ABSTRACT
The imaginal segmental abdominal hypodermis, salivary glands, fore-, mid-and hind-gut, lateral cerebral ganglia and segmental tracheal system each arise from a group of blastoderm cells situated within the blastodermal presumptive area of the corresponding larval structure.
The cyclorrhaphan blastoderm can be zoned into presumptive areas for all major parts of the body. Each area includes presumptive larval cells and also presumptive cells of the imaginal part which replaces the larval part at metamorphosis. Except for mesodermal structures, all major imaginal parts and their larval equivalents are probably determined in the blastoderm. The theory is developed that one of the fundamental steps in cyclorrhaphan evolution has been separation of the blastodermal rudiments of the major parts of the hypodermal body wall and gut into determined larval and adult components. In addition, it is suggested, the larval components have acquired a capacity to evócate differentiation of mesodermal structures independently of the adult components.
INTRODUCTION
In addition to the principal imaginal discs (cephalic, labial, wing, haltere, leg and genital), the larva in Cyclorrhapha carries as discrete components the rudiments of the imaginal segmental abdominal hypodermis, salivary glands, fore-, mid-and hind-gut, lateral cerebral ganglia and segmental tracheal system (Snodgrass, 1924; Bodenstein, 1950; Shatoury, 1956b; Anderson, 1964, etc.). The embryonic origins of these rudiments have never been firmly established. The present paper describes their origins in the embryo and larva of Dacus tryoni (Frogg.) (Try-petidae), taking a further step in the elucidation of the developmental relationship between the cyclorrhaphan larva and adult.
MATERIAL AND METHODS
The embryonic origins of the rudiments in question were traced in serial sections of embryos and larvae prepared by the methods described by Anderson (1962a, 1963b). The reader is referred to these papers for details of the methods employed.
RESULTS
The abdominal hypodermal rudiments
The larval hypodermis of each abdominal segment in D. tryoni develops from a corresponding segmental pair of ectodermal presumptive areas in the blastoderm (Anderson, 1962a, 1963b). In giving rise to hypodermis, each area undergoes cell division and spread, without radical cell rearrangement or invasion by cells from other sources. In the early larva, the resulting abdominal hypodermis is histologically uniform and grows by increase in cell size. At the middle of the 2nd larval instar, hypodermal cells at four localized sites in each of the first seven segments of the abdomen begin to divide. The sites are paired, dorso-laterally, and ventro-laterally, and equidistant from the ends of the segment. At each, a patch of small cells is formed which contrast markedly in size with the surrounding larval hypodermal cells (Text-fig. 1A). The four patches are the rudiments of the imaginal hypodermis of the segment. The cells from which they arise stem directly from cells occupying the same relative sites in the abdominal segmental ectodermal presumptive areas of the blastoderm. The approximate location of these blastodermal presumptive imaginal abdominal hypodermal rudiments is shown in Text-fig. 3.
A, transverse section through left ventral imaginal hypodermal rudiment of 2nd abdominal segment of 53-hr. larva (mid-2nd instar). B, frontal section through junction of salivary gland and duct of 33-hr. embryo. C, imaginal tracheal rudiment in transverse section through 3rd abdominal segment of 53-hr. larva. D, sagittal section through junction of stomodaeum and mid-gut of 26–hr. embryo.
A, transverse section through left ventral imaginal hypodermal rudiment of 2nd abdominal segment of 53-hr. larva (mid-2nd instar). B, frontal section through junction of salivary gland and duct of 33-hr. embryo. C, imaginal tracheal rudiment in transverse section through 3rd abdominal segment of 53-hr. larva. D, sagittal section through junction of stomodaeum and mid-gut of 26–hr. embryo.
The salivary gland rudiments
In the embryo of D. tryoni, ventro-lateral salivary gland plates differentiated in the ectoderm of the labial segment invaginate about 18 hr. after oviposition to form uniform blind-ending tubes. Within the next 3 hr., the tubes develop into larval salivary glands and ducts (Anderson, 1962a). As the gland and duct of each side differentiate, cells at the junction of gland and duct remain small and undifferentiated (Text-fig. IB). The ring of small cells so formed is the rudiment of the corresponding imaginal salivary gland. The precursor cells of the rudiment are cells of the original salivary gland plate, directly descended from cells of the blastodermal presumptive salivary gland area and located in the blastoderm in the position shown in Text-fig. 3.
