The Early Embryonic Development of Rhodnius prolixus (Hemiptera, Heteroptera)

Bhodnius prolixus Stal is a blood-sucking bug of the family Reduviidae, and is a native of South America. A great deal of work has been done on the nymphal and adult stages (Buxton, 1930; Wigglesworth, 1931, 1933, 1934), and it has proved an admirable experimental animal. For this reason it seemed desirable to work out in detail its development within the egg.

A culture of Bhodnius was obtained from Professor P. A. Buxton, of the London School of Hygiene and Tropical Medicine, and was bred by the methods described in his paper (Buxton, 1930). The insects were fed at intervals on rabbit’s blood. The adults were kept at a constant temperature, 21° C., and the blotting-paper among which they lived was examined frequently for eggs. For the earlier stages the age of the eggs was known to within 1 hour. For later stages their ages were not quite as accurately known in all cases. The eggs were incubated under constant conditions of temperature and humidity for varying lengths of time from 1 hour up to 5 days. This was considered very important, for it is well known that temperature and humidity greatly affect the rate of development. Previous workers have not incubated their eggs under controlled conditions, with the result that they have been unable to produce accurate time-tables of development. Even under conditions of constant temperature and humidity there was slight individual variation in the time taken to develop to any given stage (see Table 1, p. 5).

The eggs were kept at 21° C. and 90 per cent, relative humidity; the duration of the embryonic period of controls under these conditions was approximately 29 days. There was a slight variation in the time taken to hatch, but it was never longer than 30 days. Eighty-five per cent, of the eggs hatched successfully. The remainder all developed, but the nymphs died in the process of emerging. This is known to happen with other bugs, for instance with Cimex lectularius. The temperature chosen for incubation was one at which development proceeded very slowly. The humidity was high since Rhodnius eggs are known to dry up readily.

Of all the eggs examined by means of sections (over 130), only one was found which had failed to develop.

The eggs of Rhodnius vary somewhat in size and shape. They are elongate oval, with a cap at the anterior end (Textfig. 1). They are 1·5 mm. by 0·8 mm. approximately. The dorsal surface is slightly flattened, while the ventral side is convex. The cap slopes towards the dorsal side. As the eggs were fixed with their chorions intact (except in some cases for removal of the cap), the yolk solidified in the shape of the egg. This was of great assistance in section cutting, as it was possible to tell from the sections which side was dorsal and which ventral, and whether the orientation was correct or not. This fact was of importance when following the stages in blastokinesis.

The chorion of the egg is sculptured; in life it is hard and fairly brittle. The eggs are pale pink; the pigment is contained in the yolk and it shows through the transparent chorion.

The eggs were fixed in Carnoy, Bouin, or hot alcoholic Bouin. The first fixative gave the best results. When using Bouin the caps of the eggs were removed with a needle to ensure penetration of the fixative. After the eggs had been kept in 90 per cent, alcohol for about 3 days the chorion was dissected off with needles. The alcohol made the chorion sufficiently brittle for it to be easily removed without damaging the underlying yolk and embryo. Slight shrinkage of the yolk facilitated this process. Successive stages in the development of the embryonic rudiment were drawn as a whole while the eggs were in 90 per cent, alcohol; staining of the whole mounts was found unnecessary because the pale embryo showed up against the pigmented yolk. Eggs for sectioning were double embedded in celloidin and wax by Newth’s method.

Sections were cut 8μ thick and were stained with a variety of stains. Ehrlich’s haematoxylin and eosin were mainly used. Alcoholic haematoxylin, Giemsa, Leishman, safranin, Küll’s, and Volkonsky’s staining methods were also tried.

The earliest stages before cleavage took place were extremely difficult to stain. In the vast majority of sections no nuclei were visible. This may be due to the breaking up of the nuclei on fixation. Various bodies, possibly consisting of chromatin, appeared scattered among the cytoplasm surrounding the yolk-spheres.

