ABSTRACT
An understanding of the physiological and structural development of an embryo requires a knowledge of the precise state that development has reached. Without this knowledge it is not possible to compare findings from experiments performed on different embryos or to relate them to structural changes during development.
In previous work on Acridid embryos two methods have been used in estimating the stage of development. Some authors have defined embryonic stages by their age from the time of oviposition (Slifer, 1932; Roonwal, 1936, 1937; Hong, 1968), while others have described development in a series of arbitrarily chosen stages, which are based primarily on changes in the external morphology (Steele, 1941; Jhingran, 1947; Mathée, 1951; Shulov & Pener, 1959, 1963).
The eggs of Schistocerca gregaria are laid in pods of 30–90. Pods were collected and kept under constant conditions in order to investigate the development of the embryonic muscle (Tyrer, 1968, 1969). During this work it was found that there was considerable variation in the differentiation of eggs from different pods of the same age, even though most of the eggs in any one pod were at the same stage. This is reflected by the length of the development period of the eggs, which at a temperature of 28 °C, ranged from 13 to 20 days with a mean of 15·25 days (Fig. 1). Similar variations have also been observed by other workers on Acridid embryos (Slifer, 1932; Shulov & Pener, 1959, 1963). Because of these large variations, time alone cannot be used to define the stage of development accurately, especially in the later stages where the effects of variation will be more marked.
This difficulty can be avoided by defining development in a series of stages based on the changes in external morphology, but this has the disadvantage that internal changes may take place without changes in external features, and may therefore occur in the undefined period between stages. Shulov & Pener’s series for S. gregaria (1963) defines the early stages in detail and it is these which are the main concern of the embryologist, interested in the origin and formation of organs. The physiologist, however, is concerned primarily with the function of organs which develops towards the end of embryonic life. In this period most of the external changes are slight and often qualitative (e.g. progressive pigmentation), and Shulov & Pener give few stages which are difficult to identify precisely.
Rather than the discontinuous steps of a morphological series, investigation of organ function requires a continuous scale of development such as that provided by timing, but one which can be adjusted to the differing rates of development. A method has been devised, therefore, which enables any point in development to be defined accurately, by expressing the age of an embryo as a percentage of the development time of the rest of the eggs in the pod. The validity of the method has been checked by correlating external changes in the later stages with percentage development calculated from accurate measurements of the incubation period. From the data obtained it has been possible also to define stages for the later part of development which can be used when the more accurate method is not required, and which are more precise than those given by Shulov & Pener (1963).
METHODS USED TO MEASURE THE INCUBATION PERIOD
A large stock of Schistocerca gregaria is kept in the Zoology Department, Cambridge, using methods similar to those of the Anti-Locust Research Centre (Hunter-Jones, 1961). 50 –60 adults were kept in a standard, glass-fronted culture cage 18” × 18” × 20”, in a constant temperature room maintained at 28 ± 1 °C and a relative humidity of 35–40 %.
To record the time of oviposition
The pods of eggs of Schistocerca are normally laid in moist soil. The tubes of wet sand usually provided for oviposition (Hunter-Jones, 1961) were replaced by a trough filled with wet sand. This was drawn on a trolley below the culture cage at a rate of
by a geared-down electric meter motor fixed to the cross-bar between the rails at the end of the trolley’s run (Fig. 2). The locust had access to it via a
diameter hole cut in the floor of the cage. According to the position of the pod in the trough the time of oviposition could be determined to within 2 or 3 h. The females seemed undisturbed by having to lay in a narrow, moving trough. Initially a normal, stationary tube was provided in addition to the trough, and it was found that pods were laid in both receptacles with equal frequency.
The trough was removed once every 48 h, and the individual egg pods were transferred with some of the sand surrounding them to small covered glass jars on which the time and date of oviposition was written. The jars were placed in polythene bags to prevent evaporation from the wet sand.
To determine the time of hatching
A single pod of eggs was placed in apparatus which recorded the emergence of the first hopper as it crossed a light beam (Fig. 3). The resulting change in light intensity on a photocell switched on a transistor operating a relay which actuated a lever writing on a smoked drum. The lid of the jar containing the eggs was removed and replaced by a cover incorporating a short, vertical tube in diameter, which allowed the newly hatched hopper to enter the recording apparatus. This consisted of a Tufnol tube,
in diameter, mounted vertically and leading to a collecting jar. Half-way up the tube a narrow horizontal beam of light from a 6 V, 0 · 3 A torch bulb was directed, by means of a wave-guide made from a specially shaped piece of Perspex, across the tube on to a photoconductive cell (Mullard ORP 12) mounted in the opposite wall. The whole apparatus was kept in an incubator maintained at 28 ±0 · 5 °C.
A disadvantage of this apparatus is that it does not give a record of the variation in hatching times of the eggs in the pod since the hatchlings tend to congregate around the light so that only the time that the first one emerged can be measured. This could perhaps be overcome by using a red light beam to which the hoppers are not attracted and by providing a white light in the collecting jar to attract them.
