage. A number of conclusions were reached.

  1. Normally only cells at discrete loci express a chondrogenic potential in vivo: the periosteal cells at these sites of future articulation become committed to chondrogenesis during stage 35, more than 24 h before cartilage is identifiable in vivo.

  2. However, cells with a ‘latent’ chondrogenic potential are widespread in membrane bone periosteum and occur over most, if not all, of the surface area of the bone. This potential is expressed in the ‘permissive’ environment created by submersion of the tissue in explant culture or in submerged organ culture.

  3. This chondrogenic potential exists long before the time at which commitment of cartilage-forming cells occurs and even presumptive maxillary ectomesenchyme at stage 29 has a limited ability to form cartilage in vitro.

It is suggested that spatial position is a principal factor controlling the differentiation of secondary cartilage. Ectomesenchyme cells with the potential to form secondary cartilage are widespread but it is only those cells whose migration from the neural crest positions them and their progeny at the site of a presumptive joint which subsequently express this potential. This epigenetic interpretation is discussed in the general context of development mechanisms underlying the spatial and temporal patterns in which neural crest-derived cells differentiate to produce bone and cartilage during the formation of the head skeleton.

The major part of the bony and cartilaginous components of the head skeleton in the vertebrate groups studied to date, are formed by ectomesenchyme cells derived from the neural crest (Horstadius, 1950; Weston, 1970; Le Lievre & Le Douarin, 1975; Le Lievre, 1978). Whereas ‘mesodermal-derived’ skeleto-genic cells may be regarded as differentiating in situ, mesenchyme cells derived from the neural crest (N.C.) undergo precise migration patterns before differentiation and it is generally thought that it is the pathway along which NC cells migrate that determines the nature of their subsequent differentiation (Le Douarin & Teillet, 1974). In the case of skeletogenic cells from the N.C., some differentiate directly into bone, i.e. membrane bone, and some into primary cartilaginous models (e.g. Meckel’s cartilage and quadrate) which may or may not undergo a degree of bone replacement. The role of migration pathways, and of the range of subsequent cell-cell and cell-matrix associations generated by different migration routes, in the differentiation of skeletogenic NC cells has been discussed and reviewed elsewhere (Morris & Thorogood, 1978).

In this paper I shall describe an analysis of one such NC-derived skeletogenic system, namely the membrane bones in the maxillary process of the chick embryo. The skeleton of the margin of the upper jaw comprises three membrane bones and I shall consider the most posterior of these - the quadratojugal (QJ). It is a slender bone comprising a shaft, and a hook at the posterior end where it articulates with a groove in the quadrate. The QJ system was selected for study, not for any major intrinsic properties but because a considerable body of information concerning its developmental history has accumulated in recent years (see Fig. 1). From experiments using grafts labelled with [3H]thymidine, it is known that NC cells begin to leave the mesencephalic level of the developing brain at approximately 36 h and migrate rostrally and ventrally (Johnston, 1966; Noden, 1975). The first of these cells, or their progeny, arrive in the presumptive maxillary process at 52 h. There is subsequently a proliferation of these ‘ectomesenchyme’ cells (Johnston, 1966) and at 7 days there is a condensation of the mesenchyme cells followed at 7 days by the appearance of a calcified matrix (Murray, 1963); note that this is bone formation directly not via a cartilaginous model. Using the self-replicating marker of the heterospecific quail-chick grafting system, it has been demonstrated that most, if not all, of the bone and associated periosteum is derived from the NC (Le Lievre & Le Douarin, 1975), On the twelfth day of incubation, two pads of cartilage appear beneath the periosteum on the posterior hook. This constitutes the embryonic articulating surface with the quadrate and, because it is cartilage deposited on the surface of the bone by a periosteum, it has been termed ‘secondary’ or ‘adventitious’ cartilage (Murray, 1963). Analysis of the fate of quail mesencephalic grafts into chick embryo hosts has revealed that these chondrogenic cells, like their osteogenic neighbours, are derived from the NC at the level of the mid-brain (Le Lievre, 1978, personal communication).

Fig. 1.

A diagrammatic summary of the developmental history of the cells which comprise the quadratojugal (QJ): (A) Stages 10 –11: Cells from the mesencephalic NC migrate rostrally and ventrally. (B) Stage 16 : NC-derived cells arrive at presumptive maxillary region. (C) Stage 30: Condensation of ectomesenchyme cells to form QJ primordium followed shortly by membrane bone formation. (D) Stage 37: Secondary cartilage formation starts on the anterior and posterior aspects of the ‘hook’ of the QJ. (Not drawn to scale).

Fig. 1.

A diagrammatic summary of the developmental history of the cells which comprise the quadratojugal (QJ): (A) Stages 10 –11: Cells from the mesencephalic NC migrate rostrally and ventrally. (B) Stage 16 : NC-derived cells arrive at presumptive maxillary region. (C) Stage 30: Condensation of ectomesenchyme cells to form QJ primordium followed shortly by membrane bone formation. (D) Stage 37: Secondary cartilage formation starts on the anterior and posterior aspects of the ‘hook’ of the QJ. (Not drawn to scale).

It is now known that for NC-derived maxillary mesenchyme to become osteogenic and subsequently form membrane bone an interaction with maxillary epithelium must take place (Tyler, 1978; Hall, 1978). In vivo the cells from the NC ‘arrive’ in the presumptive maxillary region at 2 days and bone formation starts at 7 –7 days (see Fig. 1). For isolated maxillary mesenchyme to form bone in either organ culture (Hall, 1978) or when grafted onto the chorioallantoic membrane (Tyler, 1978), there is a dependency upon an epithelial presence until 3 –4 days of development. In the absence of that component, the cultured or grafted mesenchyme gives rise only to fibrous connective tissue. It has been proposed, that the causal mechanism underlying chondrogenesis is simply the biomechanical effect of movement. The embryo starts to show a dramatic increase in non-reflexogenic, spontaneous whole body movements from about 8 days onwards (Hamburger, Balaban, Oppenheim & Wenger, 1965) and it is believed that the local effects at embryonic articulations such as these, are intermittent pressure and/or shearing forces. Murray and colleagues demonstrated that following in ovo paralysis by use of post-synaptic blocking agents such as curare or decamethonium, membrane bone differentiated normally but there was an absence of secondary cartilage associated with the articulations between membrane bones (Murray & Drachman, 1967). The causal relationship between movement and secondary chondrogenesis is better demonstrated by grafting the joint to an ectopic site where the tissues are not exposed to such forces (Murray & Smiles, 1965; Hall, 1972) or by organ culture of early QJ rudiments in which the periosteal cells are not ‘determined’ for chondrogenesis. By the simple expedient of applying gentle intermittent pressure to the joint surfaces early in the culture period, chondrogenesis can be stimulated (Hall, 1968).

