ABSTRACT
By a series of explantation, transplantation (yolk syncytium left intact) and incision experiments done with the eye rudiment during the stages VI –IX (Naef, 1923) it is concluded that the yolk syncytium does not induce the differentiation of ‘the outer layer of cells’ from stage VI on as suggested by Arnold (1965 b). From the explantation and transplantation experiments the author draws the conclusion that there exists, from stage VI on, in the ‘outer layer of cells’ on each side of the embryo, an area which contains all factors necessary for eye formation and which manifests itself, under experimental conditions, in regulation. The explanted eye rudiment shows in vitro autonomous differentiation capacity only if nutritional conditions are sufficient. The incision experiments elucidate the role of ‘contractile elements’ in organogenesis. Arnold’s results are discussed.
INTRODUCTION
According to Arnold (1965b) the Loligo embryo, at early organogenetic stages (stage VI of Naef, 1923; stages XVI and XVII of Arnold, 1965a), shows three components: ‘an outer layer of cells’ (the actual embryo), ‘an inner syncytial epithelium’ (transitory formation, usually referred to as ‘yolk syncytium’), ‘a central mass of yolk’.
According to this author again, the inner layer has an inductive function. The yolk syncytium is the original egg cortex; it contains a ‘morphogenetic inductive map’ determining the differentiation, from stage VI on, of the ‘outer layer of cells’ (Arnold, 1965b, 1968). This represents a classical induction/reaction system.
Experiments done by the author (Marthy, 1970 b) suggest, however, that Arnold’s conclusions are not necessarily adequate to explain his very precise observations. We have been able to show, by means of transplantation experiments, that the eye rudiment in stage VII has a ‘differentiation tendency’ (Raven, 1938) and the capacity of autonomous retina differentiation. Thus, determination or ‘canalization’ (Waddington, 1940) towards specific organ formation appears to be irreversible. The explanted rudiment of the eye vesicle fully integrates at its new location, as auto- or allograft at the same stage, without losing any of its elements either by mechanical or other disturbances during the transplantation procedure or by an influence of the yolk syncytium on to which it is grafted (which would be, to say to the least, conceivable according to Arnold’s hypothesis).
As for the donor region, it is interesting to note that the yolk syncytium, following Arnold’s concept, would not yet have entirely lost its induction activity: the material of the ‘outer layer of cells’ growing over it forms one normal-sized eye rudiment (Arnold, 1965b) or two smaller-sized eye rudiments (Marthy, 1970a). This means that the formerly neighbouring material, which should be supposed to be determined (not for retina formation) as irreversibly as the explanted material, is in fact labile. In other words, we are facing a true regulation phenomenon. The question to be answered is then: Is the peripheral material capable of replacing the removed material by partial autonomous reversion of former determination (regulation) or does it only furnish cell material of which the former determination is reversed by the inductive action of the underlying yolk syncytium?
We have carried out our investigations in this line by a series of experiments at the earlier stages of morphogenesis when the primary eye vesicle is formed (stages VI –IX); the yolk syncytium was either left intact or damaged or removed. In particular, the morphogenetic processes were analysed: (1) in the explantation zone after complete or incomplete removal of the eye rudiment, with the yolk syncytium left intact in all cases; (2) in the heterotopic auto- or allograft grafted on to intact yolk syncytium; (3) in the prospective eye vesicle in situ after separation from the yolk syncytium or partial removal or destruction of the latter; (4) in the prospective eye vesicle in situ after incision.
We should like to anticipate the conclusion to be drawn from our experiments: An inductive influence of the yolk syncytium does not exist at stage VI or later. Instead, an ‘eye region’ in the ‘outer layer of cells’ can be defined on either side of the embryo: this region is definitely larger than Arnold’s ‘placode’ (1971, p. 297), the complete removal of this complex leads to the lack of the entire eye, although the yolk syncytium remains intact. If this complex is, however, incompletely removed, the remaining parts regulate autonomously. The morphological appearance of this regulation (single eye or duplication) depends on the stage at which the operation is performed. The process of duplication revealed the presence of ‘contractile elements’ - their important role in the shaping of the optic vesicle was demonstrated by a simple experiment in which only an incision was made in the otherwise intact rudiment.
The eye rudiment is particularly favourable for this type of experiment because of its large size (stage VII: 0·7/0·25 mm) and its later separation into a pigmented part (vesicle) and unpigmented peripheral organs, which facilitate evaluation of the various effects of operations. A discussion of Arnold’s results and conclusions is suitable as his investigations were also focused on this organ complex.
