The Imaginal Ecdysis of Blowflies. Observations On the Hydrostatic Mechanisms Involved in Digging and Expansion

In cyclorrhaphous Diptera the splitting and shedding of the old cuticle has become independent of air-swallowing and expansion of the new cuticle can be delayed for considerable periods after emergence. This makes these insects particularly suitable for studying the internal pressures created before and during expansion.

Many authors, notably Fraenkel (1935) and Knab (1909, 1911), have assumed the occurrence of positive internal pressures associated with ecdysis in insects but, so far as I am aware, only one attempt has ever been made to measure them. This was by Shafer in 1922. He used a sharpened 0·336 mm. bore capillary tube which was first touched to water to ‘satisfy capillarity’ and was then inserted into the insect. The height to which the haemolymph rose in the tube (allowing for the height of the original column of water) was taken to indicate the internal pressure. Values obtained from Anax and Aeshna in this way indicate that there is a considerable increase in internal pressure associated with the imaginal ecdysis, but such wide fluctuations were recorded from larvae that it is impossible to decide whether a similar phenomenon occurs at larval moults. Larval anisopterid dragonflies can expel water from the rectal gill chamber with sufficient force to propel themselves rapidly forwards (see Hughes, 1958) so that much of the fluctuation in pressure recorded by Shafer may have been due to the muscular movements associated with this escape reaction.

Clearly, in investigating the pressures associated with expansion it is desirable that measurements be made with an instrument which is not only as nearly isometric as possible but one which is capable of furnishing a continuous record of the pressure changes encountered.

Through the kindness of Dr R. H. J. Brown it was possible to use an electronic condenser manometer to record internal hydrostatic pressures during ecdysis. A block diagram of the experimental arrangement is shown in Text-fig. 1. The insect was connected by means of a fine glass cannula (A) to a brass chamber (B), one wall of which consisted of a circular metal diaphragm 0·11 in. in thickness (C). The whole system was filled with water which was prevented from mixing with the insect’s haemolymph by means of a short segment of paraffin coloured with Sudan III (D).

The diaphragm (C) acted as one plate of a condenser and any minute movements caused by pressure changes within the system varied the frequency of a tuned circuit. Comparison of this frequency with a standard oscillator produced an output voltage proportional to the diaphragm movement and hence to the pressure change. The voltage changes were displayed on an oscilloscope and recorded photographically so that in this way a continuous record of the internal pressure of the insect could be obtained. For purposes of calibration and checking, the condenser manometer was combined with an apparatus for measuring pressure somewhat similar to that described by Kao & Chambers (1954). The brass transducer chamber could be closed off from the insect by means of a tap (E) and by opening another tap (F) a fluid connexion was made with a mercury manometer (G) which could be set at specified pressures by means of a micrometer driven syringe (H). In this way it was possible to calibrate the oscilloscope records by applying a series of known pressures to the diaphragm (C). In addition by opening taps (E) as well as (F) the hydrostatic pressure of the insect could be measured directly by using the syringe to adjust the mercury column to such a height that the water-paraffin interface in the fine glass cannula (A) did not move. It was also found useful to have a tap opening the system to the atmosphere (I) and a reservoir (J) for flushing out the cannula. The brass parts were boiled in distilled water to remove air bubbles, allowed to cool and were assembled while still under water. All other parts were filled with boiled water and great care was taken to exclude gas bubbles.

The best place for inserting the cannula into a fly was found to be at the posterior edge of the scutellum. In a newly emerged blowfly this is largely filled with blood (the two lateral air sacs do not expand until several hours after expansion) and contain virtually no organs which can come up against the tip of the cannula and block it. Moreover, the backwardly protruding shape of the scutellum makes it comparatively easy to obtain a pressure-tight seal.

