ABSTRACT
The cephalic, wing, haltere, leg and median genital discs of D. tryoni develop in the embryo shortly after dorsal closure and involution of the head. The labial and lateral genital discs first appear in the mid-second larval instar.
The wing and haltere discs arise as outgrowths of the ventral meso- and metathoracic discs. The remaining discs arise in the positions generally accorded to them by previous authors.
Each pair of discs originates from cells in a definite location in the blastoderm.
The relationship between the blastodermal locations of the presumptive discs and the segmental pattern of the larval ectoderm suggests that imaginai disc development in Cyclorrhapha resembles regeneration of a total field by a part.
INTRODUCTION
The cephalic, labial, wing, haltere, leg and genital imaginal discs of Cyclorrhapha are histologically distinct in the late embryo or early larva (Weismann, 1864; Pratt, 1900; Auerbach, 1936; Kaliss, 1939; Poulson, 1950). Experimental investigation of Drosophila melanogaster further suggests that the determined rudiments of the discs are present in the blastoderm of the early embryo (e.g. Geigy, 1932; Howland and Child, 1935; Howland, 1941; Gloor, 1947). Almost no attempt has been made, however, to describe the embryonic origins of the discs or to follow their development before hatching. Investigation of this matter in Dacus tryoni (Frogg) (Diptera, Trypetidae) has clarified the developmental relationship between the cyclorrhaphan larva and adult.
MATERIALS AND METHODS
The origin and development of imaginal discs in the embryo of D. tryoni was followed from serial sections prepared in connection with a previous paper (Anderson, 1962). Living embryos were also further examined by the method given in that account.
RESULTS
The early development of the imaginal discs
In D. tryoni, the cephalic, wing, haltere, leg and median genital discs become histologically distinct between 24 and 32 hours after oviposition, during the em-bryonic phase of dorsal closure, completion of involution of the head and early histodifferentiation of the larval organs (Anderson, 1962).
The cephalic (eye-antenna) discs are first discernible at 32 hours as a pair of small outgrowths at the postero-lateral corners of the frontal sac (Text-fig. 9C). During the remaining 10 hours before hatching they increase slowly in length by cell division, forming narrow, blind-ending tubes with regular epithelial walls (Text-fig. 1A, B).
A, sagittal section through cephalic disc of newly hatched larva. B, transverse section through cephalic disc of newly hatched larva.
The prothoracic leg discs first appear between 24 and 26 hours as paired ventro-lateral thickenings of the prothoracic hypodermis near the edges of the ventral region from which the nerve cord has separated (Text-fig. 2A). Between 26 hours and hatching they show some increase in cell number but retain their superficial position (Text-fig. 3A).
A, transverse section through prothoracic disc of 26-hour embryo. B, transverse section throughmesothoracic leg disc of 26-hour embryo. C, transverse section through wing disc of 26-hour embryo. D, diagrammatic reconstruction of right meso- and metathoracic discs of 26-hour embryo. E, sagittal section through median genital disc of 28-hour embryo. F, sagittal section through median genital disc of 30-hour embryo.
A, transverse section through prothoracic disc of 26-hour embryo. B, transverse section throughmesothoracic leg disc of 26-hour embryo. C, transverse section through wing disc of 26-hour embryo. D, diagrammatic reconstruction of right meso- and metathoracic discs of 26-hour embryo. E, sagittal section through median genital disc of 28-hour embryo. F, sagittal section through median genital disc of 30-hour embryo.
Newly-hatched larva. A, transverse section through prothoracic disc. B, transverse section through mesothoracic leg disc. C, transverse section through metathoracic leg disc. D, transverse section through wing disc. E. transverse section through haltere disc. F, transverse section through genital disc.
Newly-hatched larva. A, transverse section through prothoracic disc. B, transverse section through mesothoracic leg disc. C, transverse section through metathoracic leg disc. D, transverse section through wing disc. E. transverse section through haltere disc. F, transverse section through genital disc.
The mesothoracic discs also become visible between 24 and 26 hours as ventro-lateral thickenings in the hypodermis of the mesothoracic segment. Unlike the prothoracic discs, however, they undergo rapid cell division and growth, each producing a long stalk penetrating postero-dorsally into the haemocoele. By 26 hours the tip of the stalk has swollen to form the first rudiment of the wing disc of its side (Text-fig. 2C, D). At the proximal end of the stalk the remainder of the thickening forms the rudiment of the mesothoracic leg disc (Text-fig. 2B, D). A single rudiment thus produces the wing and leg discs on each side of the mesothorax.