The overall objective of this study and of work to be reported in subsequent communications is to achieve an understanding of the developmental factors which determine the spatial and temporal pattern in which NC-derived cells differentiate into bone and cartilage during the formation of the head skeleton. The initial objective was to analyse the relationship between embryonic movement and the subsequent differentiation of ectomesenchyme cells into cartilage. The analysis was originally planned in two steps :

  • Firstly, to investigate the temporal relationship between the commitment of cells to cartilage formation and the duration of in ovo movement (i.e. when do periosteal cells of the QJ hook become committed to chondrogenesis). To establish this whole QJs from normal and paralysed embryos of various ages were grown in organ culture and the incidence of in vitro chondrogenesis recorded: contralateral QJs were fixed at the outset of the culture period and used as controls.

  • Having established the timing of commitment, was it then possible to grow uncommitted periosteal cells in large numbers, and by subjecting the cells to various in vitro manipulations, cause a chondrogenic differentiation? If such an in vitro system of chondrogenesis could be established then it might provide an insight into the acquisition of chondrogenic ability by the in vivo counterparts of these cells. Explants of ‘uncommitted’ periosteum were grown in order to obtain the appropriate cells but unexpectedly the primary outgrowths from such explants spontaneously became chondrogenic. Consequently the second part of the investigation centred upon the acquisition of chondrogenic potential by these cells and the factors controlling the expression of that potential.

All embryos used in this study were from White Leghorn ×Light Sussex eggs obtained locally. The eggs were incubated at 37 ·0 ± 0 ·5 °C in a humidified ‘Westernette’ incubator and the embryos were staged according to the developmental table of Hamburger & Hamilton (1951). Embryos were used at stages between 6 and 18 days of incubation; the appropriate tissues were removed and rinsed briefly in sterile Dulbecco’s phosphate buffered saline (PBS) pH 7 ·3. Any fine dissections were carried out in aliquots of culture medium (see later). Paralysis of embryos was induced by single doses of either 10 mg of D-tubo-curarine chloride (Sigma Chemical Company) or 1 mg decamethonium iodine (Koch-Light Laboratories) dissolved in 0 ·5 ml sterile PBS and injected into the air space of eggs which had been incubated for 9 days (Hall, 1972). Embryos were judged to be paralysed if they (i) showed a lack of movement on removal from the egg and (ii) exhibited the ‘baggy trouser’ syndrome due to oedematous hind limbs.

Intact QJs were cultured individually using a standard ‘Trowell-type’ organ culture system. The tissue was supported at the gas/medium interface on a platform of Minimesh FDP (Expanded Metal Company, West Hartlepool, U.K.) in 30 mm plastic tissue culture dishes (Sterilin) each containing 2 ml of the a-modification of Eagle’s Minimal Essential Medium with 10% foetal calf serum and antibiotics. The 30 mm dishes were placed inside standard 9 cm glass petri dishes which were humidified with damp filter paper and then incubated for 10 days at 37 ·5 °C at 5% CO2 in air in a Grant GC4 tissue culture incubator fitted with a Gow Mac analytical CO2 monitor. In later experiments (see Results) controlled pairs of QJs were cultured simultaneously in standard and ‘submerged’ conditions. In these experiments the tissue was placed initially on a 13 mm diameter Millipore filter, pore size 0 ·45 μm, and for standard organ culture at the gas/medium interface this was placed on the Minimesh raft. For ‘submerged’ organ culture conditions in which the tissue was grown below the interface the Millipore filter was ‘sunk’ by weighting down the filter with a piece of Minimesh.

Three categories of tissue of various development ages were used for explant cultures:

  • periosteal cells from the tip of QJ hooks : this tissue is known to contain cells which subsequently give rise to the cartilage pads associated with the hook (Thorogood & Hall, 1976); removal of these periosteal cells prior to organ culture of the QJ results in a complete absence of the cartilage pads (Thorogood, unpublished observation).

  • segments of the mid-shaft and associated periosteum of the QJ; approximately 0 ·25 –0 ·4 mm in length.

  • blocks of presumptive QJ tissue taken from the maxillary process and of a size approximately equivalent to those used in (ii).

The explants were cultured in 2 ml of medium in 30 mm plastic tissue culture dishes (Sterilin), both medium and incubation conditions being identical to those described earlier for organ culture. Outgrowths were viewed in the living state by inverted phase contrast microscopy. For cytological purposes some explants were made on glass coverslips in order that cultures could be subsequently fixed and stained for morphological analysis (see later): no basic difference in the nature of outgrowth from explants on glass or plastic substrates was observed.

Some tissue from organ culture was examined in whole mount preparations by fixation in formol ethanol, staining in methylene blue and clearing in methyl salicylate according to the Lundvall Technique (Hamburger, 1960). Other material from organ culture was fixed in situ on the Millipore filter in 80 % ethanol for histology. After decalcification in acid alcohol and paraffin wax embedding, 7μm serial sections were cut and stained with Hansens haematoxylin, alcian blue and chlorantine fast red (Lison, 1954). Explant cultures on glass coverslips were gently rinsed in two changes of PBS, fixed in absolute methanol and stained with the May-Grumwald and Giemsa procedure (Paul, 1965). A number of such explant cultures was subjected to enzymic digestion tests to check the specificity of the metachromasia produced by the Giemsa stain. Methanol-fixed material was incubated in a moist chamber at 37 °C for 2 h with chondroitinase ABC (Seikagaku Kogyo Co. Ltd., Tokyo, Japan) according to the method described by Katoh & Takayama (1977) and then stained with May-Grumwald and Giemsa solutions : control material was incubated in PBS before staining. Immunofluorescent techniques were used to investigate the distribution of two principal collagen types - type I (skin, bone and tendon) and type II (cartilage), by courtesy of Dr Klaus von der Mark, Max Planck-Institut für Biochemie, Munchen, West Germany. The antibodies used had been raised in guinea-pigs against chick type-I and type-II collagens and purified by immunoadsorption as described previously (von der Mark, von der Mark & Gay, 1976). Explant cultures grown on ‘Sterilin’ 30 mm plastic tissue culture dishes were washed twice in PBS fixed for 5 min in the vapour from an absolute ethanol : glacial acetic 3:1 solution, then immersed into the fixative solution for a further 20 min after which they were drained and air-dried. The procedure for application of the guinea-pig collagen type-I or type-II antibodies and counterstaining with fluorescein-conjugated anti-guinea-pig y globulins has been described elsewhere (von der Mark & von der Mark, 1977).