The experimental fly was kept under constant carbon dioxide anaesthesia during the operation and the cannula was sealed in position by means of a wax-resin mixture applied by an electrically heated probe. The cannula was sufficiently rigidly fixed in this way to support the fly and after the latter had been allowed to recover from anaesthesia it could be given a cotton-wool ball and would expand exactly as if it were suspended from a pin waxed to the dorsum.

After releasing the beam of an automatic balance and adjusting the zero, a short length of glass tubing of internal diameter 4 mm. was firmly clamped so that it was suspended immediately above the left-hand pan. Weights were then added to the right-hand pan bringing the left-hand one firmly against the bottom of the tube. The experimental fly was allowed to dig down the glass tube until it reached the blocked end and the weights on the right-hand pan were adjusted until it was only just able, at maximal expansion of the ptilinum, to displace the left-hand pan downwards away from the tube. Care was taken to see that flies were digging effectively against the pan and not against the glass walls of the tube. The flies were tested over at least 10 min. and the largest weight which still allowed the fly to displace the left-hand pan was taken to be the maximum force exerted by the fly.

The experiments were begun using Calliphora erythrocephala (Meigen) but it soon became obvious that because of its robustness Sarcophaga barbata Thomson was more suitable and in fact most of the measurements have been made with the latter fly. Methods of rearing and handling these insects have been described elsewhere ( Cottrell, 1962 a, b). 

The digging movements of newly emerged flies are highly characteristic and have been briefly described by Lewis (1934) for Glossina and by Fraenkel (1935) for Calliphora. Tracings of prints made from a cinematograph film of a newly emerged female Sarcophaga moving along a glass tube of rectangular cross-section are shown in Text-fig. 2. The tracings represent photographs made at intervals of sec.

At the beginning of a cycle, when the abdomen begins to contract, the head and ptilinum are kept laterally narrowed so that they are extended forwards in the approximate form of a wedge (Text-fig. 2A6, B3 and 5, C3 and 6). With further contraction of the abdomen the ptilinum expands enormously while the head becomes shorter and broader, enlarging the space into which it has been thrust (Text-fig. 2 A12, B12, C12). At this time the contraction of the abdomen is maximal ; not only has it undergone considerable longitudinal shortening but the pleural membranes of the central abdominal segments have been drawn closely together, and the sternites have virtually disappeared through being invaginated into the body (C12). Subsequently the abdomen relaxes slightly, but the head is still kept distended, pressing the compound eyes against the sides of the tube and so acting as an anchor (C15 to 24). Meanwhile the thorax and abdomen are brought up to the head partly by contraction of the neck muscles, partly by certain wriggling movements of the thorax which lead to its longitudinal shortening and to the deepening of the crease which marks the transverse suture (A 15). Particularly in loose media the drawing up of the body is often accompanied by scrambling movements of the legs. Finally, as the ptilinum is retracted the abdomen relaxes completely and the backwardly directed thoracic bristles maintain the insect in its new position while the head regains its normal shape and the neck extends to its resting position (A24 and 27, B24 and 27, C27, 30 and 33). During the relaxed phase the head may be turned from side to side (A3) as if searching for an alternative escape route, sometimes the fly will also probe in different directions with its ptilinum, extending it forwards and then retracting it without broadening it laterally.

Different individuals or the same individual in different situations may show considerable variation in the duration or expression of various phases. Thus when progress is blocked by a cotton-wool plug the ptilinum may be everted and distended several times in rapid succession without any evident ‘bringing up’.

Digging movements may occur in a variety of situations such as when a leg is trapped, when a fly encounters an obstruction or when it is suspended without tarsal contact but they are most easily evoked by contact stimuli to the body (see Fraenkel 1935). A newly emerged fly fixed to the cannula of a condenser manometer can be induced to dig by allowing it to enter a short length of tubing of suitable diameter. This is arranged so that it can slide freely and as digging continues passes backwards over the fly and along the cannula which supports it.

A record obtained from a newly emerged Sarcophaga which is typical of those obtained in this way is shown in Text-fig. 3 (1 a).