The metathoracic discs develop in a similar manner. Originating ventrolaterally in the hypodermis of the metathorax between 24 and 26 hours, they produce stalks whose tips swell to form the haltere disc rudiments while the proximal ends merge into metathoracic leg discs (Text-fig. 2D).
Like the prothoracic leg discs, the meso- and metathoracic leg discs remain superficial until hatching, showing only a slight increase in cell number (Text-fig. 3B, C). The wing and haltere discs become more distinct as the stalks on which they are borne lengthen and become thinner and both pairs of discs become attached to tracheal branches in their respective segments. They grow slightly between 26 hours and hatching and a lumen develops in the centre of each (Text-fig. 3D, E).
The median genital disc arises between 26 and 28 hours as a median évagination of the anterior wall of the distal end of the proctodaeum (Text-fig. 2E). By 30 hours the évagination is cut off and lies closely apposed to the proctodaeal wall just in front of the anus (Text-fig. 2F). Subsequent growth before hatching is slight, the disc increasing in width but not in its other dimensions (Text-fig. 3F).
The labial discs and the small lateral discs associated with the median genital disc do not begin to invaginate until the middle of the second larval instar. The details of their early development will be discussed in a subsequent paper. The labial discs arise from cells in the floor of the cephalopharynx on either side of the median salivary gland opening (Text-fig. 9C). The lateral genital discs are formed by cells in the ventro-lateral hypodermis of the eighth abdominal segment, on either side and just in front of the anus.
The blastodermal origins of the imaginal discs
By taking account of the movements of parts proceeding during gastrulation, elongation and shortening of the germ band, dorsal closure and involution of the head, it is possible to determine the blastodermal locations of the cells from which the imaginal discs arise. As a preliminary, it is necessary to trace the presumptive plan of ectodermal segmentation in the blastoderm.
The presumptive embryonic ectoderm is distributed in the blastoderm as two broad lateral bands of cells extending from the anterior almost to the posterior end (Anderson, 1962; see also Text-fig. 4A). At the beginning of gastrulation, each band is divided by a temporary cephalic furrow (described by Anderson, 1962) into an anterior and a posterior portion. During gastrulation, the anterior portion spreads down to the ventral mid-line without being displaced anteriorly or posteriorly (Text-fig. 4B). The posterior portion also moves down to the ventral mid-line, but at the same time elongates, its anterior end remaining stationary while its posterior end moves forward along the dorsal surface (Text-fig. 4B).
A, presumptive ectoderm in the blastoderm of D. tryoni, lateral view. B, distribution of ectoderm at the end of gastrulation, lateral view.
While the ectodermal bands are thus extended, ectodermal intersegmental grooves begin to form. The distribution of the grooves relative to the line of the cephalic furrow shows that the ectoderm derived from the anterior portions of the presumptive ectodermal bands is divided between the head, mandibular segment and maxillary segment (Text-fig. 5A), while that derived from the posterior portions is incorporated successively in equal portions into the 12 post-maxillary segments (Text-figs. 5B-D ; 6A). The presumptive plan of this subdivision can be superimposed on the map of the larval blastodermal presumptive areas as shown in Text-fig. 7.
Embryo, lateral view. A, 16 hours. B, 18 hours. C, 20 hours. D, 22 hours.
It now becomes possible to assign blastodermal locations to the cells from which the imaginal discs arise.
The cephalic discs can be traced to their blastodermal origins by following the formation of the frontal sac. During dorsal closure and involution of the head, the dorsal edges of the ectoderm of the head spread upwards to meet in the dorsal mid-line (Text-figs. 5A-C; 8A, B). At the same time, the dorsal edges of the prothoracic hypodermis join mid-dor sally and the prothoracic segment as a whole moves forward to meet the posterior edge of the head ectoderm. (Text-figs. 5 A-D; 6A-C; 8A-C). Without further movement of the dorsal head ectoderm, the antero-dorsal edge of the prothoracic hypodermis folds forward over the dorsal surface of the head, forming the frontal sac, while the remainder of the pro-thoracic segment continues to move forward as involution of the mouth-part segments and the anterior and lateral parts of the head ectoderm proceeds (Text-figs. 6D, E; 8C, D; 9A-C). The ventral wall of the frontal sac is thus formed by the dorsal ectoderm of the head, the dorsal wall by prothoracic hypodermis, as in Drosophila melanogaster (Tsai, 1961). The cells which give rise to the cephalic discs lie postero-laterally at the edge of the ventral wall (Text-fig. 9C), as des-cendants of cells at the postero-dorsal edges of the presumptive ectoderm of the head (Text-figs. 7 ; 8A-D).