(a) Organ culture experiments

To establish the temporal relationship between the commitment of cells to a chondrogenic fate and in ovo movement, intact QJs of various developmental ages were grown in organ culture : from each embryo the contralateral QJ was fixed at the outset of the experiment as a control. Cartilage was never observed in any control until stage 37 –the 12th day of development (Table 1). In contrast the incidence of cartilage formation during a 10-day in vitro period rose with tissue taken from progressively older embryos, increasing from 8 % in cultures of stage-34 QJs to 100 % in stage-36 tissue (Table 1). Chondrogenesis in cultures of stage-37 tissue was discounted because cartilage formation has apparently already commenced in vivo by this time, judging by the incidence of cartilage in some of the controls at this stage. In the Lundvall whole mount preparations of normal QJs at stage 39, two discrete pads of cartilage can be identified within the hook (Fig. 2); the narrow space between the two pads is occupied by the spicule of bone projecting into the hook from the shaft. In those cultured QJs in which cartilage was found in the hook, the two pads were generally less well defined and the final form ranged from two chondrogenic loci apparently fused together (Fig. 3) to an ill-defined mass of cartilage at the tip of the hook. In sections of equivalent QJs such areas possessed a characteristic cartilage phenotype comprising an extensive alcian-blue-positive matrix in which the cells were located within distinct lacunae (Fig. 4). In order that the effect of in ovo movement on subsequent chondrogenesis might be studied, QJs were cultured from embryos paralyzed by a single in ovo injection of curare. At a developmental stage at which high incidences of in vitro chondrogenesis had been observed in normal QJs, tissue from paralysed embryos failed to exhibit cartilage formation in vitro (Table 2). In two cultures of tissue taken from stage-36 paralysed embryos limited cartilage formation was observed: it had been noted at the outset of the experiment that these two embryos alone, although exhibiting the oedematous ‘baggy trouser’ syndrome, had been incompletely paralysed and had shown some slight limb movement on removal from the egg.

Table 1.

The incidence of cartilage formation in QJs grown in organ culture

The incidence of cartilage formation in QJs grown in organ culture
The incidence of cartilage formation in QJs grown in organ culture
Table 2.

The incidence of cartilage formation in organ culture of QJs taken from embryos paralysed in ovo at 9 days

The incidence of cartilage formation in organ culture of QJs taken from embryos paralysed in ovo at 9 days
The incidence of cartilage formation in organ culture of QJs taken from embryos paralysed in ovo at 9 days
FIGURES 2–5

Fig. 2. A normal stage-29 quadratojugal (QJ). Cartilage is limited to two discrete pads within the hook (arrows). Lundvall preparation. Bar represents 1 mm.

Fig. 3. A normal QJ fixed at stage-36 (A) and its contralateral partner grown in organ culture for 10 days (B). Note the absence of cartilage in the control and the partially-fused pads of cartilage in the cultured specimen. Lundvall preparation. Bar represents 1 mm.

Fig. 4. L.S. through the hook of a stage-36 QJ grown in organ culture for .10 days. A thin spicule of bone (Bo) is surrounded on both sides by the chondrocytes of the two cartilage pads (C). Alcian blue: chlorantine fast red. Bar represents 50 μm.

Fig. 5. Fibroblastic cells at the growing edge of the outgrowth from a 7-day explant of stage-36 QJ hook periosteum. Inverted phase contrast microscopy. Bar represents 100 μm.

FIGURES 2–5

Fig. 2. A normal stage-29 quadratojugal (QJ). Cartilage is limited to two discrete pads within the hook (arrows). Lundvall preparation. Bar represents 1 mm.

Fig. 3. A normal QJ fixed at stage-36 (A) and its contralateral partner grown in organ culture for 10 days (B). Note the absence of cartilage in the control and the partially-fused pads of cartilage in the cultured specimen. Lundvall preparation. Bar represents 1 mm.

Fig. 4. L.S. through the hook of a stage-36 QJ grown in organ culture for .10 days. A thin spicule of bone (Bo) is surrounded on both sides by the chondrocytes of the two cartilage pads (C). Alcian blue: chlorantine fast red. Bar represents 50 μm.

Fig. 5. Fibroblastic cells at the growing edge of the outgrowth from a 7-day explant of stage-36 QJ hook periosteum. Inverted phase contrast microscopy. Bar represents 100 μm.

(b) Explant culture experiments

The results from the previous experiments were interpreted as demonstrating that cells in the periosteum overlying the QJ hook were committed to chondrogenesis by stage 36 and that in ovo movement prior to culturing the tissue was involved (see Discussion). To provide the large numbers of uncommitted periosteal cells needed to examine this relationship further using in vitro techniques (see Introduction), explant culture technique was employed. Periosteal tissue was dissected away from the QJ hook and used as a primary explant. However, it was found that periosteal tissue taken from embryos of an age at which cells were not committed to forming cartilage, as judged by the previous organ culture results, produced outgrowths which contained an abundance of chondrogenic cells.

Irrespective of the developmental stage used as a source of explant tissue, the incidence of cartilage formation in the outgrowths was always 100 % (Table 3). Also the rate of outgrowth and the timing of the appearance of various cell phenotypes within the outgrowth were independent of developmental stage. The explant usually became attached to the substrate within the first 24 h of culture and shortly afterwards cells with a fibroblast-like phenotype grew out in an apparently random fashion.

Table 3.

The incidence of chondrogenesis in explant culture of periosteum isolated from the QJ hook

The incidence of chondrogenesis in explant culture of periosteum isolated from the QJ hook
The incidence of chondrogenesis in explant culture of periosteum isolated from the QJ hook

As more cells appeared the outgrowth became consolidated and grew in a centrifugal fashion, always with fibroblast cells at the growing circumference (Fig. 5). Behind this growing edge but within the fibroblastic outgrowth a second cell phenotype arose on the 4th day. These cells had a flattened epithelioid morphology and each cell was surrounded by a refractile extracellular space; this phenotype initially arose in small monolayer loci around the original explant and these loci subsequently expanded and coalesced (Fig. 6). This cell phenotype was interpreted as chondrogenic on the basis of its characteristic in vitro chondrocyte morphology (e.g. Solursh, Meier & Vaerwyck, 1973) and because of the composition of the extracellular matrix produced by the cells. Initially these chondrogenic cells existed in monolayers but these subsequently grew into multilayers composed of more rounded cells. After 10 –12 days secondary loci of chondrogenesis arose away from the primary outgrowth: these were assumed to be established by floaters from the primary outgrowth which settle, attach and divide. Such secondary loci exhibited the same morphological sequence of epithelioid monolayered cells, followed by cells in multilayers which subsequently grew to produce small nodules of cartilage after several weeks of culture.