In all the later stages the cytology appeared to be reasonably good. The method of double embedding usually ensured the easy sectioning of the yolk.

The structure of the egg half an hour after being laid was that of a typical insect egg-cell. A single nucleus with its surrounding cytoplasm was situated at the centre of a large quantity of yolk. Strands of cytoplasm radiated out between the yolkglobules and connected the central cytoplasm with a peripheral layer, which lay immediately underneath a very thin vitelline membrane. The yolk during life was liquid and appeared to be composed of a large number of spheres. After fixation the spheres were still apparent in the earliest stages, and were separated from one another by a very thin layer of cytoplasm. At all stages the yolk stained very readily with any of the methods used. After 2 hours’ incubation a nucleus and cytoplasm were found to have moved from their central position and to be nearer the surface. Possibly this was the egg nucleus about to undergo maturation. I was not able to find any complete nuclei in eggs of between 2 and 11 hours old. A cytoplasmic network was visible, and some bodies which stained like chromatin were seen scattered through the cytoplasm. A combination of several fixatives and staining methods gave very similar results. Possibly the nuclei at this stage are very fragile and easily broken up by fixatives. In eggs more than 12 hours old the cleavage nuclei were visible, and they steadily increased in number as time went on.

After 12-13 hours of incubation the first cleavage occurred. At 18 hours there were between 4 and 8 cleavage nuclei (Textfig. 2). By 25 hours there were at least 32 nuclei. These early cleavage nuclei lay towards the anterior end of the egg, and they had not moved far from the centre of the yolk. When the number of nuclei reached 32, they had begun to migrate to the periphery, fig. 1., Pl. 5. As in most other insects, the whole egg is a syncytium at this stage and remains such until the formation of the blastoderm is completed. The cleavage nuclei were rounded and lay in an irregular mass of cytoplasm which was continuous with a thin layer of cytoplasm round the yolkspheres (fig. 1., Pl. 5). Their position in the egg was similar to those of Pieris at the same stage (Eastham, 1927), but they never showed the conspicuous comet-like appearance as shown by the nuclei of that species. The chromatin in the nuclei of Bhodnius does not stain deeply until blastoderm formation has begun.

There was no cleavage of the yolk in Bhodnius. The structure of the yolk changed as development proceeded. At first it was a mass of small spheres of 40 μ, in diameter. Later, when the cleavage nuclei had reached the periphery and there was no longer a layer of cytoplasm round the yolk-globules, these seemed to run together, but their continuity was broken by large spaces. These spaces were presumably due to some ether soluble substance having been dissolved out during the technique of celloidin embedding.

Not all the cleavage nuclei migrated to the periphery. Some remained behind and gave rise to the yolk-nuclei (fig. 4, Pl. 5). Those which migrated moved through the yolk with their surrounding cytoplasm and came to lie in the peripheral layer of cytoplasm underneath the vitelline membrane. At first they were situated at a considerable distance from one another though they were connected by the peripheral cytoplasmic layer (fig. 2, Pl. 5). Soon the nuclei divided tangentially, so reducing the space between each nucleus (figs. 2 and 3, Plate 5). As a result of the first few cleavage nuclei having been formed in the centre of the anterior half of the yolk, it was in the anterior part of the egg that the nuclei first reached the periphery. It was several hours later before they reached the posterior surface. In this Rhodnius resembled Pieris and the beetle Europe terminalis (Paterson, 1931), but differed from the holly tortrix moth (Huie, 1917), and the beetle Hydrophilus (Heider, 1885).