CORRELATION OF PERCENTAGE AGE WITH DEVELOPMENT
Expressing the age as a percentage of the development time is a precise definition of the stage that the embryo has reached only if two criteria are fulfilled. First, all the embryos in one pod must develop at the same rate, and second, the development of each embryo must proceed at a constant rate.
(1) Synchronous development in the pod
Salzen (1960) found in Locusta migratoria migratorioides that the eggs in any one pod were usually at the same stage of development. To check that this is also true for Schistocerca, five eggs were removed from each of 50 pods. The egg shell was dissected away under the binocular microscope and the external features of the embryos compared. Usually all five embryos were morphologically identical although occasionally there were some retarded individuals which were usually abnormal and were discarded. Papillon (1960) described pods laid by the ‘gregaria’ phase of S, gregaria containing a small percentage of’ solitaria’ forms which occurred in the lower third of the pod. She found that these eggs hatched ca. 20 h later than the ‘gregaria’ at 30 °C. Although this phenomenon was not observed here, differences in phase character were obviated by using eggs only from the upper two-thirds of the pod.
Further evidence that all the embryos in one pod develop at the same rate comes from measurement of the variation in their development time. Although the hatching recorder did not give a record of this variation (see above) this was checked in 12 pods by removing the hatchlings from the apparatus at hourly intervals and counting them (Table 1). These results show that 49% of the eggs hatch within 2 h of the first hatchling and 82% have hatched by 3 h.
(2) Constant rate of development of individuals
It is generally accepted that eggs of tropical locusts such as Schistocerca develop continuously (Lees, 1955; Shulov & Pener, 1961). Development is arrested only if the soil in which the eggs are laid dries up, and is resumed when the soil is moistened again (Shulov & Pener, 1963). In this study, particular care was taken to prevent drying of the eggs (see above) to avoid this. Even so, it is possible that the rate of development of an embryo could fluctuate during incubation, and that embryos from different pods could have different fluctuations in the rate of development. This was excluded by an analysis of the external features of embryos at the later stages of development.
Six egg pods between 9·5 days (227 h) and 13-5 days (323 h) old were chosen for a detailed analysis. They were kept in an incubator at 28 ± 0·5 °C. Five eggs were taken from each pod every day at exactly 24 h intervals. These were dissected in Hoyle Ringer (Hoyle, 1953) and the embryos examined alive under the binocular microscope. The changes which occurred in 16 characters were noted (Table 2), and the animals fixed in neutral formol saline. The remaining eggs in each pod were kept until they hatched. The time of hatching was recorded and the total length of the development period was calculated. This period was taken as 100 % and the percentage of this period that 24 h represented calculated for each pod of eggs. In this way the age of each embryo was calculated as a percentage of the total development time of the pod from which it had been removed. All the embryos were then arranged in order of their percentage age, together with their tables of developmental characters. The fixed material provided a means of direct comparison of each embryo with its neighbours in the series. Both the fixed animals and the tables of characters formed an orderly development sequence, despite the fact that they came from six pods whose incubation times varied from 14 · 0 days (337 h) to 16 · 6 days (400 h). The rate of development in any one pod, therefore, appears not to fluctuate, even though different pods develop at different rates. The major changes which occur during development are listed in Table 3. Some of these are also shown in drawings made from photographs of the fixed material (Figs. 4, 5) and of eggs in which the shell was made transparent by dissolving the chorion in 3% sodium hypochlorite solution (Fig. 6) (Slifer, 1945).
Development of the metathoracic leg. The age of the animals from which the leg was taken is expressed as a percentage calculated from the development time of the rest of the embryos in the pod.
Development of the mandible. The anterior view of the left mandible is shown. The age of the embryos from which the mandible was taken is expressed as a percentage calculated from the development time of the rest of the embryos in the pod.
Drawings from photographs of eggs staged on morphological criteria. The shell was made transparent by treating with 3% sodium hypochlorite (Slifer, 1945).
Drawings from photographs of eggs staged on morphological criteria. The shell was made transparent by treating with 3% sodium hypochlorite (Slifer, 1945).
DISCUSSION
Calculating the percentage development from measurement of the incubation period enables any point in development to be defined accurately. For some purposes, however, the method is more elaborate than necessary and the construction of special apparatus may not always be justified. Where precision is not required it is convenient to distinguish five morphological stages. These are easily identifiable and have approximately equal intervals of development between them. Drawings of these stages of development are shown in Fig. 6. Work on embryos staged by either of the two methods can be related since the percentage development of the morphological stages is defined.
The 60 % stage
The most important criterion for defining this stage is the small size of the head and legs in relation to the thorax and abdomen which have swollen to accommodate the yolk. The metathoracic femur is not as long as the body is wide, and the head is so small that the eyes protrude laterally and dorsally. The eye pigment is confined to a small orange crescent dorsally and the mandibles are not differentiated into teeth. The tissues are sufficiently transparent for the yolk to be visible through the legs. The range of development over which embryos display these characters was not determined accurately as it was found in physiological and behavioural investigations (Tyrer, 1968) that this was not a particularly important stage.