FIGURES 6–9.

Fig. 6. A monolayer of epithelioid chondrogenic cells and refractile extracellular matrix within the outgrowth from a 12-day explant of a stage-35 (early) QJ hook periosteum. Inverted phase contrast microscopy. Bar represents 100μm.

Fig. 7. Fixed and stained preparation of outgrowth from 14-day explant of stage-34 QJ hook periosteum. Note the metachromatic matrix organized into distinct lacunae (arrows), each occupied by a cartilage cell. May-Grunwald and Giemsa. Bar represents 40 μm.

Fig. 8. Immunofluorescent staining of the outgrowth from a 14-day explant of stage-35 (early) QJ hook periosteum; fixed in absolute ethanol: glacial acetic, air-dried and incubated with antibody to type-II collagen. The extracellular material around each lacunae stains positively, indicating the presence of the cartilagespecific collagen type in the matrix. Bar represents 100/mi.

Fig. 9. Myotubes (arrows) overlying a loose monolayer of fibroblastic cells in the outgrowth of an 8-day explant of stage-29 presumptive maxillary mesenchyme. A. number of dark cells with fine elongate processes (arrowheads) abutting onto the myotubes are also present and these were interpreted as nerve cells. Inverted phase contrast microscopy. Bar represents 100 μm.

FIGURES 6–9.

Fig. 6. A monolayer of epithelioid chondrogenic cells and refractile extracellular matrix within the outgrowth from a 12-day explant of a stage-35 (early) QJ hook periosteum. Inverted phase contrast microscopy. Bar represents 100μm.

Fig. 7. Fixed and stained preparation of outgrowth from 14-day explant of stage-34 QJ hook periosteum. Note the metachromatic matrix organized into distinct lacunae (arrows), each occupied by a cartilage cell. May-Grunwald and Giemsa. Bar represents 40 μm.

Fig. 8. Immunofluorescent staining of the outgrowth from a 14-day explant of stage-35 (early) QJ hook periosteum; fixed in absolute ethanol: glacial acetic, air-dried and incubated with antibody to type-II collagen. The extracellular material around each lacunae stains positively, indicating the presence of the cartilagespecific collagen type in the matrix. Bar represents 100/mi.

Fig. 9. Myotubes (arrows) overlying a loose monolayer of fibroblastic cells in the outgrowth of an 8-day explant of stage-29 presumptive maxillary mesenchyme. A. number of dark cells with fine elongate processes (arrowheads) abutting onto the myotubes are also present and these were interpreted as nerve cells. Inverted phase contrast microscopy. Bar represents 100 μm.

The chondrogenic nature of these cells was confirmed by a number of other criteria. Firstly, in the outgrowths of explant cultures grown on glass coverslips and stained by the May-Grumwald and Giemsa technique, there were extensive areas of metachromatic extracellular matrix (Fig. 7). The matrix had a lacunar organization and resembled the characteristic morphology of sectioned cartilage. The metachromasia was abolished by preincubation in chondroitinase ABC prior to staining whereas control preparations preincubated in PBS retain the metachromatic staining property. Secondly, the ultrastructural appearance of these cells was characteristic of chondroblasts and chondrocytes – a scalloped cell profile and a cytoplasm packed with rough E.R. and surrounded by a matrix containing granular and fibrillar components (unpublished observations). Lastly, the immunofluorescent analysis of collagen types revealed that the extracellular matrix was strongly positive for type-II collagen - the collagen species characteristic of cartilage, but contained very little type I (Fig. 8). Type I was found sparsely distributed among the fibroblasts at growing edge but more strongly located in a border around the nodules of cartilage formed in older cultures.

It had been previously noticed that in organ culture of intact QJs small diffuse patches of cartilage occasionally occurred at loci away from QJ hook and therefore it was felt necessary to test the chondrogenic potential of periosteum from other regions in explant culture. Segments of QJ shaft and associated periosteum were removed from embryos ranging from stage 34 to stage 43 and grown as explants. Like the previous cultures of QJ hook periosteum, the present set of cultures exhibited virtually a 100% incidence of chondrogenesis which again was independent of the developmental stage used as a source of tissue (Table 4). The manner of outgrowth, range of cell phenotypes and differentiation of cartilage were all identical to those described for the previous explant cultures.

Table 4.

The incidence of chondrogenesis in explant culture of segments of periosteal tissue taken from the QJ shaft

The incidence of chondrogenesis in explant culture of segments of periosteal tissue taken from the QJ shaft
The incidence of chondrogenesis in explant culture of segments of periosteal tissue taken from the QJ shaft

As all tissues grown as explants had exhibited chondrogenesis in vitro, including the earliest tissue – taken from stage-34 embryos, it was necessary to establish how early this chondrogenic potential was acquired by the cells concerned. Tissue for explant culture was taken from embryos of progressively earlier developmental stages (see Fig. 1):

  • QJ rudiments at stage 33: that is, soon after membrane bone formation commences.

  • The QJ condensation at stage 31 when it consists of a mass of tightly packed cells and uncalcified matrix.

  • Samples of maxillary mesenchyme at stage 29, that is approximately 12 h before condensation, were selected as ‘presumptive’ QJ tissue.

Cultures of stage-33 QJ rudiments grew in a manner identical to that described previously for periosteum explant cultures and a 100% incidence of chondrogenesis was recorded (Table 5). In contrast, in cultures of tissue taken from the developmentally earliest source, that is stage-29 maxillary mesenchyme, the incidence of cartilage formation was far lower –36 ·4% (Table 5) and the chondrogenic loci were much smaller, grew slowly and generally failed to coalesce, apparently being overgrown by vigorous proliferation of fibroblastic cells. Myotubes were always observed in maxillary mesenchyme cultures by the 6th day of culture and were regularly found lying over a layer of fibroblastic cells (Fig. 9). Irregularly shaped cells with slender, elongated processes were frequently found in association with the myotubes and at the growing edge of the outgrowth these were tentatively interpreted as nerve cells. By approximately 12 days of culture both the myogenic and neurogenic phenotypes could no longer be identified and the outgrowths were composed largely of rapidly proliferating fibroblastic cells and the occasional locus of chondrogenic cells. Explants of stage-31 QJ condensations grew in a similar fashion to the maxillary mesenchyme cultures although the incidence of chondrogenesis was higher at 50% and the incidence of myogenesis lower at 62 ·5% (Table 5). It should be noted that myogenesis was confined to these cultures of younger tissues and that myotubes were never seen in outgrowths of explant cultures of tissues taken from embryos older than stage 31. Their occurrence was interpreted as the result of inadvertent inclusion of cells which in vivo would have contributed to facial muscle but could not be excluded from the presumptive maxillary ectomesenchyme, or from the ill-defined stage-31 QJ condensation, during explantation.

