The work presented, in this and the subsequent papers of a series, was begun in order to re-examine the properties of the amphibian primary embryonic field, in the light of current theories concerning the nature of individuation fields in developing animal systems. A detailed description is given of the basic operation whose results are described in this and the subsequent paper. This involves the transplantation, into a late blastula or stage-10 gastrula host, of a supernumerary stage-10 organizer region. The consequences of such operations during the following 4–6 h, up to the late gastrula stage, are also described.

Evidence is presented that, from a time some 2·5 h before the organizer site first becomes externally visible, its presumptive region is immune from interference by the proximity of another, implanted organizer, even one which is itself 2·5 h older. That is to say, the final site of development of host organizer activity is not altered by the presence of such an implant.

Pairs of early organizers at comparable stages of activity appear to set up competing fields of cellular orientation and immigration, which show a fairly sharp boundary at their interface. This is most obvious for pairs of organizers fairly close together, since the cell polarization and stretching is most pronounced in the region near to the apex of the field, i.e. the initial site of cell immigration. Independent initial fields of immigration due to two organizers can reliably be distinguished in cases where they are as little as 30° of angular distance apart in the marginal zone of the host.

These results are to be considered in relation to those of Paper II, for the same series of operations, where the final patterns of cell differentiation are studied, and to those of Paper III, where evidence is given for the development of autonomous polarity in the region of the organizer.

Classical work in experimental embryology, particularly that of Spemann & Mangold (Spemann, 1938) (see Paper III in this series (Cooke 1972b)), has established the region around the earliest visible organizer activity in the amphibian embryo as the dominant or apical zone, for the subsequent organization of an accessory axial field, following its transplantation into host gastrulae. This region includes the presumptive anterior head endoderm and mesoderm and the earliest-forming dorsal lip of the blastopore and is capable in Xenopus, when isolated and transplanted as a cell group into a site far from the comparable cells of a host embryo, of setting up a field for a complete accessory set of axial structures, including the tailbud, and involving host material.

According to the concensus of classical work using the differentiation of transplanted cells and of parts of embryos explanted into culture (e.g. Holtfreter & Hamburger, 1955), this period of primary organization comes to a close during the latter part of gastrulation or in the early neurula (after some 15 h development at 21 °C in Xenopus), when the zones of autonomous differentiation into endodermal and mesodermal structures become fixed in a mosaic manner. However, the final effects of disturbances imposed upon the field before this time are best observed and recorded at considerably later stages, where the pattern of secondarily induced neural and other surface structures affords much more visual information. There is evidence that it reflects rather closely the form of the field in the underlying mesoderm and endoderm.

The beginning of the period is harder to define. Recent work has tended to push this back into earlier stages of development (e.g. Curtis, 1962 a, b; Nieuwkoop, 1969 a, b), and it is now considered that the maintenance of the primary field, which has an apex or centre of physiological dominance in some sense in the region of the presumptive boundary between endoderm and mesoderm, begins at least in early cleavage. For the purpose of these studies, however, the earliest time that it can be investigated experimentally is early in the stage-8 blastula, the large cell size making cutting and transfer of cell groups rarely successful before this time.

Most of the experiments described in this and the subsequent two papers are classical in nature, although perhaps new in detail. They were undertaken in the belief that it is worth re-examining such classical material in the light of current theories concerning the detailed nature of the homeostatic fields of information that lead to spatial patterns of cell behaviour and differentiation during embryogenesis. These are the ‘individuation fields’ of Waddington (see Waddington, 1966), and the whole of the organization coming to exist during amphibian gastrulation can probably be considered as due to the operation of a single such field. No empirical description of what happens in the hands of any one experimenter is ever complete, and we have reason to expect that, watching the results of various systematic disturbances of this early embryonic field, we shall be led by current theory to consider different types of information as relevant, and perhaps to notice new phenomena.

The current state of theory concerning fields underlying pattern formation has been recently reviewed by Wolpert (1969), and, with differing emphases, by Cohen (1971), Lawrence (1970), and Crick (1970), in the course of adding their own contributions. The formulation of the problems developed by Wolpert, and particularly his term ‘positional information’ for the variable, perceived by cells, and supplying at each point within a field a measure of relative position within that field, will be utilized throughout the discussion in these papers.