The 70% stage
The head is almost as large as in the hopper and the metathoracic femur is as long as, or slightly longer than, the body is wide. The tissue of the legs now makes it difficult to see the yolk through them, although this is still clearly visible through the body wall of the abdomen. The eye is two-thirds pigmented but the mandibles do not yet show differentiation into teeth. The range over which embryos show this combination of characters is small. The eye is less than half pigmented in the 67· 5% stage and by the 71· 2% stage the mandibles show a marked crinkling along the edge where the teeth will form.
The 80 % stage
The metathoracic femur is now half as long again as the body is wide and near the joint with the tibia a crescentic mark is just distinguishable on the cuticle of the femur. Soft unpigmented spines can be seen on the tibia if the embryonic cuticle is dissected off (see Fig. 4). The tips of the tarsi are just pigmented and pale pigment spots have appeared on the legs. The mandibular teeth are easily distinguishable. These characters are visible from the 77· 8% stage onwards, but an upper limit can be put on this stage when the pigment spots on the legs become as dark as, or darker than, the yolk. At this stage also (81-9%) the tibial spines are easily visible through the embryonic cuticle.
The 92 % stage
The stage between 90· 9 and 92· 8 % is the most easily defined. Black pigment begins to appear in the femur crescent at the 91 % stage and this region is fully pigmented by the 93 % stage. A very dark pigment spot also appears at the base of the cercus during the same period. The ratio of the metathoracic femur length to body width is nearly at the hatchling value of 1· 76. Pigmentation of the tibial spines and the tarsal claws is complete and that of the teeth almost complete. The head, thorax and legs are well pigmented and the tissues are opaque, so the yolk is no longer visible except in the dorsal mid-line where the heart lies. The intensity of pigmentation of the head, thorax and legs continues to increase up till hatching, but few pigment spots appear on the abdomen, the colour of which appears green to the naked eye.
The 100% stage
This stage is defined as the period between emergence from the egg and the full darkening of the cuticle, which normally lasts about 2 h. The light green colour of the newly emerged hopper darkens within an hour and by 2 h the animal is almost black.
Using these criteria it was found that the percentage development could be estimated usually to within 2 % of the development time (Tyrer, 1968), even in the early stages in which few samples were taken. Definitions of intermediate stages, however, are best obtained by accurate measurement of the age of the embryo and calculating this as a percentage of the development period as described above.
Shulov & Pener’s descriptions of their stages for the same period of development described here are given in Table 4 together with the equivalent percentage age which has been calculated from their data. From their figures the calculated percentage age of their XX stage is 67 · 2 % ± 2 · 8 %, of the XXI stage 78 · 4 % ± 2 · 8 %, of the XXII stage 84 % ± 2 · 1 % and of the XXIII stage 89 · 6 % (standard deviation not given). The morphological criteria for the stages they define, although imprecise, correspond well with the characteristics defined in this study.
Relation between the stages defined by Shulov & Pener (1963) and those defined in this investigation

SUMMARY
The length of embryonic development in Schistocerca gregaria was measured using specially constructed recorders to determine the time of oviposition and the time the first egg in a pod hatched.
A variation in development time of 330 – 410 h was found in eggs kept at 28 °C.
To relate embryos from pods with widely differing development times, their age was expressed as a percentage of the development time of the rest of the eggs in the pod.
The validity of the method was checked by comparing the external features of embryos removed at 24 h intervals from 6 egg pods with widely differing development times. When arranged in order of their percentage age the external features formed an orderly development sequence.
From the data obtained in this study, five easily identifiable stages have been defined in the later part of development in which external changes are slight. These stages can be used when the more accurate method, which requires exact measurement of the development period, is not required.
RÉSUMÉ
Estimation quantitative du stade de développement embryonnaire du Criquet’. Schistocerca gregaria
La durée du développement embryonnaire de Schistocerca gregaria a été mesurée en utilisant des enregistreurs construits spécialement pour déterminer le moment de la ponte et le moment de l’éclosion du premier œuf.
On a trouvé, dans des œufs maintenus à 28° une variation de durée de développement s’étendant de 330 – 410 h.
Pour exprimer les relations entre les embryons des pontes avec des durées différentes de développement, l’âge de chaque embryon a été défini en pourcentage de durée de développement par rapport à celle du reste de la ponte.
La validité de la méthode a été contrôlée en comparant les formes extérieures d’embryons prélevés à 24 h d’intervalle et provenant de 6 pontes à durées de développement très différentes. Lorsque ces œufs ont été classés suivant le pourcentage de durée, les formes extérieures se rangeaient suivant une séquence ordonnée de développement.
D’après les données obtenues par cette étude, cinq stades bien identifiables ont pu être définis pour la partie-avancée du développement, au cours de laquelle les changements externes sont faibles. Ces stades peuvent être utilisés lorsque la méthode plus exacte, qui requiert la mesure exacte des périodes du développement, n’est pas exigée.
ACKNOWLEDGEMENTS
This work was carried out as part of a Ph.D. research project at the Department of Zoology, at the University of Cambridge and was supported by a grant from the Agricultural Research Council. I should like to thank my supervisor, Dr J. E. Treherne, for his advice and encouragement.