After 30 hours’ incubation a large number of the cleavage nuclei had migrated to the periphery of the egg and had taken up their position near the surface. The formation of the blastoderm was very similar to that described for Pieris (Eastham, 1927) and for the beetle Europe (Paterson, 1931). The peripheral cleavage nuclei increased in number by tangential divisions. The cytoplasm belonging to adjacent nuclei became continuous, forming a syncytial layer (fig. 3, Pl. 5). The external surface of the egg remained smooth, except at the posterior end where the nuclei protruded beyond the surface of the egg as they do in some insects (fig. 6, Pl. 5) (Eastham, 1927; Marshall and Dernhehl, 1905; Nelson, 1915). The vitelline membrane was not distinct at this stage. The syncytial layer was very thin and gave the appearance of being stretched over the egg surface. Cell-walls dividing each nucleus and its cytoplasm from its neighbours began to appear after 50 hours. They were only distinct in the parts of the blastoderm destined to form the embryonic rudiment. The remaining, extraembryonic, blastoderm became clearly separated from the yolk, but the limits of the cells were difficult to distinguish. The nuclei in this part became flattened and elongated very early, giving the characteristic appearance of serosa cells.

The blastoderm was complete anteriorly before it was formed at the posterior end of the egg. Differentiation of the blastoderm began on the ventral side of the egg. The ventral and lateral portions of the blastoderm became differentiated into cubical cells, while the dorsal and lateral parts remained covered by flattened epithelium (Text-fig. 4). The cubical cells gave rise to the embryonic rudiment, while the flattened cells became the extra-embryonic blastoderm.

The yolk-nuclei were derived from cleavage nuclei which remained behind instead of migrating to the edge of the yolk. I have not found another example of an exopterygote insect in which the origin of the vitellophages was definitely stated to be from cleavage nuclei (fig. 6, Pl. 5). Some nuclei migrate nearly to the periphery before lagging behind the others. The vitellophages were large and conspicuous at first, and they stained more deeply with haematoxylin than did cleavage nuclei of the same age. They multiplied by mitosis up to the age of 2 days. After that age no evidence of mitosis was seen, though it was not uncommon to find yolk-nuclei in pairs or in threes, giving the impression that they might have recently divided. After formation of the embryonic rudiment, large numbers of cells were given off into the yolk. There they gradually disintegrated.

The differentiation of the blastoderm into embryonic rudiment occurred on the ventral side of the egg. The blastoderm covering the egg ventrally, laterally, and posteriorly, was transformed into cubical epithelium; the remainder of the blastoderm was made of very flattened epithelium. The cells of the cubical epithelium or ventral plate became thickened along two ventro-lateral areas (Text-figs. 3 and 4). These areas were separated at the anterior end of the yolk by a median portion of ventral plate which was made of smaller, less cubical, cells with more elongated nuclei. These cells were not as flattened as those forming the extra-embryonic blastoderm. The ventro-lateral areas of cubical epithelium ran parallel until they reached the posterior end of the egg. Here they turned towards one another and met. The total effect was that of a U with the open end facing anteriorly, the cells inside the U being smaller. The cubical epithelium then extended its area round the posterior end on to the dorsal surface (Text-fig. 5). This was very similar to the first appearance of the embryonic rudiment in the bug, Pyrrhocoris apterus (Seidel, 1924). This stage could be seen in whole mounts because the parallel areas of cubical epithelium, being made of large cells, showed up white against the pinkish colour of the underlying yolk.

Longitudinal sections of this stage (Text-fig. 5) showed a number of small cells with very conspicuous nuclei situated at the posterior pole of the egg. These cells were budded off from the blastoderm and were continuous with it (fig. 8, Pl. 5). I assume that these are the germ-cells of Rhodnius, because germ-cells, cytologically similar, have been found in the same position in other insects (Imms, 1925 ; Hirschler, 1909). Poluszynski (1911) found the same arrangement in a Coccid bug, but apparently the germ-cells in Pyrrhocoris apterus (Seidel, 1924) were not differentiated at this early stage. It is interesting to observe that Pyrrhocoris does not resemble Rhodnius although the two species belong to the same suborder of the Hemiptera.

In Rhodnius the germ-cells made their first appearance between 66 and 76 hours after the eggs were laid. They increased in number by further immigration and by occasional mitosis until they formed a mass of cells projecting inwards into the yolk.