Embryo, lateral view. A, 23 hours, B, 24 hours. C, 25 hours. D, anterior end, 26 hours. E, anterior end, 28 hours.
Presumptive plan of ectodermal segmentation and location of presumptive imaginal discs in the blastoderm of D. tryoni, lateral view.
The second pair of discs in the head, the labial discs, arises from the presumptive ectoderm of the labial segment, just dorsal to the presumptive salivary gland plates (Text-fig. 7). This ectoderm remains adjacent to the salivary gland openings throughout involution of the head (Text-fig. 8A-D). When the labial segment inturns to form the anterior part of the cephalopharyngeal floor (Text-figs. 6A-E ; 9A-C) and the paired salivary gland openings fuse in the mid-line, it then lies on either side of the median opening (Text-fig. 9C). Cells within it later form the labial discs.
A, anterior end of 16-hour embryo, diagrammatic lateral view. B, anterior end of 22-hour embryo, diagrammatic lateral view. C, anterior end of 25-hour embryo, diagram-matic lateral view. D, anterior end of 26-hour embryo, diagrammatic lateral view. E, posterior end of blastoderm, diagrammatic lateral view. F, posterior end of 25-hour embryo, diagram-matic lateral view. G, diagrammatic transverse section through blastoderm at level of pro-thorax. H, diagrammatic transverse section through 8|-hour embryo at level of prothorax. I, diagrammatic transverse section through 12-hour embryo at level of prothorax. J, diagram-matic transverse section through prothorax of 26-hour embryo.
A, anterior end of 16-hour embryo, diagrammatic lateral view. B, anterior end of 22-hour embryo, diagrammatic lateral view. C, anterior end of 25-hour embryo, diagram-matic lateral view. D, anterior end of 26-hour embryo, diagrammatic lateral view. E, posterior end of blastoderm, diagrammatic lateral view. F, posterior end of 25-hour embryo, diagram-matic lateral view. G, diagrammatic transverse section through blastoderm at level of pro-thorax. H, diagrammatic transverse section through 8|-hour embryo at level of prothorax. I, diagrammatic transverse section through 12-hour embryo at level of prothorax. J, diagram-matic transverse section through prothorax of 26-hour embryo.
A, diagrammatic sagittal section through anterior end of 24-hour embryo. B, diagrammatic sagittal section through anterior end of 26-hour embryo. C, diagrammatical sagittal section through anterior end of 33-hour embryo.
In each segment, the thoracic disc and surrounding hypodermis of each side develops directly from the presumptive ectoderm bordering the presumptive nerve cord (Text-fig. 8G-J). The blastodermal antecedents of the thoracic disc cells thus lie approximately in the centre of this presumptive ectodermal region (Text-fig. 7).
The cells of the region from which the median genital disc arises move down to it during gastrulation from either side, originating from blastoderm cells at the postero-ventral corners of the presumptive ectoderm of the eighth abdominal segment and the antero-ventral corners of the presumptive proctodaeum (Text-figs. 7 ; 8E, F). The antecedent cells of the median genital disc thus have a bilateral origin, later reflected in the bilateral form of the developing disc.
The cells of the regions from which the lateral genital discs arise can be similarly traced from blastoderm cells of the presumptive ectoderm of the eighth abdominal segment, antero-dorsal to the presumptive median genital disc regions (Text-figs. 7; 8E, FJ.