The last five cycles on this trace correspond with five typical cycles of digging movements similar to the one described above. The record of each cycle can be divided into three parts :

1. First there is a small rise in pressure amounting to slightly less than 1 cm. of mercury. This appears to correspond with the protrusion of the laterally narrowed ptilinum see [Text-fig. 2 (A 6, B3 and 6, C3 and 6) and compare Text-fig 1b].

2. The small primary rise quickly goes over into an abrupt, spikelike pressure increase which may amount to 10 cm. of mercury or more. This appears to correspond with the phase when the abdomen is maximally contracted, while the head is shortened and laterally broadened [Text-fig. 2 (A 12, B12, C12)]. As the earlier parts of trace 1 a show, the upward spike may be quite variable in its development.

3. Finally there is a series of small pressure changes, not reaching much above 1 cm. of mercury, which appear to correspond with the wriggling movements involved in bringing the thorax up to the head and in withdrawing the ptilinum [Text-fig. 2 (A 15 to 27, B15 to 27, C15 to 33)].

The arrangement of the movements which go to make up a cycle such as is seen when a fly is moving down a glass tube may be altered according to circumstances. For instance, Text-fig. 3 (1 c) shows a record obtained from a fly whose progress had been blocked by a cotton-wool plug. Here the ptilinum was thrust forward and sharply broadened a number of times in succession without any intervening movements of bringing up the body. An interesting feature is that the pressure records of these ‘bouts’ of ptilinum movements are all strikingly similar, each consisting of five clearly defined upward pressure spikes arranged in an almost identical manner.

The development of the relatively high pressures associated with the spikes in normal digging is not dependent on the fly being enclosed within the supporting walls of a glass tube. Thus the record shown in Text-fig. 3(1d) was obtained from a freely suspended fly which was attempting by means of its legs to pull itself forwards away from the cannula and on to a wooden support held in front of it. Whilst struggling in this way it was also making irregular digging movements extending and broadening its ptilinum in the normal manner. It can be seen that the internal hydrostatic pressures so created were of the order of 11 cm. of mercury, and if anything greater than those produced within a glass tube [see Text-fig. 3 (1 a, b and c)].

Records of different flies moving along glass tubes do not differ greatly in pattern [see Text-fig. 3 (1 a, 2a and 4) for three examples], but the arrangement may be considerably altered when the same flies are made to dig in different situations. Records 2 b and c and 3 a and b (Text-fig. 3) were obtained by allowing flies to dig down plastic tubes which had been slit along one side. This gave the tubes a springy quality and resulted in considerable modification of the pattern characteristic of movement in glass tubes. In 2b the modifications take the form of an exaggeration of the pressure associated with the thrusting forward of the ptilinum and a reduction of the pressure spike. In 2 c and 3 a and b the spike shows a series of divisions, possibly the result of the inward pressure produced by the walls of the tube. Records 2 c and 36 also show the emergence of the head of the digging fly from its plastic tube (after the 4th cycle in 2 c and after the downward arrow in 3b) and associated with this emergence a change in pattern to that characteristic of a fly moving down a glass tube.

The maximum pressures recorded during struggling or digging in ten specimens of Sarcophaga ranged from 6 to 12 cm. of mercury (mean + s.e. = 9·9 ±0·7 cm. of mercury). It is interesting to compare these results with the maximum force which specimens confined in glass tubes of 0-4 mm. internal diameter could exert against a plate (balance pan) blocking the end of the tube. In ten specimens this was found to range from 9 to 23 g. (mean±s.E. = 16·0+ 1·7 g.). When the ptilinum is fully expanded and broadened it completely fills the bore of a tube of this diameter and as the maximum internal hydrostatic pressure appears to occur at just this moment it seems reasonable to assume that the area available for transmitting such a pressure to the balance pan is approximately equal to the cross-sectional area of the tube. This was 0·1267 cm². Table 1 shows the forces which a fly might be expected to exert as a result of various internal pressures ranging from 6 to 12 cm. of mercury acting over this area. The estimated values are in fair agreement with those which have actually been determined and this indicates the efficiency, in tubes of this diameter, of the bulging thorax and backwardly directed bristles in preventing slipping.