MATERIALS AND METHODS
The eggs of Loligo vulgaris used in the experiments were obtained by bottom trawling off the coast of Catalonia (Western Mediterranean). The egg strings were kept in running sea water or in large receptacles with well-aerated sea water. For preparation, operations and culture we used the method described earlier (Marthy, 1970b) with one modification: the agar was sterilized (autoclave) and stored in 50 ml flasks and was re-liquified and distributed in Petri dishes immediately before an experiment. After removal of the innermost egg case (chorion), embryos were prepared for operation by several rinses in sterile sea water. In order to diminish the pressure of the yolk mass, the yolk envelope was perforated at the vegetative pole, with the result that a large quantity of yolk leaked out and formed a good attachment for the embryo, preventing its rolling over. Also, the loss of yolk pressure facilitates removal of tissue without impeding normal development. The translucency of the agar allows operating with illumination from below. Platinum wire (0·1 mm) and ‘Wolfram’ wire were used for explantation and incision. The tip of the ‘Wolfram’ wire was sharpened in an electrolyte. Grafts were implanted on to the denuded, intact yolk syncytium in the stomodeal region or in the ventral midline between the arm rudiments. In stages VI and VII the embryo was turned over the graft for 3 –4 h in order to facilitate integration of the graft by gravity. In stages VIII and IX the embryos were immobilized by means of glass bridges for several hours. In general, ciliar activity of the embryo detaches the surface cells from the glass, thus avoiding damaging of tissue by removal of the glasses.
Operations were made at room temperature (21 –24°C) in sterile sea water to which penicillin was added (50 –100 i.u./ml). The culture dishes (6 cm in diameter, 1 cm deep) with an agar bottom (2 –3 mm thick) were kept at 18·5 –20°C. The operated embryos were transferred to new culture dishes with fresh sea water at 4-day intervals, since the lack of oxygen disturbs development, slowing down its rate and eventually leading to malformations (Marthy, 1970b). Embryos were in general fixed later than stage XVI in Halmi fixation solution. Throughout this paper we will use the developmental stages defined by Naef (1923), indicating the corresponding stages of Arnold (1965a), if necessary.
For a better understanding of the processes we have investigated it will be useful to give a brief description of the normal eye development (detailed studies by Faussek, 1900; Naef, 1928; Sacarrao, 1954; Arnold, 1967 a,b, 1971).
Around stage VI the early eye rudiment can be recognized on either side of the embryo as an ectodermal area, the ventral part of which subsequently becomes limited by an inconspicuous semilunar ridge (Fig. 1) that progresses dorsally so as to form a complete, closed ring of elongate shape by stage V1U. Meanwhile mitotic activity and rearrangement of nuclei lead to a thickening of the ectoderm within this ringfold. The latter grows over (like a closing curtain) the central part of the rudiment until the optic vesicle is completely closed at stage IX (Fig. 2). This primary optic vesicle, first a flattened pocket, acquires its definitive, nearly spherical shape in subsequent stages (Fig. 3). Histological differentiation includes thickening of the proximal part (retina) until stage IX, pigmentation and separation into two layers (within and beyond the limit of the membrana limitans exterior (von Lenhossék, 1894; Faussek, 1900)) of the retina from stage XI on. Starting at stage IX, the distal part of the eye vesicle shows differentiation into a Tentigenic area’ (Arnold, 1971) and a secondary annular fold which forms the iris (pigmentation from stage XI on). The large lentigenic cells are distinct before morphological separation of these two areas. The cornea arises from a prolongation of the bases of the arms two and four (from stage XII on); these dorsal and ventral folds of each side grow laterally, unite behind the eye vesicle and finally close at the base of the arm crown (stage XVIII).
Dorsal view of an embryo in stage VIII –IX. o, Almost closed primary optic vesicle; m, mouth (stomodeum).
It will be useful to divide the morphogenetic phase that is the particular interest of our investigation into the following steps:
Stage V –VL Ectodermal eye rudiment monolayered, not distinguishable in living embryos (Fig. 4); topographical extent not histologically defined.
Stage VI. Cells of eye rudiment slightly elongate, nuclei arranged on different levels. Strong mitotic activity. Limit of primordium histologically defined only in ventral part by an inconspicuous elevation. (About stage XVI of Arnold.)
Stage VI –VII. Peripheral elevation more distinct, extending dorsally to the midline of the primordium.
Stage VII. This structure encloses the entire primordium as an annular elevation, the ventral part of which becomes a distinct ridge.