The present paper is the first of a series in which it is planned to present the results and discuss the implications of a series of experiments, using embryos of Xenopus laevis, the South African clawed toad, during the period of primary organization. This period is defined as that during which the first of the hierarchy of fields of positional information (see Wolpert, 1969), that underlie patterns of cellular differentiation, is (a) erected and maintained in a regulatory manner, and (b) expressed. Such expression firstly takes the form of a pattern of cell locomotory and adhesive behaviour, causing gastrulation, and subsequently of the division of the mesodermal and endodermal mantle of the early neurula into zones of fixed differentiation potency.

Here, the basic experiment forming the subject of the first three papers is described in some detail. It consists of supplying a blastula or beginning (stage-10) gastrula with an extra beginning organizer (or head organizer region) from a stage-10 donor, situated in the marginal zone at varying distances from the future or actual site of host organizer activity. Observations are then given on the interactions of such pairs of organizer regions with respect to the early, cell behavioural aspects of their activity.

The results are to be considered in relation to those in the next paper, where the pattern of cellular differentiation tendencies finally obtaining in the mesodermal mantle at the end of gastrulation, and thus the pattern of axial organization seen at the late neurula stage, is considered for the same series of embryos. The approach of these experiments towards relevance, in discriminating between theories, will then be discussed.

Supply of embryos

Xenopus laevis, obtained as adults in Africa via Harris Biological Supplies Ltd., were kept in the laboratory at 21 °C, being fed on fresh chopped beef heart. Batches of eggs were obtained by matings following on the injection of chorionic ganadotrophin (‘Chorulon’ Organon Laboratories, Ltd.), 300 i.u. for females and 150 i.u. for males. Eggs were demembranated with the aid of forceps during stages 7–9, and allowed to heal in half strength Holtfreter solution for 20 min before storage, until use, on a 2 % agar bed in 0·1 strength Holtfreter. The span of time available for operations was increased by keeping groups of eggs at 19 and 25·5 °C, but all embryos were equilibrated to 21 °C before use and, unless otherwise stated, operations were allowed to develop at an ambient temperature of 21 °C.

The work was carried out using repeated layings of eggs from 10 pairs of toads, and operations of all of the different types used for comparison in this and 2 subsequent papers of the series, were obtained from each batch of eggs. Thus any effect due to inter-female variance in the properties of the eggs was randomized over the whole pattern of results. Staging of embryos was according to the table of normal development for Xenopus, due to Nieuwkoop & Faber (1956).

Solutions used

  • Holtfreter solution, buffered to pH 7·2 with 0·56 g/1. of Tris (Sigma, ‘puriss.’) and HC1, and with 20 mg/1. of NaHCO3 added after autoclaving. Used at half strength and at 0·1 strength.

  • Ca2+ and Mg2+ free Holtfreter, +150 mg/1. EDTA, buffered to pH 8·2 with 0·56 g/1. Tris and HC1. Used at half strength.

Solutions were autoclaved, but used diluted with new, unautoclaved doubleglass-distilled water.

Treatment of embryos during and after operation

Only embryos apparently perfectly healed after demembranation were used as hosts, although donors included some embryos with slight unhealed wounding or malformation in regions far from the developing organizer at stage 10.

Immediately before all operations, both host and donor embryos were transferred by pipette to the disaggregation solution (EDTA, pH 8·2), remaining in this for 2·5–3 min before transfer to an agar bed, equilibrated with half strength Holtfreter which just covered the embryos. This procedure was found to enable excision of stage-10 organizers and their implantation into a gap made in the host cell sheet, with a minimum production of cell debris, and to cause neater and more consistent annealing of the cellular structure of graft and host, as both together recovered the adhesive and contractile properties of their cells. Control experiments had shown that development was completely normal after this treatment and simple wounding.

Following operations, which were done with tungsten microknives kept sterile in boiling water, hosts were rotated, site of implantation downwards, in shallow depressions in the agar, and left for a 35-min healing period. At inspection following this, all embryos showing continuing loss of yolky material and unclosed wounding of the surface, after gentle washing with a micropipette, were discarded, and the rest left to develop, animal pole uppermost, on glass, covered by a shallow layer of 0·1 strength Holtfreter. In cases where continuing observation of the marginal zone was required for some hours following operation, embryos were kept instead on agar under 0·1 strength Holtfreter, thus avoiding damage consequent upon movement during the period of temporary cell adhesion to glass, which lasts until the close of gastrulation.