Table 5.

The incidence of chondrogenesis and myogenesis in explant cultures of tissue taken from the presumptive maxillary process and from the early QJ rudiment

The incidence of chondrogenesis and myogenesis in explant cultures of tissue taken from the presumptive maxillary process and from the early QJ rudiment
The incidence of chondrogenesis and myogenesis in explant cultures of tissue taken from the presumptive maxillary process and from the early QJ rudiment

Tissue removed from in ovo paralysed embryos had been previously found to lack the ability to form cartilage in organ culture, even at a stage when QJs from normal embryos are apparently committed to chondrogenesis (see earlier – Tables 1 and 2). The elimination of in ovo movement by paralysis had been thought to block the event leading to chondrogenic commitment of periosteal cells (see later – Discussion). Explant culture had been found to be ‘permissive’ to chondrogenesis by this same population of cells irrespective of developmental stages and in order to find out if the permissiveness of explant conditions would overcome the in ovo block, explants of QJ periosteum from paralysed embryos were set up. Embryos were paralysed by a single in ovo injection of curare or decamethonium at stage 35, and the QJ hook periosteum removed at stage 36 and cultured (Table 6). All explants exhibited a pattern of outgrowth and chondrogenic differentiation identical to that shown by previous explant cultures of stage-33 and older tissues.

Table 6.

The incidence of chondrogenesis in explant culture of QJ hook periosteum taken from embryos paralysed in ovo at 9 days

The incidence of chondrogenesis in explant culture of QJ hook periosteum taken from embryos paralysed in ovo at 9 days
The incidence of chondrogenesis in explant culture of QJ hook periosteum taken from embryos paralysed in ovo at 9 days

(c) ‘Submerged’ organ culture experiments

Previous results indicated that using organ culture techniques the commitment of periosteal cells to a chondrogenic fate was stage-dependent, and that the dependency was apparently based upon the existence and duration of in ovo movement prior to culture. In contrast any tissue grown in explant culture failed to show a stage-dependent commitment to a chondrogenic fate and explant culture conditions appeared to be ‘permissive’ in that cells of virtually any age, from any region of the periosteum, always formed cartilage. This dramatic difference in the in vitro expression of chondrogenic potential was investigated using a third culture technique. A principal difference between tissue grown in organ culture and tissue cultured as an explant is that the former is maintained at the gas phase/medium interface whereas the latter becomes attached to the floor of the culture vessel and is submerged beneath the medium. In order to test whether or not the factor of submersion was important intact QJs were grown in organ culture conditions but submerged beneath the gas phase/medium interface (see Materials and Methods). All QJs were taken from stage-36 embryos previously paralysed by in ovo injection of curare or decamethonium at stage 35 and consequently the tissue lacked cells committed to chondrogenesis (see earlier – Table 2). From each embryo the QJ from one side was grown in conditions of submerged organ culture and the contralateral QJ grown in standard organ culture as a control. In all other respects conditions were identical to the organ culture experiments described previously.

On examination of the control tissue a 15 ·8% incidence of in vitro chondrogenesis was observed and in each of these positive cases the chondrogenesis was limited to single small nodules of cartilage (Table 7). The controls in this experiment are identical to the experimental cultures recorded in Table 2 and the results are essentially the same, that is a low incidence of limited chondrogenesis. However, the submerged QJs whose control partners produced little if any cartilage in standard organ culture, were found to regularly contain abundant cartilage (Table 7 and Fig. 10). The positive staining for cartilage was generally heavy and much more widespread than the cartilage found in previous experiments: compare Fig. 10 with Fig. 3.

Table 7.

The incidence of cartilage formation in ‘submerged’ organ culture of QJs taken from stage-36 embryos paralysed in ovo at 9 days

The incidence of cartilage formation in ‘submerged’ organ culture of QJs taken from stage-36 embryos paralysed in ovo at 9 days
The incidence of cartilage formation in ‘submerged’ organ culture of QJs taken from stage-36 embryos paralysed in ovo at 9 days
FIGURES 10–12

Fig. 10. Intact QJs from a single stage 36 embryo paralysed at stage 35 with decamethonium iodide and cultured for 10 days ; fixed and processed by the Lundvall technique as in Figs. 2 and 3. The lower specimen was grown in ‘submerged; organ culture and the upper specimen was the contralateral control grown in standard organ culture. The submerged specimen contains extensive cartilage associated with the hook (arrow) and along the greater part of the shaft; cartilage is absent in the control. Bar represents. 2 mm.

Fig. 11. L.S. through the shaft of a QJ from a stage-36 embryo paralysed at stage 35 and grown in submerged organ culture for 10 days. The darker staining bone (Bo) is surrounded by, and invested with, cartilage which has a lighter staining matrix (c). The lacunar organization of the cartilage is more clearly seen in the upper part of the figure (arrows). Alcian blue and chlorantine fast red. Bar represents 100 μm.

Fig. 12. L. S. through the shaft of the contralateral partner of the specimen in Fig. 11, grown in standard organ culture. The bone (Bo) is surrounded by fibroblastic cells and no cartilage is present. Filter (f). Alcian blue and chlorantine fast red. Bar represents 100 μm.

FIGURES 10–12

Fig. 10. Intact QJs from a single stage 36 embryo paralysed at stage 35 with decamethonium iodide and cultured for 10 days ; fixed and processed by the Lundvall technique as in Figs. 2 and 3. The lower specimen was grown in ‘submerged; organ culture and the upper specimen was the contralateral control grown in standard organ culture. The submerged specimen contains extensive cartilage associated with the hook (arrow) and along the greater part of the shaft; cartilage is absent in the control. Bar represents. 2 mm.

Fig. 11. L.S. through the shaft of a QJ from a stage-36 embryo paralysed at stage 35 and grown in submerged organ culture for 10 days. The darker staining bone (Bo) is surrounded by, and invested with, cartilage which has a lighter staining matrix (c). The lacunar organization of the cartilage is more clearly seen in the upper part of the figure (arrows). Alcian blue and chlorantine fast red. Bar represents 100 μm.

Fig. 12. L. S. through the shaft of the contralateral partner of the specimen in Fig. 11, grown in standard organ culture. The bone (Bo) is surrounded by fibroblastic cells and no cartilage is present. Filter (f). Alcian blue and chlorantine fast red. Bar represents 100 μm.