Stage VII –VIII. Ridge transforms into a fold in the ventral area. Shaping of the ridge in the dorsal part. Detached cells lie between anterior part of the eye primordium and the yolk syncytium.
Stage VIII. Annular fold entirely surrounds prospective retina and starts growing over it (this is not an invagination!) (Fig. 5). Optic ganglion a flat band between anterior part of eye primordium and yolk syncytium.
Stage VIII –IX. Closure of the optic vesicle progressing.
Stage IX. Porus of the eye vesicle closed (Fig. 6). Histological differentiation of the outer wall into a central area of lens formation and peripheral area of iris formation. Retina thick, but not pigmented. Optic ganglion a compact mass attached to the retina. (Stage XXI of Arnold.)
The same embryo as in Fig. 5 in stage IX (14 h later at 19°C). Primary optic vesicle (o) closed, a, Arm; e, embryo; m, mantle; 5, outer yolk sac; st, stomodeum; y, yolk (leaks out).
RESULTS
1 Morphogenetic behaviour of the explantation zone, with intact yolk syncytium, after removal of eye material
Operations made at different stages were always identical: ‘subtotal’ (a), total (b) and partial removal (c, d, e) of the eye rudiment without damaging the underlying yolk syncytium. Wound closure took place without exception. According to the amount of material removed, morphogenetic behaviour of the explantation zone varied. The number in parentheses following the stage number gives the interpretable cases in the series of experiments.
(a) The ectodermal rudiment was removed including the annular elevation (stage VI: estimation of dorsal part):
1. Stage VI (8). Wound closure within 20 h. Subsequently, a normal eye vesicle was formed, which by stage IX could not be distinguished from a normal control eye. The adjacent optic ganglion had normal size. (Table 1: 5 →16).
Schematic representation of the experiments performed on the eye rudiment of the stages VI and VII.

2. Stage VI and VII (3). Wound closure within 15 –10 h. Further development resulted in formation of two optic vesicles, the common size of which was equal to a normal vesicle. (Table 1: 5 (Fig. 7) →17 (Fig. 8).)
Dorsal view of the head region of an embryo in stage XII. On one side two optic vesicles are formed.
3. Stage VII –VIII (2). A single, small optic vesicle was formed (Fig. 9). Duplication never occurred in the optic ganglion; it was of normal size and structure, both in the case of vesicle duplication and with a single small vesicle.
Ventral view of the head region of an embryo in stage XIV –XV. On one side a small optic vesicle is formed.
4. Stage VIII and later (6). Complete lack of optic vesicle. Growth and development of the optic ganglion, however, were not influenced by the operation.
(b) The ectodermal rudiment was removed together with a circular area surrounding the annular elevation (on the site of its appearance):
1. Stage VI(3). Wound closure after more than 24 h. No eye rudiment was formed (Table 1: 11 →18). Optic ganglion was smaller than the normal size (Fig. 10). (This may be due to excessive removal of cell material, because of the difficulty of locating exactly the limit of explantation.)
Ventral view of an embryo in stage XIV–XV. On one side lack of the optic vesicle. Optic ganglion reduced in size, a, Arm; g, optic ganglion; m, mantle; o, optic vesicle; 5, outer yolk sac.
2. Stage VII and later (3). Wound closure after about 20 h. No eye rudiment was formed. The optic ganglion was of normal size (this may be due to more precise explantation made possible by the greater extent of the annular elevation). (Table 1:11 (Fig. 11) →18 (Fig. 12).)
Dorsolateral view of an embryo in stage VII with the eye rudiment removed together with surrounding tissue; yolk syncytium left intact.
Ventral view of an embryo in stage XIV–XV. On one side lack of the optic vesicle. Optic ganglion normal sized.
(c) Only parts of the ectodermal rudiment were removed:
1. Stage VI (2). Wound closure according to extent of explantation, not later than 15 h. A normal eye vesicle and a normal optic ganglion were formed (Table 1: 6 →16).
2. Stages VII and VII-VIII (f). Identical effect after removal of dorsal or ventral part of the eye rudiment respectively. Wound closure within 10 h. A complete optic vesicle was formed and reached normal size (Table 1: 6 (Fig. 13) →16).
Ventral view of an embryo in stage XV. o, Optic vesicle formed from a half of the eye rudiment.