Of the movements which occur in phases I and II that are not directly related to expansion, only the cleaning of the ptilinum with the forelegs produces definite pressure changes. These consist of short pressure pulses, not greater than 2 cm. of mercury in magnitude, which are apparently due to the partial extrusion of the ptilinum as it is rubbed with the forelegs. Running merely produces a slight irregularity of the traces.

Once a fly has begun to swallow air it remains relatively still so that, except for the cleaning of the ptilinum which occurs in phase II, all pressure changes can be attributed to the process of expansion.

Extracts of a record obtained from a specimen of Sarcophaga barbata Thomson are shown in Text-fig. 4. The scale is in centimetres of mercury and the numbers on the time-marker line refer to the time after emergence in minutes. Two phenomena can be distinguished. First there is a gradual increase in the basic haemolymph pressure which reaches a peak a few minutes after full wing extension and then falls gradually away after air-pumping ceases. Secondly, superimposed on the gradual rise, there is a rhythmic series of brief pressure pulses which cease at about the time of full wing expansion. In timing, these pressure pulses correlate with the ‘muscular effects’ or simultaneous contractions of ptilinal and abdominal musculature which have been described in a previous paper ( Cottrell, 1962a).

Text-fig. 5 has been drawn from data obtained from the complete record extracts of which are shown in Text-fig. 4. For the period during which the brief pressure pulses occur, the highest and lowest pressures recorded in any minute were plotted. This procedure is thought to be justified because in any given minute the pressure pulses are all of approximately the same size. The difference between the maximum and minimum pressures recorded in any minute gives a very approximate indication of the magnitude of the pressure pulses, but it must be remembered that it also includes a large contribution due to the slope in the record of basic haemolymph pressure during that minute. Once the pressure pulses have ceased only the pressure at the mid-point of each minute interval is plotted.

In addition to the above representation of the pressure changes, curves showing the number of ‘muscular efforts’ per minute and the rate of pulsation of the pharyngeal muscle have also been plotted.

Such plots show rather more vividly the two phenomena already noted in the pressure record. In the case of the individual illustrated the basic haemolymph pressure reached about 105 mm. of mercury some 3 min. after full wing extension and then returned gradually to atmospheric pressure during a further 20 min. The pressure pulses gradually diminish in size towards the achievement of maximum pressure, but at their largest represent a transitory pressure increase over the basic haemolymph pressure of some 2·5 cm. of mercury. Comparative values for pressure curves obtained from six specimens of Sarcophaga are shown in Table 2.

Table 3 shows that Calliphora generally produces rather lower maximum pressures.

The maximum pressures detected by means of the condenser manometer have been confirmed by direct measurement using a modification of the method described by Kao & Chambers (1954) (see Methods).

Both pressure pulses and the rise in the basic haemolymph pressure occur during air-pumping in flies which have had their cuticles prehardened by transfusion with active blood (see Cottrell, 1962b). The pressure pulses may be somewhat reduced in magnitude (none greater than 15 mm. of mercury has been recorded) but they are still quite distinct. The maximum height reached by the basic haemolymph pressure may also be somewhat lower than normal in some individuals but in others pressures greater than 70 mm. of mercury have been recorded.