Histological examination of the tissues from the experiment revealed that control QJs contained healthy membrane bone surrounded by osteoblastic and fibroblastic cells whereas submerged tissue contained, in addition to these components, extensive cartilage which not only covered and invested the periosteal surfaces but also filled the spaces within the membrane bone (Figs. 11 and 12). The alcian blue-positive matrix was abundant and in some instances could also be found within osteocytic lacunae and surrounding the osteocytes.

(a) Commitment and chondrogenesis

It is evident from the organ culture experiments described in section (a) of the Results that commitment of periosteal cells to chondrogenesis is normally a time or stage-dependent phenomenon. The incidence of in vitro chondrogenesis rose progressively with the stage of embryo used as a tissue source until stage 36 - the 11th day development, when all cultured membrane bones subsequently exhibited cartilage formation. Evidently cells in the QJ periosteum have become committed to chondrogenesis at least 24 h before secondary cartilage is histologically identifiable, as cartilage in vivo is first seen at stage 37. In the light of extensive work demonstrating a causal relationship between the biochemical effects of movement and secondary chondrogenesis (see Introduction) it is possible that the pattern of commitment revealed in this experiment reflects a quantitative dependence upon the amount of in ovo movement and intermittent pressure to which the tissue is exposed prior to culture. If this is the case then by stage 36 a threshold has been passed and a state of commitment attained by the cells. When analysing the non-reflexogenic spontaneous activity of chick embryos Hamburger et al. (1965) found that although motility is exhibited as early as 3 days of development the proportion of overall time occupied by such activity is low and only starts to increase sharply at stage 32 when it rises from 25 % to 70 % at stage 37 and subsequently peaks at 80% by stage 39. Furthermore, the actual phases of activity start to lengthen dramatically from stage 34 and by this time spontaneous motility is exhibited not only by the neck, trunk and limbs but also by the mouth, tongue and eyelids. Thus if periosteal cells at articulations between membrane bones are exposed to intermittent pressure during embryonic movement it is appropriate that the incidence and duration of embryonic motility increases shortly before the time at which periosteal cells become committed to forming secondary cartilage. That movement is involved in establishing commitment is clearly demonstrated by the culture of QJs from paralysed embryos (Table 2) where elimination of in ovo movement from 9 days of incubation onwards, results in the elimination of an ability to form cartilage in vitro, presumably as a consequence of blocking the events leading to commitment. Precisely how intermittent pressure and/or shearing forces affect the periosteum at the cellular and subcellular levels is not known (see discussion in Hall, 1978).

In connexion with this causal relationship between movement and secondary chondrogenesis it should be noted that the present system is very different from the chondrogenesis occurring in ‘mesodermal’ mesenchyme. If one eliminates movement from a system of developing diarthrodial joints (such as in the limb) by paralysis or by ectopic grafting, cartilage formation is actually enhanced to the extent that secondary fusion of the articulating surfaces occurs (e.g. Drach-mann & Sokoloff, 1966). In the present system of NC-derived cells the absence of movement results in a lack of chondrogenesis – a point which emphasizes the difference in developmental histories and underlying developmental mechanisms of tissues which are phenotypically alike, but which are, in an operational sense, ‘non-equivalent’ (Lewis & Wolpert, 1976).

(b) Chondrogenesis in explant culture

The original objective of using the explant culture technique described in Section (b) of the Results was to grow large numbers of uncommitted and non-chondrogenic periosteal cells. It was intended that such cells should be subjected to various in vitro manipulations in an attempt to cause a chondrogenic differentiation thereby establishing an in vitro system which might provide some insight into the acquisition of chondrogenic ability by these cells in vivo. However, this had to be revised as a result of the unpredicted finding that virtually all periosteal explant cultures spontaneously formed cartilage in the primary outgrowths. Irrespective of age of embryo providing the source of tissue (Tables 3 and 5) or the precise site from which the periosteal tissue was taken (Tables 4 and 5) all cultures showed a similar pattern of differentiation. The temporal sequence of cell phenotypes appearing in the outgrowths and the type of extracellular matrix produced was virtually identical in all types of explant (with the exception of the incidence of myogenesis and overgrowth by fibroblastic cells observed in cultures of presumptive QJ tissue –see Table 5). Even the timing of cartilage formation in explant culture is not stage-dependent as the onset of chondrogenesis was the same in all categories of the experiment, that is at 4 days after explantation. These findings are in marked contrast to the earlier conclusion made from the organ culture results. For the same tissue now grown in explant culture the concept of temporal ‘commitment’ seems inappropriate in that the age of tissue, or stage of embryo from which the tissue is removed prior to explantation, is not related to cartilage formation. Periosteal tissue of any stage-even from the presumptive tissue of the early maxillary process, is capable of expressing a chondrogenic potential in explant culture. In some way explant culture conditions appear to mimic or simulate the in ovo events which lead to a chondrogenic commitment and the subsequent formation of secondary cartilage in vivo or in organ culture. Indeed explant culture can apparently reverse the effects of blocking these in ovo events : periosteal tissue from paralysed embryos, which fails to form cartilage in organ culture, regularly produces cartilage in explant culture. The explant culture environment may therefore be regarded as ‘permissive’ in the sense that a simple change in culture conditions and one lacking any element of specific informational content, results in a specific change in the fate of the cells concerned and in their pattern of gene expression. The ability to form cartilage, which in the QJ system in vivo is ostensibly restricted to a precise region of the hook, and both in vivo and in organ culture is only expressed in a stage-dependent fashion, is now expressed by cells from any part of the QJ and without relationship to the age of tissue concerned.

The results shown in Table 4 and to a lesser extent those in Table 5, indicate that latent chondrogenic potential is widespread throughout the periosteum and not restricted to the periosteum overlying the hook. In the intact embryo this potential to form secondary cartilage is normally expressed only by the periosteum overlying the hook but it can be concluded from the present results that ectomesenchyme cells with a latent chondrogenic potential are widely distributed in association with membrane bone. One interpretation of the present results is that only those cells which are finally located at the sites of presumptive joints subsequently express this potential and proceed to form secondary cartilage. Cells with equivalent potential and located elsewhere do not do so but when placed in explant culture will form secondary cartilage. Some evidence in support of this interpretation of widespread but unexpressed chondrogenic potential is provided by the observation that during the repair of fractured membrane bones cartilage is formed in spite of membrane bones not going through a cartilaginous phase during their development (Girgis & Pritchard, 1958). One source of such cartilage is apparently the periosteum and if the periosteum is removed prior to experimental fractures being created, this ‘periosteal’ cartilage does not form (Hall & Jacobson, 1975).