Surface observation of immediate post-operative events was performed under the dissecting microscope, with schematic drawing of typical or otherwise interesting cases.

The basic operation

Fig. 1 depicts the basic operation giving rise to the results which are the subject of this and of much of the next paper in this series (Cooke, 1972 a). Variants of this operation, and new types of operation, for all of which the basic treatment of embryos described above was also used, will be described in the results section as they arise.

Fig. 1.

The basic operation, (a) Site of extirpation of early organizer from a stage-10 donor, sagittal section, and of implantation into the marginal zone of a host, frontal view, (b) Vegetal aspect of host at stage 10, showing own organizer just visible (top) and an implant in situ at ca. 80°.

Fig. 1.

The basic operation, (a) Site of extirpation of early organizer from a stage-10 donor, sagittal section, and of implantation into the marginal zone of a host, frontal view, (b) Vegetal aspect of host at stage 10, showing own organizer just visible (top) and an implant in situ at ca. 80°.

The situation under study is that where a plug of cells representing the head organizer at stage 10, is implanted into a gap made in the presumptive zone of marginal intucking of a host embryo, which may itself be anywhere between the stages 8 and 10 at the time of operation. All donor organizers were used within 20 min of their first becoming visible, including external signs of the dorsal lip, and internally, the bottle-shaped cells of Holtfreter together with most of the presumptive head endoderm. In Xenopus, such a plug of cells is potentially capable of organizing a field in the host to give rise to a completely individuated second axis. However, if due to later excision of the graft, a few of the inner, most ‘apical’ cells are missed and left in the donor, capacity to promote individuation of the most cephalad levels in the induced axis is sometimes lost.

Once the host’s own gastrulation activity commences, the angular distance between host and grafted dorsal lips, along the circle in the marginal zone lying in a plane parallel to the equator of the blastula, can be estimated to within approximately 10° of arc, and is referred to as the inter-organizer angle. Obviously, this angle can only be controlled and measured at the time of opera-tion in the limiting case where the host is already itself at stage 10. In the other limiting case, that of early stage-8 blastula hosts, it does not become visible for approximately 3 h after operation, or 2·5 h after healing inspection. For midstage-9 hosts at operation, the corresponding times are 1 h and 0·5 h.

It will also be apparent from Fig. 1 that the orientation of the implanted organizer may be ‘normal’ with respect to the presumptive orientation of the host’s own organizer, or else ‘reversed’ through 180° in the sense that the presumptive axial mesodermal component of the graft is now nearer the host’s vegetal pole, and the presumptive head endoderm nearer the host’s animal pole. Heterogeneity of cell size, etc., on the outer surface of the graft, enabled this orientation to be controlled with a high degree of reliability in operations.

The excised plug of organizer cells was allowed to heal up and become compact, by cell re-adhesion, before implantation into a newly made recipient site in the host. In this way, graft-host contact and internal –external positioning of the graft were found to be least variable.

For operations performed on early hosts there will clearly occur, in a few cases (though see below) a co-incidence of the graft site with the presumptive organizer site of the host. Here a singly organized field of gastrulation activity might be expected to result, and this is indeed seen after a significant proportion of operations. Lines of host cell intucking activity then progress as in normal gastrulation, from the grafted organizer, but do not commence until gastrulation in controls synchronous with the host embryo is also commencing.

Grafts which lie in positions more distant from that of the host organizer induce their own fields of intucking activity, which also progress in time with, or slightly retarded relative to, those of the hosts. This occurs after a quiescent period varying according to the age of the host at operation (and absent in the case of operations on stage-10 hosts). During this period the graft, after having healed in and formed a very local site of intucking, shows little further activity except for a slow progressive deepening of the flask-shaped cavity in the surface of the marginal zone, which continues until gastrulation begins in the host.

The subsequent field of intucking, involving host cells around the graft, in most cases extends until it coalesces with the extending blastopore rim due to the host organizer. In some cases, this coalescence requires slight regulation along the zone of the presumptive blastoporal lip, in the sense of a slight alteration in the progressing ends of the two lips so that they meet, when their presumptive lines would not always have caused this. Cases where such regulation has to be excessive, or fails due to animal-vegetal misplacement of the graft, are rejected.