To check the origin of the two pressure phenomena, the probosces of a number of flies were blocked with wax before attachment to the cannula. The pulsations of the pharyngeal muscles could still be observed but no air was swallowed and expansion did not occur. Records obtained from such flies show that although the pressure pulses are still present, there is no rise in the basic haemolymph pressure. The maximum differences between the maximum and minimum pressures recorded in any one minute in five specimens of Sarcophaga treated in this way were 18, 20, 22, 25 and 28 mm. of mercury (mean = 22·6 mm.). This is rather lower than the mean maximum difference (42·0 mm.) recorded from untreated Sarcophaga and might therefore suggest that the pressure-pulse size had been reduced. However, it must be remembered that the latter figure includes a contribution due to the slope of the basic haemolymph pressure and this has been found almost to make up the difference. The rise in basic haemolymph pressure is therefore attributable to the swallowing of air.

Further evidence was obtained by denervating the abdominal muscles. This was done by tying a very tight ligature about the base of the abdomen and then undoing it again. The abdomina of such insects remained completely flaccid until after cuticular hardening had occurred. The flies were unable to expand their ptilina even under appropriate stimulation (see Text-fig. 6B) although they would go through cycles of narrowing and broadening the head. If the abdomen was compressed between the fingers at the appropriate point in the cycle, the ptilinum would expand and the fly was then capable of withdrawing it again. This was considered to indicate that the abdominal muscles had been put out of action. No attempt was made to denervate the ptilinal musculature although, as has been indicated, certain of its components participate in the muscular efforts. Denervation of the abdominal musculature does not prevent the swallowing of air which passes into the midgut as usual.

Records obtained from specimens of Sarcophaga treated in this way showed that rhythmic pressure pulses had been virtually eliminated and that the basic haemolymph pressure rose smoothly to a peak with only very minor irregularities. The maximum differences between the maximum and minimum pressures recorded in any given minute for five flies were 12, 10, 18, 17 and 22 mm. of mercury (mean = 15·8), these differences being almost entirely due to the slope of the record of basic haemolymph pressure over the given minute.

Taken together with previous evidence this makes it almost certain that the pressure pulses result from the ‘muscular efforts’ while the rise in the basic haemolymph pressure is due to the pumping into the gut of quantities of air.

Flies which are prevented from swallowing air by blocking of their probosces do not expand and it is clear that in cyclorrhaphous Diptera air-swallowing is an essential part of the process of expansion. By contrast, flies in which ‘muscular efforts’ have been largely eliminated by denervation of their abdominal musculature do expand although their expansion is always in some degree abnormal. What then is the part played by the’ muscular efforts’ ? Although a direct answer to this question cannot be given, a more detailed consideration of expansion in denervated flies at least suggests some possible explanations.

In denervated flies full wing extension always occurs late in the air-pumping cycle and in contrast to untreated flies the membrane of the wing expands before the costal elbow is opened out (Pl. la and b). Indeed in many cases traces of the elbows remain after expansion has been completed. Correlated with this the frequency of pulsation of the pharyngeal muscles is maintained at a high rate for a considerable period of time producing ‘flat-topped’ rather than ‘peaked’ curves. This indicates that, even though expansion in denervated flies is slower than usual, the flies are in fact swallowing more air than would untreated specimens. The explanation of this apparent paradox lies in the gross distension of the abdomen exhibited by denervated flies. The intersegmental membranes of the genital segments become fully extended (in normal flies they remain more or less folded) and in some cases the rectum and parts of the female genital organs become everted. Apparently in unoperated flies certain of the abdominal muscles serve to prevent this distension and so enable expansion of other parts to take place after the swallowing of smaller quantities of air.

The maximum pressures recorded for denervated flies (40, 43, 57, 65 and 78 mm. of mercury; mean = 56·6) are always lower than those recorded for untreated flies (mean 94·8 mm. of mercury). This would account for the lateness of the wing extension; the probable reason as to why it brings about the expansion of the wing membrane before the opening of the costal elbow will be dealt with in a subsequent paper ( Cottrell, 1962 c). It is worth stressing that the recorded pressures were measured in the thorax and not in the abdomen. In denervated flies the air-filled midgut, which becomes tightly pressed against the abdominal wall, would trap large quantities of haemoloymph in the posterior part of the abdomen resulting in its distension. Thus the ‘muscular efforts’ may serve to force blood forwards out of the abdomen making it available for the transmission of pressure and hence for the expansion of the appendages. It seems that the production of large but undirected pressures by means of air-swallowing is insufficient for normal expansion. The regulatory activities of the muscular system as reflected in the ‘muscular efforts’ and their accompanying pressure pulses are also required.