The fore-gut rudiment
In the embryo of D. tryoni, the stomodaeum arises by ingrowth from a blasto-dermal stomodaeal presumptive area and gives rise to the larval oesophagus and the central core and reflexed intermediate cell layer of the larval proventriculus. The outer epithelium of the proventriculus is formed by the anterior end of the mid-gut (Anderson, 1962a). Between 26 and 28 hr. after oviposition, after dorsal closure of the embryo but before penetration and reflexion of the end of the stomodaeum within the mid-gut, a terminal ring of small cells is differentiated in the stomodaeal wall at its junction with the mid-gut (Text-fig. ID). Between 28 and 30 hr., as penetration and reflexion occur and the larval pro ventricular structure is established, the ring comes to occupy the junction between the proventricular intermediate cell layer and outer epithelium. Here it forms the rudiment of the imaginal fore-gut (Text-fig. 2A). The precursor cells of the rudiment are cells of the blastodermal stomodaeal presumptive area (Text-fig. 3).
A, frontal section through proventriculus of 33-hr. embryo. B, sagittal section through junction of proctodaeum and mid-gut of 30-hr. embryo. C, transverse section through mid-gut wall of 40-hr. embryo. D, transverse section through mid-gut wall of newly-hatched larva. E, parasagittal section through right cerebral hemisphere of 7–10 hr. larva.
A, frontal section through proventriculus of 33-hr. embryo. B, sagittal section through junction of proctodaeum and mid-gut of 30-hr. embryo. C, transverse section through mid-gut wall of 40-hr. embryo. D, transverse section through mid-gut wall of newly-hatched larva. E, parasagittal section through right cerebral hemisphere of 7–10 hr. larva.
The hind-gut rudiment
The imaginal hind-gut rudiment in D. tryoni also becomes histologically distinct in the embryo, as a ring of small cells differentiated in the wall of the inner end of the proctodaeum, adjacent to the bases of the Malpighian tubules. It first becomes conspicuous between 28 and 30 hr. after oviposition, when the proctodaeum differentiates into larval hind-gut (Text-fig. 2B). The proctodaeum arises by ingrowth from a blastodermal proctodaeal presumptive area (Anderson, 1962a). The precursor cells of the imaginal hind-gut rudiment are cells of this area (Text-fig. 3).
The mid-gut rudiment
The larval mid-gut of D. tryoni originates from paired anterior and posterior mid-gut strands, formed by immigration and proliferation of blastodermal anterior and posterior mid-gut presumptive areas. During dorsal closure, the paired strands spread to enclose the yolk sac as a cuboidal epithelium (Anderson, 1962a). The cuboidal condition persists during the major part of mid-gut morphogenesis and a uniform epithelium and yolk-filled lumen are still present 40 hr. after oviposition (Text-fig. 2C). During the final period of rapid utilization of the yolk before hatching at about 42 hr., the majority of the mid-gut epithelial cells enlarge, but some persist as scattered small cells at the base of the general epithelium (Text-fig. 2D). Subsequent events show that the small cells are the rudiment of the imaginal mid-gut epithelium. They stem directly from cells of the anterior and posterior mid-gut presumptive areas (Text-fig.3).
The cerebral ganglionic rudiments
The larval central nervous system of D. tryoni arises from neuroblasts whose presumptive distribution in the blastoderm is shown in Text-fig. 3. At hatching, it comprises the typical paired cerebral hemispheres and ventral nerve cord of a cyclorrhaphan maggot, with a cellular cortex overlying a fibrous core (Anderson, 1962a). Shortly after hatching an outgrowth of the fibrous core in each hemisphere penetrates the cortex postero-laterally, breaks through the surface and turns forward to grow towards its corresponding cephalic disc as an optic nerve (Anderson 1963b). Between 7 and 10 hr. after hatching, early in the 1st larval instar, cortical cells around the root of each optic nerve enlarge and give rise to the rudiments of an outer and inner patch of cuboidal epithelium, closely adjacent to each other near the surface of the cortex (Text-fig. 2E). From these subsequently arise the massive lateral optic ganglionic systems of the imaginal brain. The origin of the rudiments is from cells descended from blastodermal cells of the cerebral ganglionic presumptive areas, located approximately as shown in Text-fig. 3.
Larval and imaginal presumptive areas in the blastoderm of D. tryoni, diagrammatic lateral view. Larval areas are labelled above the diagram, adult areas, together with the presumptive mesoderm and ventral nerve cord, below the diagram.