Normally, such dual fields of cellular activity co-ordinate, so that by midgastrula stages the double origin of intucking is no longer apparent, and such embryos will appear normal until neural induction is visible. The mesodermal mantle infolds and is carried forward, behind these two leading points, in a co-ordinated way, to give an early neurula having both a normal, median antero-posterior axis of elongation, and an additional head organizer region lying parallel to the host’s and at the presumptive anterior end, as shown by vital staining of grafts with Nile blue sulphate.

The organization of embryos having such grafts, during subsequent developmental stages, forms the main subject of the second paper in this series.

Fig. 2 shows the distribution of inter-organizer angles finally observed, in the total sample of operations where the host was found not to have reached stage 10 (appearance of the beginning dorsal lip) when examined 35 min after implantation of the graft. Thus hosts in this series ranged from stage 8 at time of operation, when in the extreme case inter-organizer angle could not be observed until 3 h later at 21 °C, up to stage 9, when their own organizer first became visible some 45 min later. 30° is an estimate of the mean smallest angle at which two discrete early dorsal lips, and the cellular activity associated with each of them, could reliably be distinguished at host stage 10. Nine out of the 12 operations showing two organizer apices, where angle was estimated as less than 10° above this threshold, had in fact been made in stage-8 hosts 2 ·5 –3 h previously. The distribution observed for classes of angles between this and 180°, over the whole sample, does not deviate significantly from that to be expected on the hypothesis that an additional organizer, implanted into the marginal zone of a blastula, exerts no influence upon the position of development of the host’s own organizer, if this is presumptively at least 30° away from the position of the graft. This remains true even in cases where the grafted organizer has been healed in place for up to 2 ·5 h. It is suspected that the very initial sites of organizer activity, with respect to this cell behaviour (i.e. apparent cell polarization or at least stretching, and then immigration), remain autonomous at distances below 30°, but observation does not permit certainty on this point (see Discussion).

Fig. 2.

Total numbers of organizer pairs observed within various ranges of angular distance in the marginal zone at onset of host stage 10, following operations performed on hosts of stages 8 and 9.

Fig. 2.

Total numbers of organizer pairs observed within various ranges of angular distance in the marginal zone at onset of host stage 10, following operations performed on hosts of stages 8 and 9.

Comparison with control embryos synchronous with those to be used as hosts in operations gave no evidence that either the operative procedure per se, or the subsequent presence of the additional organizer tissue, caused any retardation or acceleration in the onset of host gastrulation

Fig. 3 shows schematic drawings taken from two hosts operated on at stage 8, and two at stage 9, at each of three times separated by 20 min of development, and beginning with the onset of host stage 10. It can be seen that in the case of stage-8 operations, by the first observation the graft has sunk further into the host marginal zone, maintaining contact with the exterior via a patent ‘neck’ of cells, and that for both stage-8 and stage-9 operations there are no lines of cellular intucking proceeding from the grafts as would be occurring in the case of normal gastrulation, begun 1 or 2 h previously. A further feature is the tendency, once host gastrulation has commenced and the graft been re-activated some short while later, for lines of cellular intucking to be extended faster in the stretch of marginal zone between two organizers situated relatively close together. The drawings shown are selected as demonstrating particularly clearly these tendencies, which are suggested throughout the series of operations. In the limiting case, however, where the host reaches stage 10 between operation and the healing inspection after 35 min, cellular activity at graft and host sites usually keeps pace.

Fig. 3.

Drawings, from observation, of four operated embryos, made at three successive times 20 min apart, commencing with earliest host organizer activity (stage 10). Embryos 1 and 2, stage 9 at time of operation, embryos 3 and 4, stage 8 at time of operation. H = site of host organizer activity. 1 = implanted organizer.

Fig. 3.

Drawings, from observation, of four operated embryos, made at three successive times 20 min apart, commencing with earliest host organizer activity (stage 10). Embryos 1 and 2, stage 9 at time of operation, embryos 3 and 4, stage 8 at time of operation. H = site of host organizer activity. 1 = implanted organizer.