It has already been noted that in untreated flies ‘muscular efforts’ cease at about the time of full wing extension. In flies which have had their probosces blocked there is no rise in the basic haemolymph pressure and neither the wings nor any other parts of the cuticle expand; yet the ‘muscular efforts’ and their accompanying pressure pulses cease at about the same period in the air-pumping cycle as they would in an unoperated fly. Further, the pressure pulses diminish in size towards the time of cessation in just the same way as they normally do. Evidently the regulation and cessation of ‘muscular efforts’ is not dependent on the achievement of expansion.

It has been noted that a ‘muscular effort’ consists essentially of a contraction of the abdomen very similar to that which produces eversion of the ptilinum during digging but which in this case is accompanied by a contraction of the ptilinal muscles. Denervation of the abdomen by means of a tight ligature (which is subsequently removed) almost entirely eliminates the pressure pulses which accompany ‘muscular efforts’ and also prevents the development of the pressures characteristic of a normal digging cycle (Text-fig. 6 B). Very probably the same abdominal muscles are concerned in the production of both these pressure phenomena. Though it is not at present possible to evoke ‘muscular efforts’ at will, digging movements can be readily produced in newly emerged flies by subjecting them to contact stimuli. In view of the cessation of ‘muscular efforts’ which occurs naturally during the course of expansion it is of considerable interest that after full wing extension contact stimuli are no longer capable of bringing about digging movements.

Text-fig. 6 A is a record obtained from a newly emerged fly (with its proboscis blocked) digging along a glass capillary tube. After this fly had been removed, placed in conditions suitable for expansion and allowed to complete the full cycle of airpumping movements (without of course expanding), it was again made to enter the tube and the pressure changes shown in Text-fig. 6C were recorded. It is evident that these bear no resemblance to the ones produced before air-pumping and certainly the fly made no movements which even remotely resembled normal digging. The small pressure increases which were recorded appeared to be due to a rather vigorous backwards and forwards movement of the head against the thorax combined with a slight buckling of the prothorax which accompanied leg movement and seemed to be due to the contraction of the extracoxal depressor muscle. It is thus clear that the in effectiveness of contact stimuli in causing digging after air-swallowing is not dependent on the expansion of the cuticle. However, it might be argued that even though the fly with its proboscis blocked was transferred to the digging tube immediately after the cessation of air-pumping movements, the cuticle might already have hardened sufficiently to prevent adequate compression of the abdomen by the appropriate muscles. This possibility can be readily disproved by transfusing a digging fly with active blood to preharden the cuticle (see Cottrell, 1962 b) and then recording the pressure changes during digging. Text-fig. 6D shows a trace obtained from such a fly. The normal pattern is somewhat distorted, but all the component pressure changes of a normal digging cycle are still present.

These experiments show that, just as with the cessation of ‘muscular efforts’, the ineffectiveness of contact stimuli in provoking digging after air-pumping is not dependent on either the expansion of the cuticle, the attainment of maximum pressure or the hardening of the cuticle. It seems that at about the time when the wings would expand in a normal individual the fly loses the ability to excite those abdominal muscles which produce the compression of the abdomen.

Digging movements and ‘muscular efforts’ are produced by the activities of both the head and the abdomen, and it is of some interest to examine the muscular systems which bring them about.