The tracheal rudiments
The larval tracheal system of D. tryoni originates as a paired series of segmental invaginations of the lateral ectoderm of the three thoracic and first seven abdominal segments, the cells of the rudiments being presumptively distributed in the blastoderm as shown in Text-fig. 3 (Anderson, 1962a, 1963a). As in Drosophila melanogaster (Robertson, 1936; Bodenstein, 1950), the imaginal tracheal rudiments first become histologically distinct in the larva as paired segmental clusters of small dividing cells in the epithelium of the segmental branches of the larval system. Two pairs occur in each thoracic segment and one pair in each of the first seven abdominal segments. Like the imaginal abdominal hypodermal rudiments, they begin to show cell multiplication at the middle of the 2nd larval instar (Text-fig.1C). The precursor cells of the imaginal tracheal rudiments are cells of the larval tracheal epithelium, originating from blastodermal cells of the segmental tracheal presumptive areas (Text-fig.3.)
DISCUSSION
Although the blastoderm in D. tryoni is histologically uniform, as in all Cyclorrhapha, the descriptive evidence presented here and in previous papers (Anderson, 1962a, 1963a) allows it to be zoned into the presumptive areas shown in Text-fig.3. These areas have still to be subjected to detailed experimental analysis of the state of determination of their cells. However, a considerable body of evidence from Cyclorrhapha supports the view that the larval areas are determined very early, probably by oöplasmic segregation in the egg cortex (Hathaway & Selman, 1961; Anderson, 1962b). Regulation within the subsequent products of an area may occur following experimental disturbance (Nitschmann, 1958; Counce, 1961), but no regulative switch of fate of cells of one area to that of another appears to be possible. There is also evidence that preliminary determination of imaginal areas in Cyclorrhapha has occurred by the blastoderm stage (Geigy, 1932; Howland & Child, 1935; Howland, 1941; Henke & Maas, 1946; Gloor, 1947; Counce, 1961), even though regulation within the subsequent products of the areas can easily be experimentally produced (Ursprung, 1963). It is tempting, therefore, to regard the areas shown in Text-fig. 3 as already determined. The determinative differences between them remain obscure, but the map expresses well the degree of complexity existing in the cyclorrhaphan blastoderm.
If blastodermal mosaicism can be accepted, the map also provides new insight into the developmental relationship in Cyclorrhapha between larva and adult. It shows that the initial step in the development of the cyclorrhaphan egg is the setting out of the rudiments of the major components of the animal’s body (preoral hypodermis, mouth-part hypodermis, segmental thoracic hypodermis, segmental abdominal hypodermis, posterior abdominal hypodermis, brain, ventral nerve cord, segmental tracheae, salivary glands, fore-gut (stomodaeal), mid-gut, hind-gut (proctodaeal), mesoderm and germ cells) as a series of blastodermal presumptive areas, arranged relative to one another approximately in the positions they will occupy in the future animal. Only the germ cell rudiment, in the form of a part of the complement of pole cells, is distant at this stage from its final relative position. The map further shows, however, that each area, with the exception of the mesodermal and neural areas and the pole cells, is already separated into cells of two types, those which will give rise to the adult product of the area and those which will give rise to its larval equivalent. A summary of this can be made as seen in Table 1.
Primitively, each of areas 1–12 would be expected to give rise more or less directly to the adult product. In Cyclorrhapha, it seems that selection has occurred, simultaneously for each area, for the transference of this potentiality to a sub-group of cells contained within the area, leaving the remainder free to develop along specialized, expendable larval lines. The implication of this is that in the evolution of the Cyclorrhapha, fundamental changes have taken place in the processes of determination in the very early embryo.
At the same time it can be seen that in respect of its hypodermal body wall and gut, to which imaginal areas 1–12 give rise, the development of the adult in Cyclorrhapha retains the general fines common to all insects, in spite of the apparent complexities of metamorphosis. The major parts of the adult hypodermal body wall (including the tracheal system) and gut (including the salivary glands) are established as rudiments in the blastoderm in the pattern typical of an insect germ band and develop directly through to their adult condition. The modifications they show during development can be regarded as adaptations which give space for the temporary development of the corresponding larval parts. Together with the reduction in relative size of the initial adult blastodermal rudiments, these adaptations permit the interpolation of extreme larval modification of the body wall and gut into a developmental sequence yielding as a final product a relatively unmodified adult insect.