3. Stage VIII –IX (1). Same effect, except for formation of iris fold, which was inhibited.
(d) The ventral part of the ectodermal rudiment was only partially removed leaving the ventral tip intact:
1. Stages VII (1) and VIII (1). Wound closure within 10 h. Two eye vesicles were formed, the ventral one smaller in size than the dorsal one, the latter smaller than normal size (Table 1: 8→17). Both were surrounded by a common iris fold.
2. Stage VIII–IX (2). Formation of two vesicles with differentiation of lens, retina and pigmentation; lack of iris fold.
(e) The anterior or posterior part of the ectodermal rudiment was removed:
1. Stage VII (3). Formation of normal-sized eye vesicle and iris fold.
2. Stages VII–VIII to VIII–IX (3). Formation of normal-sized eye vesicle; lack of iris fold.
2 The development of the eye primordium when grafted as a heterotopic auto- or allograft on to denuded, intact yolk syncytium
The differentiation capacity of the graft depends on the stage at which transplantation is done:
1. Stage VI (3). The eye primordium grafted on to embryos at stage VII (thus obtaining shorter time of wound closure) differentiates into a compact retina, which becomes pigmented at later stages. No further differentiation into the typical retina layers. No peripheral thickening (which marks beginning of vesicle formation). (Table 1: 5a→10.)
2. Stage VII(5). Explanted eye primordium is incorporated into the surrounding tissue and differentiates into a compact retina with pigmentation occurring later. In contrast to our earlier description (Marthy, 1970a), differentiation into lentigenic cells occurred in 3 of 5 cases; these cells (Fig. 14) were located in the peripheral elevation. Further thickening at the periphery did not occur again.
Marginal region of an eye rudiment grafted in stage VII, now on stage XII. r, Retina cells; c, lentigenic cells, × 140.
3. Stage VIII (ringfold distinct) (2). Incorporation of the graft and formation of an optic vesicle within a few hours, with normal structural and histological differentiation. No formation of iris fold.
4. Stage IX (2). The already closed optic vesicle is incorporated and undergoes normal differentiation including spherical shape (Fig. 15).
3 The development of the eye primordium in situ after operation of the underlying yolk syncytium
Embryos at stage VI and at stage VII (period of reconstitution ability) were used in these experiments in which the eye primordium was left intact, its normal contact with the underlying yolk syncytium being interrupted. These experiments are complementary to the others in which, on the contrary, the eye primordium was operated on and the yolk syncytium left intact.
(a) The primordium is cut out except for a small connexion and lifted off the yolk syncytium (Table 1: 1). Under the yolk pressure circumference of the wound increases. After a few minutes the embryo is turned over so as to bring the eye rudiment into its original position. Wound closure is completed after a few hours. This operation does not influence normal eye development except for retardation which is apparent until stage XVI (Table 1: 16).
(b) Damaging (4), partial removal (2) or complete destruction (5) of the yolk syncytium after lifting up the eye primordium has the same effect (Table 1: 2, 3, 4→16). The loss of yolk more or less markedly influences the wound closure (wound contracts beyond the circumference of the primordium) and leads to some deformation of the eye rudiment, the general development of which is normal except for retardation as observed in the above experiment.
These experiments again demonstrate that the eye primordium from stage VI on carries all the information necessary for further development that is independent of the yolk syncytium. We will look in more detail into this phenomenon by analysing the results of point (a) 2 and (d) in the Results section 1.
4 Splitting of the eye primordium (yolk syncytium left intact or damaged)
(a) Splitting horizontally (division into a dorsal and a ventral part).
1. Stage VI (3). The dorsal and the ventral part of the eye rudiment divided by a simple horizontal incision ‘fuse’ within a few hours after operation. One normal optic vesicle was formed.
2. Stage VI–VI I and earlier stage VII (2). Formation of a complete optic vesicle showing a slight constriction of the posterior wall at the site of the incision (Fig. 16).
3. Stage VII–VIII (5). Constriction more marked in the posterior part becoming visible in the anterior part. The beginning of this process can be recognized about 10 min after operation.
4. Stage VIII (6). Strong constriction along the line of incision (Fig. 17) resulting in complete division into dorsal and ventral part, each developing into an independent optic vesicle with a well-defined lens (Fig. 18). In four cases the lenses of the two vesicles were connected to each other (Fig. 19). The iris fold is always common to both vesicles, but does not form normally, however, if the operation is done later than V11I-1X.
Splitted eye rudiment of an embryo in stage VIII (division into a dorsal and a ventral half) 15 min after operation. → Direction of the contracting forces along the wound edges.