The area of the egg covered by the embryonic blastoderm now began to decrease. The thickened ventro-lateral portions became withdrawn towards the mid-line. The result of this was that the embryonic cells took on a more columnar form, and the cells in the mid-ventral part became columnar too, though they had previously been flattened.

A few hours later invagination of the germ-band began. At a point near the dorsal side of the egg, where the superficial embryonic rudiment curved round the posterior pole, there began a proliferation of cells inwards into the yolk. An intucking of the embryo followed at this point, and proceeded very rapidly (Text-figs. 6 and 7). The intucking took place just dorsally to the mass of germ-cells.

The invagination caused the ventral part of the embryonic rudiment to move towards the posterior end of the egg, since the posterior end of the rudiment was becoming tucked into the yolk. The cells of the extra-embryonic blastoderm became more flattened as this process took place, as they had to cover an increased area of yolk, and their number did not increase. The process of invagination continued until all but the extreme head end of the embryo was surrounded by yolk. The head-end (which was by this time at the posterior pole of the egg) remained in its superficial position for a short time; then it too became surrounded by yolk (Text-fig. 8 A), but it remained flexed towards the ventral side of the egg. The part of the embryonic rudiment other than the head lay very much nearer the dorsal surface of the egg than the ventral.

When the embryo had taken up this new position it had moved through 180° and was completely reversed in its relation to the chorion. It lay in a dorsal position with its head facing the posterior end of the egg and with its posterior extremity facing the anterior end. The original ventral surface of the embryonic rudiment was now dorsally placed. This marked the end of the first stage in blastokinesis, which is a characteristic of exopterygote insect development when there is much yolk

The invaginated portion elongated very quickly towards the anterior end of the egg, at the same time becoming very narrow and nearly cylindrical in cross-section (Text-fig. 8). The germcells were carried forward through the yolk, always occupying a position near the extreme posterior end of the embryo (Text-fig. 8). Invagination and narrowing of the embryonic rudiment took place synchronously, transforming the broad superficial rudiment into a long thin germ-band, almost all of which was sunk into the yolk. The anterior end of the germband remained superficially placed longer than the rest of the rudiment. At first it covered quite a large area (Text-fig. 9 A), but soon its edges contracted (Text-fig. 9 B), causing this part of the rudiment to thicken, and by 90 hours it also had become sunk into the yolk (Text-fig. 8). At this stage, therefore, there was a marked cephalic flexure. This flexure remained until 12 days of incubation. There was no caudal flexure in Rhodnius like that which occurs in Pyrrhocoris, or the bedbug Cimex lectularius (Heymons, 1899).

The method of invagination is shown in Text-fig. 6, and Text-figs. 6 B, c, and D show transverse sections taken at three different levels during the beginning of blastokinesis. There was active division of the cells of the embryonic rudiment during this process. Text-fig 6 c shows the position of the germcells in transverse section. They lay below the layer of cells from which the rest of the embryo was to be developed, and they were next to the main part of the yolk. The invaginated part of the embryo was at first a simple sac made out of cubical epithelium (Text-fig. 6 c). The sac was flattened in a dorso-ventral plane, and its internal cavity was the future amniotic cavity. The dorsal portion of cubical epithelium was the amnion, while that which was nearest the main part of the yolk was the ventral plate or embryonic rudiment. This invaginated ventral plate was made of much more columnar cells than the original cubical epithelium of the embryonic blastoderm. The cells were now much more tightly packed together, and this caused the germ-band to be thicker but much narrower than before. At first the cells forming the amnion were the same shape and size as those of the embryonic rudiment (Textfig. 6 c); some time after invagination was completed the amnion cells became elongated and thin anteriorly, while those of the ventral plate grew larger (fig. 9, Pl. 5). The meeting of the amnion folds and the broadening of the embryo in the future head and thoracic regions caused the amnion to be stretched and pulled out until it was a layer of much flattened cells.