(c) Chondrogenesis and tissue submersion

It is not possible to decide if explant culture conditions promote chondrogenesis by stimulating the microenvironment created by movement and intermittent pressure at a developing joint or mimic this effect by some other mechanism. However, the fact remains that cartilage does form in explant culture when it would not do so in vivo or in organ culture. By considering the fundamental differences between the two culture techniques can we learn why explant conditions promote chondrogenesis so readily? In standard organ culture tissue is grown at the gas medium interface on a filter supported by a stainless steel mesh platform and skeletal tissues tend to grow as a solid tissue rather than spread out. In contrast tissue in explant culture is allowed to sink, attach to the floor of the culture vessel and the cells grow out, initially using the floor of the vessel as a substrate; growth is essentially two-dimensional in nature and centrifugal in direction. In addition to the fact that there is a size discrepancy between the two techniques –larger tissue samples were grown in organ culture, the two principle differences are therefore the mode of growth and the degree of submersion. In explant culture the growing tissue is submerged with the medium whereas in organ culture it is maintained at the surface of the medium. It was the second parameter –submersion, which was tested experimentally and described in Section (c) of the Results. Tissue from paralysed embryos, which in standard organ culture generally does not form cartilage, was grown in an organ culture apparatus, identical in all respects to the standard pattern except that the tissue was submerged. Whereas the contralateral controls in standard organ cultures showed virtually no cartilage formation the submerged organ cultures exhibited a high incidence of extensive in vitro chondrogenesis. Not only was there a virtual 100 % incidence of chondrogenesis but the cartilage which did form was not confined to the ‘normal’ loci found in vivo or in the earlier standard organ culture experiments. The cartilage covered a much larger surface area of the bone and examination of sectioned material revealed that extensive chondrogenesis had occurred within the membrane bone itself. Evidently the parameter of submersion resolves the difference in results from the two techniques but to what extent factors of pH, oxygen and carbon dioxide concentrations and ionic gradients may be involved is not known. Given the proposed causal relationship between anaerobiosis and chondrogenesis in some systems (e.g. Bassett & Hermann, 1961 ; Pawalek, 1969), the possibility that submersion generates relatively anaerobic conditions and thereby promotes chondrogenesis, may provide a link with events in vivo, if one effect of movement at a developing joint is to generate an anaerobic environment. In fact it has recently been demonstrated that preceding the formation of secondary cartilage in the QJ in vivo there is an increase in the specific activity of lactate dehydrogenase (LDH) –a key enzyme in anaerobic glycolysis, within the periosteal tissue of the hook and that this increase fails to occur in the absence of movement (Thorogood & Hall, 1976). Furthermore, the maintained increase in LDH activity occurring during chondrogenesis is associated with a proportionate increase in the ‘anaerobic’ isozymic forms of LDH, both in this system (Coffin & Hall, 1974) and during limb chondrogenesis in the mouse (Thorogood & Law, 1979). However, this isozyme transition probably reflects an adaptation by chondrocytes to the anaerobic environment created by the subsequent accumulation of extracellular matrix and is unlikely to be characteristic of early chondrogenesis. To what extent the microenvironment created by submersion in vitro simulates or mimics the microenvironment created by movement and pressure at a developing joint in vivo remains to be investigated. Nevertheless this simple manipulation in culture does provide a means of causing large numbers of periosteal cells to become chondrogenic and to differentiate in a way in which they would not have done otherwise.

(d) The neural crest and chondrogenesis

The subject of cell lineages and the relationships between various cell types in the periosteum has not been investigated in this work. The question remains as to whether cartilage and bone formation in this system is the result of modulation in synthetic activity by a single skeletal progenitor population or is due to the selective amplification and repression of the activities of two discrete progenitor populations, one chondrogenic and the other osteogenic. Hall and his colleagues have presented considerable circumstantial evidence for the former interpretation (reviewed Hall, 1978) but in the absence of a clonal analysis the second interpretation remains equally plausible. However, the cells forming cartilage in this system, like their osteogenic counterparts forming the membrane bone, are ectomesenchyme cells derived from the cranial N.C. or more precisely, from the mesencephalic level of the cranial N.C. (Le Lievre, 1978, personal communication). Thus the cells from this level of the crest differentiate into three principal skeletal tissues – the primary cartilage of the sclera and of the quadrate, the membrane bones of the upper jaw and the secondary cartilages associated with these bones (Le Lievre, 1974,1978 ; Noden, 1978). In the differentiation of the first two of these tissue types, epithelio-mesenchymal interactions have been shown to be involved (Newsome, 1972, 1976; Tyler, 1978) although precisely what the role these interactions play in the progressive restriction of the cells, their commitment and the subsequent expression of the differentiated state, is unclear at present. For the third tissue type—secondary cartilage, tissue interactions are apparently not involved in commitment or expression. It has been proposed in the present work that the differentiation of this particular type of cartilage is controlled in an epigenetic fashion as it is merely the spatial position of the cell after migration which determines whether or not it will be permitted to express its chondrogenic potential. It has been previously suggested that N.C. cells differentiate, not according to their point of origin along the anterioposterior axis but according to morphogenetic factors encountered along the migration pathways and/or at the site where migration is arrested (Le Douarin & Teillet, 1974, Morris & Thorogood, 1978). If that generalization is correct then this mechanism controlling secondary cartilage formation provides a clear example of morphogenetic factors at the final site controlling the development of a N.C.-derived tissue.

I express thanks to my former colleagues in the Zoology Department, Oxford, for helpful discussion during the course of the work, and to Jim Bee for carrying out the immuno-fluorescent analysis for me using facilities very kindly provided by Klaus von der Mark at the Max Planck-Institut fur Biocheme, Munich. The work was supported by a Scientific Investigations Grant from the Royal Society.