When the region of an early dorsal lip is observed under fairly high power, and also in time-lapse cine films of the commencement of grastrulation made in this laboratory, the cells of the marginal zone are seen to be elongated parallel to the surface, and with their long axes orientated towards the small area of initial intucking. The arrangement of this local area of markedly anisodiametric cells is approximately radial for about the first 20 min after the inception of stage 10. Subsequently as the lines of marginal inrolling extend, they are seen at this cellular level to consist of the juxtaposition of two zones of cells, each orientated towards the initial organizer position or positions, as shown in the drawing of Fig. 4b. Orientation is usually much more exaggerated in the upper (prospective mesodermal mantle) cells. At later stages of gastrulation, when the ring-shaped zone of inrolling is almost or quite complete, and the invagination of the mesodermal mantle has proceeded some way at least dorsally, the orientation of the cell boundaries becomes much less extreme, and is more normal to the blastopore lip. The impression gained is that the cells are being passively stretched into these conformations. However, they may in fact represent, or be the basis for the subsequent erection of, intrinsically polar properties on the part of individual cells, set up under the influence of some type of orientating stimulus from the organizer apex. There is evidence for other types of expression of intrinsic polarity on the part of small groups of cells taken from near the beginning dorsal lip (see Paper III of this series).

Fig. 4.

Drawings to show configurations of cell borders in (a) earliest activity at a single organizer site, high-power view, and (6) marginal zone containing two organizers relatively close together, soon after onset of stage 10. Heavy arrows = centre of each organizer activity. Light arrows = apparent boundary between zones of influence of the two centres.

Fig. 4.

Drawings to show configurations of cell borders in (a) earliest activity at a single organizer site, high-power view, and (6) marginal zone containing two organizers relatively close together, soon after onset of stage 10. Heavy arrows = centre of each organizer activity. Light arrows = apparent boundary between zones of influence of the two centres.

In the sector of marginal zone existing between pairs of organizers in relative proximity, the initial line of cell intucking, which presumably also demarcates zones of differing prospective fate in normal embryogenesis, gives evidence of progressing precociously (see above). In cellular terms, the ‘watershed’ between the respective zones of cellular orientation due to the two organizer apices is often plainly visible in the prospective mesodermal mantle area (see Fig. 4), and to a lesser extent in the nearby endoderm cells, in the very early stages of gastrulation. There also seems to be an effect whereby the course of the zone of intucking is shortened in the region between organizers by its becoming more nearly a part of a great circle on the gastrula surface, passing through both organizer apices and thus shifting vegetally relative to its normal prospective location. This can occur to a considerable extent without gastrulation being otherwise disturbed.

As seen in Fig. 2, the proportion of the total sample of operations made into hosts at stages 8 –9, where there was the subsequent appearance either of single organizer activity or of two centres fusing within minutes of the onset of their activity, was 22/140. Since the minimum angular distance at which two separable centres could reliably be observed was estimated at 30°, the above-mentioned proportion is consistent with the hypothesis that the presence of an implanted stage-10 organizer exerts no influence upon the development of the host organizer at its own presumptive site, even if this is within 30° of the graft site, and even though the graft may have been healed in for up to 2 ·5 h. Such a hypothesis leads to a theoretically expected proportion of 30/180 apparently singly organized gastrulations.

More than half of the hosts used in this sample were in fact at stage 8 at the time of operation, and many of the more extreme instances of the closeness of two active sites were observed in these cases.

There can be no direct evidence as to whether the non-interference between organizer sites extends within the 30° zone. However, in more than half the cases of apparently single organization, the spreading of intucking activity among host cells was markedly asymmetrical on either side of the graft site in its initial stages, suggesting that autonomy still applies even in these limiting cases.

On the hypothesis of an ultimate universality of mechanism for regulative pattern-forming fields, and considering that the stability requirements for the developing bilaterality of the amphibian mesodermal field are formally similar to the requirements for stability in the axial organization of hydroids, one might expect inhibitory properties on the part of relatively advanced apical sites, such that the development of similar sites within a certain distance in the same cell sheet might be prevented. Classical work on the properties of the dorsal lip as organizer has established that it fulfils the criteria of an apex for both cephalo-caudal and bilaterality axes of positional information in amphibian development. The operation of an inhibitory mechanism, by analogy with the behaviour of regenerating Hydra (Webster, 1966 a, b,), would make an already visible stage-10 organizer region equivalent to a determined hypostome region, and the presumptive organizer site during the 2 or 3 h before this, equivalent to a subhypostomal region as yet undetermined to be hypostome, but moving metabolically in that direction. The pattern of results to be expected from a sample of operations such as the present one would then be that of a great deficit in relatively small observed angular distances, and/or a larger than expected number of both single organization and relatively large angular distances, near 180°, dependent upon whether stage-10 organizers merely suppressed organizer development within a certain inhibitory field around them or whether some more complex relationship obtained.