The serosa was derived from the cells of the extra-embryonic blastoderm (Text-figs. 5 and 8). After the germ-band had invaginated and sunk completely into the yolk, the serosa formed an uninterrupted outer layer to the yolk. The serosa was a very thin cellular layer, and its nuclei were extremely flattened. The first formation of the amnion has already been described (p. 83). It was made of blastoderm which was pulled in with the germ-band when blastokinesis occurred. When the head sank into the yolk, the amnion was completed, but it remained connected with the serosa for some time (Text-figs. 8 and 10). The germ-band later lay free inside the yolk with the amnion covering its ventral side. All connexion with the surface was lost.

Mesoderm in insects may arise in three ways (Imms, 1925): (i) by invagination of the central part of the embryonic rudiment to form a tube; (ii) by overgrowth of the central part by two lateral portions; or (iii) by the proliferation of cells from the ventral plate along the mid-ventral line. In Rhodnius the mesoderm arises by a combination of the first two methods.

While the germ-band was in its superficial position (Text-fig. 7), the anterior part of the embryonic plate had become divided into a middle and two lateral plates (Text-fig. 6 D). From the middle plate the lower layer was differentiated by the formation of a groove or gastral furrow which developed in the midline (Text-fig. 8 c); this was overgrown by the lateral portions of the ventral plate, the gastral groove being obliterated by the process of overgrowth (Text-fig. 8 B, c, and D). This process took place immediately after the first stage in blastokinesis was completed.

The gastral groove first began just behind the anterior end of the embryo and proceeded from before backwards. A single longitudinal section gave all stages in lower layer formation, from the completed lower layer at the anterior end to the differentiation of the middle plate at the posterior end. During this process cells from the middle plate were given off into the yolk all along the germ-band (Text-fig. 8 A)

After the lower layer was completed, the germ-band became surrounded at its head-end by the amnion, the amnion and serosa having been continuous previous to this stage (Textfig. 8). The growth of the amnion and the serosa pulled the flexed head end of the embryo into a more superficial position at the hind-end of the egg (Text-fig. 10). The embryo was now similar to that of Pyrrhocoris apterus (Seidel, 1924), except that it was dorsally situated in the yolk, had no caudal flexure, and the germ-cells were already differentiated.

The formation of the endoderm in insects has been the subject of much controversy, and the various interpretations of its origin have been summarized by Eastham (1930). In Bhodnius the endoderm was formed by proliferation from both an anterior and a posterior area of the lower layer. This is somewhat similar to the mode of origin of the endoderm in a number of other insects, for instance Apis (Nelson, 1915), Calliphora (Noack, 1901), and Pieris (Eastham, 1927). In Pyrrhocoris Seidel found that the mid-gut was formed from proliferating areas at either end of the lower layer. In Bhodnius the anterior endoderm rudiment arose slightly earlier than the posterior. Its position was in the centre of the region of the cervical flexure. Here active proliferation of the cells in the mid-ventral line resulted in a complete disturbance of the regular arrangement of cells of the ventral plate (Text-fig. 10). Many of the cells of the proliferation were given off into the yolk. The others spread out over the ventral surface of the embryo. The cervical flexure made the anterior endoderm rudiment difficult to study in transverse sections.

The posterior endoderm rudiment arose slightly later, near the extreme end of the embryo (Text-fig. 10). It was made by a proliferation of the lower layer, which took place between the lower layer and the germ-cells, but extended farther towards the anterior end of the embryo than did the germ-cells. A large number of these proliferated cells were also given off into the yolk. This posterior rudiment was convenient for examination in transverse sections, there being no caudal flexure in Bhodnius.

In a recent paper by Mansour (1934) further evidence is brought forward in support of the view that the mid-gut has an ectodermal origin in some insects. This paper deals with the development of the adult mid-gut in a large number of Coleoptera, and it is stated that here the adult mid-gut is developed from ectodermal cells of the larva. Mansour claims that this supports his view regarding the embryological ectodermal origin of the mid-gut in the beetle Calandra (Mansour, 1927).