Bassett
,
C. A. L.
&
Herrmann
,
I.
(
1961
).
Influence of oxygen concentration and mechanical factors on differentiation of connective tissues in vitro
.
Nature, Lond
.
190
,
460
461
.
Coffin
,
P. A.
&
Hall
,
B. K.
(
1974
).
Isozymes of lactate dehydrogenase (LDH) in skeletal tissues of the embryonic and newly hatched chick
.
J. Embryol. exp. Morph
.
31
169
181
.
Drachman
,
D. B.
&
Sokoloff
,
L.
(
1966
).
The role of movement in embryonic joint development
.
Devi Biol
.
14
,
401
420
.
Girgis
,
F. G.
&
Pritchard
,
J. J.
(
1958
).
Experimental production of cartilage during the repair of fractures of the skull vault in rats
.
J. Bone Jt. Surg
.
40B
,
274
281
.
Hall
,
B. K.
(
1968
).
In-vitro studies on the mechanical evocation of adventitious cartilage in the chick
.
J. exp. Zool
.
168
,
283
306
.
Hall
,
B. K.
(
1972
).
Immobilization and cartilage transformation into bone in the embryonic chick
.
Anat, Rec
.
173
,
391
404
.
Hall
,
B. K.
(
1978
).
Developmental and Cellular Skeletal Biology
,
New York
:
Academie Press
.
Hall
,
B. K.
&
Jacobson
,
H. N.
(
1975
).
The repair of fractured membrane bones in the newly hatched chick
.
Anat. Rec
.
181
,
55
70
.
Hamburger
,
V.
(
1960
).
A Manual of Experimental Embryology
.
Chicago
:
The University of Chicago Press
.
Hamburger
,
V.
,
Balaban
,
M.
,
Oppenheim
,
R.
&
Wenger
,
E.
(
1965
).
Periodic motility of normal spinal chick embryos between 8 and 17 days of incubation
.
J. exp. Zool
.
159
,
1
14
.
Hamburger
,
V.
&
Hamilton
,
H.
(
1951
).
A series of normal stages in the development of the chick embryo
.
J. Morph
.
88
,
49
92
.
Horstadlus
,
S.
(
1950
).
The Neural Crest. Its Properties and Derivatives in the Light of Experimental Research
.
Oxford University Press
.
Johnston
,
M. C.
(
1966
).
A radioautographic study of the migration and fate of cranial neural crest cells in the chick embryo
.
Anat. Rec
.
156
,
143
156
.
Katoh
,
Y.
&
Takayama
,
S.
(
1977
).
Establishment of clonal cell lines of hamster chondrocytes maintaining their phenotypic traits
.
Expl Cell Res
.
106
,
285
291
.
Le Douarin
,
N. M.
&
Teillet
,
M. A.
(
1974
).
Experimental analysis of the migration and differentiation of neuroblasts of the autonomic nervous system and of neurectodermal mesenchymal derivatives, using a biological cell marking technique
.
Devi Biol
.
41
,
162
184
.
Le Lievre
,
C.
(
1974
).
Role de cellules mesectodermiques issues des crêtes neurales céphaliques dans la formation des arcs branchiaux et du squelette visceral
.
J. Embryol. exp. Morph
.
31
,
453
477
.
Le Lievre
,
C.
(
1978
).
Participation of neural crest-derived cells in the genesis of the skull in birds
.
J. Embryol. exp. Morph
.
47
,
17
37
.
Le Lievre
,
C.
&
Le Douarin
,
N. M.
(
1975
).
Mesenchymal derivatives of the neural crest: analysis of chimaeric quail and chick embryos
.
J. Embryol. exp. Morph
.
34
,
125
154
.
Lewis
,
J. H.
&
Wolpert
,
L.
(
1976
).
The principles of non-equivalence in development
.
J. theor. Biol
.
62
,
479
490
.
Lison
,
L.
(
1954
).
Alcian blue 8G with chlorantine fast red 5B: a technic for selective staining of mucopolysaccharides
.
Stain Technol
.
29
,
131
138
.
Morris
,
G. M.
&
Thorogood
,
P. V.
(
1978
).
An approach to cranial neural crest migration and differentiation in mammalian embryos
.
In Development in Mammals
, vol.
3
(ed.
M. H.
Johnson
).
Amsterdam
:
North Holland Publishing Company
.
Murray
,
P. D. F.
(
1963
).
Adventitious cartilage in the chick and the development of certain bones and articulations in the chick skull
.
Aust. J. Zool
.
11
,
368
430
.
Murray
,
P. D. F.
&
Drachman
,
B. D.
(
1969
).
The role of movement in the development of joints and related structures: the head and neck in the chick embryo
.
J. Embryo, exp. Morph
.
22
,
349
371
.
Murray
,
P. D. F.
&
Smiles
,
M.
(
1965
).
Factors in the evocation of adventitious (secondary) cartilage in the chick embryo
.
Aust. J. Zool
.
13
,
351
381
.
Newsome
,
D. A.
(
1972
).
Cartilage induction by retinal pigmented epithelium of the chick embryo
.
Devi Biol
.
27
,
575
579
.
Newsome
,
D. A.
(
1976
).
In vitro stimulation of cartilage in embryonic chick neural crest cells by products of retinal pigmented epithelium
.
Devi Biol
.
49
,
496
507
.
Noden
,
D. M.
(
1975
).
An analysis of the migratory behaviour of avian cephalic neural crest cells
.
Devi Biol
.
42
,
106
130
.
Noden
,
D. M.
(
1978
).
The control of avian cephalic neural crest cyto-differentiation. I. Skeletal and connective tissues
.
Devi Biol
.
67
,
296
312
.
Paul
,
J.
(
1965
).
Cells and Tissues in Culture
, 3rd edition.
Edinburgh and London
:
E and S Livingston Ltd
.
Pawalek
,
J. M.
(
1969
).
Effects of thyroxine and low oxygen tension on chondrogenic expression in cell culture
.
Devi Biol
.
19
,
52
72
.
Solursh
,
M.
,
Meier
,
S.
&
Vaerwyck
,
S.
(
1973
).
Modulation of extracellular matrix production by conditioned medium
.
Amer. Zool
.
13
,
1051
1060
.
Thorogood
,
P. V.
&
Hall
,
B. K.
(
1976
).
The use of variable lactate/malic dehydrogenase ratios to distinguish between progenitor cells of cartilage and bone in the embryonic chick
.
J. Embryol. exp. Morph
.
36
,
305
313
.
Thorogood
,
P. V.
&
Law
,
N.
(
1979
).
Lactate dehydrogenase isoenzyme transitions during skeletal differentiation in the mouse embryo
.
J. Develop. Physiol
.
1
,
81
90
.
Tyler
,
M. S.
&
Hall
,
B. K.
(
1977
).
Epithelial influences on skeletogenesis in the mandible of the embryonic chick
.
Anat. Rec
.
188
,
299
240
.
Tyler
,
M. S.
(
1978
).
Epithelial influences on membrane bone formation in the maxilla of the embryonic chick
.
Anat. Rec
.
192
,
225
234
.
Von Der Mark
,
H.
,
Von Der Mark
,
K.
&
Gay
,
S.
(
1976
).
Study of differential collagen synthesis during development of the chick embryo by immunofluorescence. I. Preparation of collagen type I and type II specific antibodies and their application to early stages of the chick embryo
.
Devi Biol
.
48
,
237
249
.
Von Der Mark
,
K.
&
Von Der Mark
,
H.
(
1977
).
Immunological and biochemical studies of collagen type transition during in vitro chondrogenesis of chick limb mesodermal cells
.
J. Cell Biol
.
73
,
736
747
.
Weston
,
J. A.
(
1970
).
The migration and differentiation of neural crest cells
.
Adv. Morphog
.
8
,
41
114
.