DISCUSSION
Imaginai discs in cyclorrhaphan embryos
In D. tryoni, with the exception of the labial and lateral genital discs, the imaginal discs develop in the embryo shortly after dorsal closure and involution of the head. Save for Pratt’s (1900) description of the embryonic origins of the discs in Melophagus ovinus and Poulson’s (1959) description of theformation of the labial discs in the embryo of Drosophila melanogaster, other workers have failed to trace the discs before hatching of the larva. The present work, however, confirms conclusions drawn on the basis of the initial larval positions of the discs in Calliphora vomitoria, Melophagus ovinus and Drosophila melanogaster, that the cephalic discs arise at the postero-lateral corners of the frontal sac, the labial discs adjacent to the median salivary gland opening, the median genital disc in the mid-line of the anterior wall of the anus, the lateral genital discs in the hypodermis adjacent to the anus and the thoracic leg discs ventro-laterally in the hypodermis of the thoracic segments (Pratt, 1900; Auerbach, 1936; Kaliss, 1939; Poulson, 1950; Bodenstein, 1950). At the same time, it shows that the evidence presented by Pratt (1900) in support of these conclusions is erroneous in several respects. Although his description of the origins of the leg discs in Melophagus ovinus is probably correct, according with the condition in D. tryoni, it is plain that in describing the initial rudiments of the cephalic discs as two deeply invaginated sacs in front of the mouth he confused disc formation with frontal sac formation. The origin of the cephalic discs in D. tryoni from two small cell clusters at the posterior end of the frontal sac is more probably typical of Cyclorrhapha. Pratt’s median labial disc at the anterior end of the ventral wall of the cephalopharynx is also incorrectly interpreted, since by comparison with D. tryoni it can be seen to be the thickened anterior end of the wall itself. The bilateral origin of cyclorrhaphan labial discs from cells on either side of the median salivary opening described by Poulson (1950) for the embryo of Drosophila melanogaster is confirmed by the present work, although the labial discs in D. tryoni do not appear in section until the middle of the second instar. Lastly, the invaginated median genital disc described by Pratt is probably the infolded anterior anal lip. A median origin of the disc from a small cell cluster within the anterior anal lip, as in D. tryoni, is more likely to be the typical condition.
Previous workers (Pratt, 1900; Auerbach, 1936; Kaliss, 1939; Poulson, 1950; Bodenstein, 1950) have further assumed that the wing and haltere discs of Cyclo-rrhapha develop as invaginated dorso-lateral hypodermal thickenings in their respective segments. The assumption is based exclusively on the evidence of Pratt (1900) that such thickenings are present in the embryo of Melophagus ovinus. The description by Pratt of the further development of the thickenings is unconvincing and their identification as wing and haltere discs is open to question. It seems more probable that the outgrowth of wing and haltere discs in the embryo of D. tryoni from the ventral meso- and metathoracic discs is typical and that only a single pair of imaginal discs is initially present in each thoracic segment. The dorsal prothoracic pair described by Weismann (1864), Pratt (1900) and Auerbach (1936) in Calliphora vomitoria, Melophagus ovinus and Drosophila melanogaster give rise in D. tryoni to the anterior larval spiracles, not to imaginal structures.
The blastodermal origins of the discs
In D. tryoni, each pair of imaginal discs originates from cells lying in a definite presumptive ectodermal region of the blastoderm. In primitive insects the corresponding regions, head ectoderm, mouth-part ectoderm, segmental thoracic ectoderm and posterior abdominal ectoderm, give rise to the adult structures formed in D. tryoni by the discs. For example, the mesothoracic ectoderm primitively gives rise to the adult mesothoracic body wall, legs and wings, which in D. tryoni arise from a single pair of mesothoracic discs. Similar considerations apply to the cephalic, labial, prothoracic and metathoracic discs, and to the genital discs if the eighth abdominal segment is regarded, not as a true segment, but as a condensed posterior end of the abdomen. If, as seems likely, the primitive development of the ectoderm of each of the above regions expresses the potentialities of a regional ectodermal field, it appears that each disc has evolved as a group of cells retaining the capacity to establish the whole field, while the remaining cells of the region have become modified to develop as larval structures. Development of cyclorrhaphan adult hypodermal structures from imaginal discs may thus fall into the same causal category as regeneration of a total field by a part and, as such, is not as distinct from the primitive direct development of insects as seems at first sight.
RÉSUMÉ
Embryologie de Dacus tryoni. 2. Développement des disques imaginaux chez l’embryon
Les disques imaginaux de la tête, de l’aile, du balancier et des pattes, et le disque génital médian, se forment chez l’embryon peu après la fermeture dorsale et 1’involution de la tête. Les disques labial et génitaux latérals apparaissent d’abord au milieu de la seconde intermue larvaire.
Les disques de l’aile et du balancier naissent sous forme d’excroissances des disques ventraux méso-et métathoraciques. Les disques restants se forment aux emplacements que leur ont généralement attribués les auteurs précédents.
Chaque paire de disques naît de cellules ayant une localisation définie dans le blastoderme.
Les rapports entre la localisation blastodermique des disques présomptifs et la structure segmentaire de l’ectoderme larvaire fait penser que le développement des disques imaginaux chez les Cyclorrhaphes ressemble à la régénération d’un champ complet par une de ses parties.
ACKNOWLEDGEMENT
This work was supported by a research grant from the University of Sydney.