At the time when the endodermal proliferations were formed in Rhodnius there were no stomodaeal or proctodaeal invaginations. The fate of the endodermal proliferations, and their connexion with the ectoderm of the foreand hind-gut, will be discussed when the organogeny of Rhodnius is described.

About the time that the endoderm began to be formed the head and thorax underwent segmentation, the lower layer becoming constricted between the segments. This segmentation was visible in whole mounts of the embryo (Text-fig. 9 B). The abdominal region remained quite unsegmented at this stage. Certain of the ectodermal cells on either side of the mid-line had become differentiated by this time into specially large conspicuous cells. These were the neuroblast cells, which later formed the ventral nerve-cord (fig. 9, Pl. 5). A few hours later the paired appendages developed (Text-fig. 11).

The cells which were given off into the yolk from the endoderm gradually disintegrated, like those of the mesoderm (see p. 84 above). Presumably their function was to render the yolk substance more easily assimilable. The amount of yolk in Rhodnius is very large compared with the amount of embryonic tissue, and this may account for the very large number of embryonic cells given off into the yolk during all stages of the early development.

In conclusion I should like to thank Professor D. M. S. Watson (in whose department this work was done) and Professor L. E. S. Eastham for reading the manuscript of this paper and for giving me many helpful suggestions.

1. Eggs of Rhodnius prolixus were incubated at constant temperature and humidity (21° C. and 90 per cent, relative humidity). Eighty-five per cent, was the lowest record of the controls hatched successfully under these conditions.

2. The processes of maturation and fertilization were not studied.

3. Cleavage begins 12-13 hours after incubation. At 25 hours there are 32 nuclei. Yolk-cells are derived from cleavage nuclei, and they multiply by mitosis up to 50 hours. Blastoderm formation is complete after 55-60 hours of incubation.

4. The ventral embryonic rudiment is similar to that of many other insects. As soon as it is formed, germ-cells are budded off at the posterior pole of the egg.

5. The first stage in blastokinesis is fully described.

6. The formation of the mesoderm is by invagination and overgrowth.

7. The endoderm arises from two proliferating areas situated anteriorly and posteriorly.

8. Numerous cells are given off into the yolk during the early development of the embryo. There they disintegrate.

Abbreviations used for Plate

Am., amnion; AZ., blastoderm; C.N., cleavage nuclei; Gy., cytoplasm; Div.nuc., nucleus undergoing mitosis; Ect., ectoderm; E.R., embryonic rudiment; Ex.E.Bl., extra-embryonic blastoderm; G.G., germ-cells; L.L., lower layer; Near., neuroblast cell; Awe., nucleus; P.O., peripheral layer of cytoplasm; Ser., serosa; S.Y., space in yolk; Y., yolk; Y.N., yolk-nucleus (vitellophage).

Fig. 1.—Cleavage nuclei at 16-32 stage after 24 hours of incubation. ×375.

Fig. 2.—Early blastoderm formation. Cleavage nuclei have reached peripheral layer of cytoplasm. One nucleus shows tangential division. ×930.

Fig. 3.—Blastoderm formation just later than Fig. 2. × 930.

Fig. 4.—Syncytial layer after 50 hours of incubation, in region which will form embryonic blastoderm. × 930.

Fig. 5.—Syncytial layer after 50 hours of incubation, in region which will form extra-embryonic blastoderm. × 930.

Fig. 6.—Syncytial layer after 50 hours of incubation, posterior end of egg. × 620.

Fig. 7.—Junction of embryonic rudiment with extra-embryonic blastoderm. × 620. After 76 hours of incubation.

Fig. 8.—Longitudinal section. Embryonic rudiment posterior pole showing origin of germ-cells after 76 hours of incubation. × 620.

Fig. 9.—Transverse section embryo. Thoracic region after about 110 hours of incubation. × 250. Shows the neuroblast cells.