It had seemed at least plausible to look for such a pattern of labile interaction at the outset, in view of the results of Curtis (1962 a, b) which indicate that, although in very early few-celled stages as in the uncleaved egg, the presumptive site of organizer formation behaves as a mosaic, chemo-determined entity, yet between the large-celled blastula stage and the onset of gastrulation, some 6 h later, the organizer field is regulative, restoring presumptive apical parts removed from it. Also, Huxley & de Beer (1934) quote results to the effect that lithium treatment, and possibly temperature gradients, will alter in an animal direction the site of dorsal lip formation in urodeles. In most classical grafting experiments, as in those done on the cortex of few-celled stages of Xenopus by Curtis (1962 a), the position of the graft has been at what corresponds in the present experiments to a large inter-organizer angle. However, the conclusion from the present results is that any state of chemo-differentiation leading to the initial cell-behavioural activity of the head organizer, is sufficiently well established in these cells by stage 8 to render them autonomous in face of any adjustments of positional information that can be caused by proximity of a group of their stage-10 homologues. This could mean either that the period of 2 ·5 h allowable is below that required by the relatively slow dynamics of positional-information adjustment in this system, even over the short distances involved, or that by stage 8 the presumptive initial dorsal lip region is equivalent to a ‘determined’ hypostome region of Hydra. The types of experiment performed so far in attempts to estimate the absolute times required for inhibitory-type re-adjustments in other systems (e.g. Wilby & Webster, 1970), have been unable to eliminate the unknown constant due to the time taken, by the site to be inhibited, to reach the developmental threshold for autonomy against further inhibitory influences. Thus we are unable to assess the relative plausibility of the two alternative explanations for this early organizer autonomy.

If further experiments should lead to the conception that, in the amphibian mesodermal/endodermal zone, a field is established, many hours before the onset of visible cell activity at its apex, of such a nature that there cannot be inhibitory interaction between two presumptive apices even where one is considerably further advanced than the other, this nevertheless would not imply a lack of inhibitory interaction between the apex and the subapical levels within each unitary field. Indeed any theory, such as that of Rose (1952, 1957), utilizing a gradient of rate of physiological progression or activity, or one utilizing a diffusion gradient of substance emanating from a localized source at the apex, which must nevertheless explain the regulative capacity that the amphibian field shows (see, for example, Curtis, 1962 b), more or less requires the added concept of inhibition in order to maintain stability of the positional information over the number of hours required.

A further feature of the organization of gastrulation that emerges from the present observations is that time of onset of visible gastrulation activity is host controlled, and unaffected by the presence of a stage-10 organizer during the 2 ·5 h preceding. Grafts implanted in pre-stage-10 embryos become quiescent after initial healing and ‘pocket-forming’ activity, only to be re-activated, usually detectably later than the onset of host gastrulation. The signal for onset must thus be some global state of the embryo (such as cell number, cell size in the region of the organizer, or else some intrinsic time-keeping mechanism in the cells) rather than that of the organizer apex itself.

Furthermore, the host’s own organizer appears to be at the summit of some gradient such that activity starts there first, whereas a foreign organizer, even one that has been healed in for some 2 ·5 h, is unable to activate cells near to it for some minutes. This again suggests the presence of a deep-seated field of some nature in the marginal zone, centred round the host presumptive dorsal lip and only susceptible to relatively slow modification by a new, local apex. The precocious lip formation that tends to occur between relatively close organizers also fits with the concept of an underlying gradient determining the rate of onset of a new cell state, and perhaps itself caused by the organizer as its summit. Such phenomena would seem to be further expressions of the dorsolateral ‘gradient’ postulated for amphibian embryos at this stage by Dalcq & Pasteels (1937).