It is well known, of course, that Cyclorrhapha also develop a specialized temporary larval mesodermal organization (musculature and fat body), later replaced at metamorphosis by a mesodermal organization more typical of an adult insect. In spite of this, it is not possible to make a distinction between larval and adult components in the mesodermal presumptive area of the blastoderm. The explanation of this probably lies in the fact that, in insects, the determination of embryonic mesoderm as individual structures often rests with evocation by the structures with which the mesoderm becomes associated (Counce, 1961). After gastrulation in Cyclorrhapha, the mesoderm becomes mesenchymatous and part of it then associates with and differentiates, apparently as a result of evocation, relative to already established gut wall and body wall rudiments (Poulson, 1950; Counce, 1961; Anderson, 1962a). In metamorphosis, a similar association of mesenchyme with newly established adult body wall and gut wall structures occurs (e.g., Bodenstein, 1950) and a similar evocator relationship can be postulated (e.g., Shatoury, 1956a). Although the origin of adult mesenchyme from embryonic mesoderm in Cyclorrhapha involves complications which require further study (e.g., Anderson, 1963b), it can be suggested with reasonable confidence that determination of mesoderm as a whole is probably established in the blastoderm, but that sub-determination of its larval and adult components occurs at separate later phases of development, as an evocated consequence of blastodermal sub-determination of separate larval and adult body wall and gut wall systems. It is the existence of this evocator relationship which seems to have permitted, in association with the evolutionary modification of the body wall and gut wall into larval and adult parts, the evolution in Cyclorrhapha of a specialized new larval musculature and the retention at the same time of an adult musculature whose constituents can be fitted into the basic system of insect muscle homologies.
The place of neural and germ cell development in this scheme needs little comment. The central nervous system passes, with modification but without large scale destruction and replacement, directly from larva to adult (e.g., Bodenstein, 1950). No sub-division of its presumptive areas into larval and adult components is therefore to be expected. The existence of an imaginal group of cells in the cerebral presumptive area on each side, later giving rise to the massive imaginal optic ganglia (Shatoury, 1956b; Anderson, 1964), is explainable as a case of retarded development of a structure which does not become functionally significant until the adult head, with its paired compound eyes, is well developed. The pole cell germ cells are an adult rudiment without a larval counterpart. As such, their development, with its migration from the posterior pole to paired sites in the fifth abdominal segment (Poulson, 1950; Anderson, 1962a) and subsequent multiplication and differentiation at these sites (Bodenstein, 1950), is independent of any considerations of larval functional organization and remains generally unmodified.
In general, therefore, it can be suggested that among the fundamental steps in cyclorrhaphan evolution have been separation of the blastodermal rudiments of the major parts of the body wall and gut each into a determined larval and adult component, involving major changes in the determinative processes in the early embryo, and the acquisition by the larval components of a capacity to evócate differentiation of mesodermal structures independently of the adult components.
RÉSUMÉ
Développement embryonnaire de Dacus tryoni (Diptera), 3. Les origines de ébauches imaginales à l’exclusion de les disques principaux
L’hypoderme imaginal des segments abdominaux, les glandes salivaires, l’intestin antérieur, moyen et postérieur, les ganglions cérébraux latéraux et le système trachéal segmentaire ont chacun pour origine un groupe de cellules blastodermiques situées à l’intérieur de l’aire présomptive de la structure larvaire correspondante.
Le blastoderme des cyclorrhaphes peut être subdivisé en zones présomptives de toutes les parties importantes du corps. Chacune de ces aires comprend des cellules larvaires présomptives et en outre des cellules présomptives de la région de l’imago qui remplace la région correspondante de la larve à la métamorphose. A l’exception des structures mésodermiques, toutes les parties importantes de l’imago et les parties larvaires correspondantes sont probablement déterminées dans le blastoderme. Ces résultats conduisent à l’hypothèse selon laquelle une des étapes fondamentales de l’évolution des cyclorrhaphes a été la séparation des ébauches blastodermiques en constituants déterminés de la larve et de l’adulte, en ce qui concerne les parties les plus importantes de la paroi hypodermique et de l’intestin. En outre, d’après l’auteur, les ébauches larvaires ont acquis le pouvoir d’évoquer la différenciation de structures mésodermiques indépendamment des ébauches de l’adulte.
ACKNOWLEDGEMENTS
This work was supported by a research grant from the University of Sydney. My thanks are due to Professor L. C. Birch for critical discussion of the manuscript and to Dr M. A. Bateman for the provision of material from his laboratory stocks.