Lateral view on the head region of an embryo in stage XVIII. Duplication on one side after incision in stage VIII of the eye rudiment.
Eye vesicle. The two lenses (l) of the two primary optic vesicles are connected. Aspect at hatching stage, r, Retina, typically layered; c, cornea; i, iris, × 87·5.
5. Stage IX (2). Splitting of the now closed eye vesicles does not result in constriction and separation. Wound closure leads to complete reconstitution of the eye vesicle; lens and iris fold form normally. Lack of pigmentation in the retina up to hatching stage marks the site of incision.
(b) Splitting vertically (division into a posterior and anterior part):
1. Stage VI (1). No effect.
2. Stage VII (1). Constriction of the retina on the dorsal side.
3. Stage VII–VIII and later (5). Dorsal constriction, no separation into two separate optic vesicles. The absence of the process in the earlier experiment (splitting horizontally) is apparently related to the elongate shape of the eye rudiment.
DISCUSSION
The findings given in Results, §§ 1–3, lead us to the conclusion that the yolk syncytium, at stage VI, does not induce the formation of the eye rudiment and, consequently, that the eye rudiment situated in the ‘outer layer of cells’ does not have a ‘reactive power’ (Raven, 1938). At stage VI the ‘outer layer of cells’ of the embryo has two areas that carry all factors necessary for eye formation, providing the capacity of self-organization and autonomous differentiation which manifests itself, under experimental conditions, in regulation. For an area having these characteristics, we may use the term ‘field’ as defined by Nieuwkoop (1967a, b).
We do not yet know how the field is established. The question is, whether certain cells are programmed to form ‘eye’ or whether originally undetermined cells receive, prior to stage VI, a morphogenetic programme from the yolk syncytium. Our recent results from experiments prior to stage VI are in favour of the former interpretation (Marthy, 1972).
At present, our experiments demonstrate that an eye field does exist at stage VI. The extent of this field is shown by the experiments in the Results section 1 a and b: It is considerably larger than the morphologically recognizable area of the future retina and the annular elevation surrounding it. We therefore think that Arnold’s results (1965 b, p. 75: no yolk syncytium explanted) were due to incomplete removal corresponding to our experiments (a) 1, 2, 3 in Results §1 (Table 1:5).
As to the explant (cf. Table 1: 5a→9 and 12a →15), we may add here that the nutritive conditions in Arnold’s experiments (1965b): presence or absence of yolk syncytium including yolk material!) suggest that the difference observed in further development may be due to the presence or absence of nutritive material, respectively. This aspect was apparently not taken into account by Arnold, who stated that yolk material did adhere to the explanted eye primordium (1965b, pp. 73, 75), but did not mention this fact in his summary, where the supposed role of the yolk syncytium alone is defined (1965b), p. 77).
Denying an inductive function of the yolk syncytium does not mean that a morphogenetic action is not conceivable on a mechanical basis. We think that Arnold’s results (1965b), p. 75: yolk syncytium explanted), which are fully confirmed by our own experiments (Table 1: 12, 13, 14), can be interpreted as the morphogenetic ‘response’ to the excessive disturbance of the material involved. It appears that wound closure has to proceed in an ‘organized’ way; the yolk syncytium serving as a ‘stage’ of rearrangement.
In addition to this, we note that the arrangement of the cells of a rudiment is mechanically guided by elements that are integrated in the cell complex. Our incision experiments (no removal of cell material) give an idea of the dynamics of the spatial differentiation of the vesicle. The constrictions observed after incision show the contracting force exerted by (supposedly cellular) elements. It is uncertain at present whether the entire cell complex or a few specialized elements take part in this process. The formation of two optic vesicles gives the most marked evidence of this process which intensifies from stage VI-VI I onward. The degree of morphological aberrance reflects the ‘competition’ between wound closure and organogenetic contraction forces. The combined effect of these forces from stage VI-VII on in all organogenetic areas spread out over the yolk results first in a separation of the organ rudiments and later, from stage VIII (Naef, 1928, p. 267) on, in an increased contraction of the entire embryo ‘assembling’ the organs to form a compact organism.
ACKNOWLEDGEMENTS
I thank Dr Katharina Mangold for her encouragement during the investigations presented here and for her help with preparing the manuscript. I am also very much indebted to Dr Sigurd von Boletzky for valuable discussions and his help during preparation of the manuscript. Mrs Leslie Rowe also kindly read the manuscript. The present investigation was partially supported by a grant of the Swiss National Fund for the Advancement of Scientific Research.