I am grateful for discussion during the course of the work presented in this and the following two papers of the present series and for criticism during preparation of the manuscripts, to Dr B. C. Goodwin, Professor Lewis Wolpert and Dr Louie Hamilton. The work was carried out under the support of the Science Research Council, received as part of a research grant to Dr Goodwin.

Cohen
,
M. H.
(
1971
).
Models for the Control of Development
.
Cambridge University Press
.
Cooke
,
J.
(
1972a
).
Properties of the primary organization field in the embryo of Xenopus laevis. II. Positional information for axial organization in embryos with two head organizers
.
J. Embryol. exp. Morph
.
28
,
27
46
.
Cooke
,
J.
(
1972b
).
III. Retention of polarity in cell groups excised from the region of the early organizer
.
J. Embryol. exp. Morph
.
28
,
47
56
.
Crick
,
F. H. C.
(
1970
).
Diffusion in embryogenesis
.
Nature, Lond
.
225
,
420
422
.
Curtis
,
A. S. G.
(
1962a
).
Morphogenetic interactions before gastrulation in the amphibian Xenopus laevis. The cortical field
.
J. Embryol. exp. Morph
.
10
,
410
422
.
Curtis
,
A. S. G.
(
1962b
).
Morphogenetic interactions before gastrulation in the amphibian Xenopus laevis - Regulation in blastulae
.
J. Embryol. exp. Morph
.
10
,
451
463
.
Dalcq
,
A.
&
Pasteels
,
J.
(
1937
).
Une conception nouvelle des bases physiologiques de la morphogenease
.
Archs Biol
.,
Paris
.
48
,
669
710
.
Holtfreter
,
J.
&
Hamburger
,
V.
(
1955
).
In Analysis of Development
(ed.
Willier
,
Weiss
&
Hamburger
).
Philadelphia
:
Saunders
.
Huxley
,
J. S.
&
De Beer
,
G. R.
(
1934
).
Elements of Experimental Embryology
.
Cambridge University Press
.
Lawrence
,
P. A.
(
1970
).
Polarity and pattern in the post-embryonic development of insects
.
Adv. Insect Physiol
.
7
,
197
266
.
Nieuwkoop
,
P. D.
(
1969a
).
Formation of the mesoderm in urodelean amphibians. I. Induction by the endoderm
.
Wilhelm Roux Archiv. EntwMech. Org
.
162
,
341
373
.
Nieuwkoop
,
P. D.
(
1969b
).
Formation of the mesoderm in urodelean amphibians. II. The origin of the dorso-ventral polarity of the mesoderm
.
Wilhelm Roux Archiv. EntwMech. Org
.
163
,
298
315
.
Nieuwkoop
,
P. D.
&
Faber
,
J.
(
1956
).
Normal table of Xenopus laevis (Dau din)
.
Amsterdam
:
North Holland Publishing Company Ltd
.
Rose
,
S. M.
(
1952
).
A hierarchy of self-limiting reactions as the basis of cellular differentiation and growth control
.
Am. Nat
.
86
,
337
354
.
Rose
,
S. M.
(
1957
).
Cellular interactions during differentiation
.
Biol. Rev
.
32
,
351
382
.
Spemann
,
H.
(
1938
).
In Embryonic Development and Induction
.
Yale University Press
. Reprinted 1967.
New York
:
Hafner Publishing Company
.
Waddington
,
C. H.
(
1966
).
Fields and Gradients
.
In Major Problems in Developmental Biology
(ed.
M.
Locke
).
New York and London
:
Academic Press
.
Webster
,
G.
(
1966a
).
Studies on pattern regulation in Hydra. II. Factors controlling hypostome formation
.
J. Embryol. exp. Morph
.
16
,
105
122
.
Webster
,
G.
(
1966b
).
Studies on pattern regulation in Hydra. III. Dynamic aspects of factors controlling hypostome formation
.
J. Embryol. exp. Morph
.
16
,
123
141
.
Wilby
,
O. K.
&
Webster
,
G.
(
1970
).
Studies on the transmission of hypostome inhibition in Hydra
.
J. Embryol. exp. Morph
.
24
,
583
593
.
Wolpert
,
L.
(
1969
).
Positional information and the pattern of cellular differentiation
.
J. theor. Biol
.
25
,
1
48
.