The ptilinal musculature of newly emerged Calliphora has been described in detail by Laing (1935), but there appears to be no detailed description of the abdominal musculature of a newly emerged blowfly. That of Sarcophaga was investigated by dissection and by means of flattened, formalin-fixed, whole mounts of the body wall examined under a polarizing microscope. Only the musculature of the visceral segments (numbers 1 to 5) was examined because the remaining segments play little or no part in either muscular efforts or digging movements.

The internal layer of body musculature (Snodgrass, 1935) is well represented (Text-fig. 7) and consists of long fibres grouped together to form discrete muscles. In each segment there are: four pairs of dorsal muscles (10a, b, c and d); one pair of paradorsal muscles (11) ; two pairs of lateral muscles (12a and b); one pair of oblique lateral muscles (13); and one pair of ventral muscles (14).

The tergite of segment 1 is much reduced and all the muscles which belong to this segment extend forward from the anterior margins of the tergite and sternite of segment 2 to the thorax. Segment 1 differs from the basic pattern in that the two pairs of lateral muscles are absent and in the presence of an additional pair of muscles (15) extending from the anterior ventral angle of the tergite to the metathoracic sternal apophysis. Segment 5 differs in the possession of an additional pair of median dorsal muscles (16) and some rather complex ventral muscles (17 a, b and c) associated with the lobes of the 5th sternite. The two pairs of lateral muscles extend from near the midline of the sternites to well above the level of the spiracles on the tergites. Their contraction causes the sternites of the newly emerged fly to fold along the midline and become drawn into the body [see Text-fig. 2 (B12)]. All the other internal muscles (except two of the pairs of dorsals in segment 2) are of segmental length and stretch from the anterior edge of one sternite or tergite to the anterior edge of the next sternite or tergite. In this respect they resemble the musculature of a soft-bodied holometabolous larva and because the sternites and tergites of a newly emerged blowfly are soft and unsclerotized they can act in the same way, shortening and compressing the abdomen by buckling the presumptive sclerites.

The external layer of the musculature consists of fibres of much shorter length than those of the internal layer and the individual fibres are all more or less widely separated. Basically, they fall into three groups: dorsal (18), lateral (19) and ventral (2); but the ventral fibres are poorly represented and have been clearly recognized only in segment

2. The dorsal group is developed only in segments 2, 3 and 4, but the lateral group is present in all segments.

As we have seen, muscular efforts cease, and digging movements can no longer be induced, after a newly emerged fly has reached the point in the air-pumping cycle at which its wings would normally expand. From a teleological point of view this is clearly an advantage to the fly since the actions of many of the muscles concerned in these activities result in considerable changes in shape. Thus, if a fly attempted to make these movements after expansion had occurred but before hardening was complete, buckling of the segments and malformations of other parts would result.

Once hardening and darkening has been completed, what becomes of the muscles which were concerned in the newly emerged fly with compression of the head and abdomen? Laing (1935) found that in Calliphora, within 2 days of emergence, all but two of the muscles connected with the ptilinal mechanism had disappeared. In Sarcophaga the abdominal musculature of the mature fly (Text-fig. 7 B) is very different from that of the newly emerged one. The external layer of muscles is carried over unchanged but the internal layer is considerably reduced. The muscles of the internal layer which remain are as follows: in all segments, the ventral muscles; in segment 1, the four pairs of dorsal muscles and the pair of additional muscles; in segment 5, the pair of additional median dorsal muscles and the muscles associated with the sternal lobes. All of the other internal muscles disappear completely within 4 or 5 days of emergence.

Those muscles which persist in the adult fly may or may not have been concerned in the characteristic movements of the newly emerged insect: the degeneration of others indicates that they were almost certainly involved in those movements which occur during the brief period in the imaginal instar when the blowfly depends partially on a hydrostatic skeleton and the mode of locomotion which goes with it.

A very approximate value for the blood volume of a newly emerged blowfly can be obtained by determining the loss of weight after slitting the insect open and removing as much as possible of the blood by gentle compression and blotting with filter paper. Because it is obviously impossible to remove all the blood by these means, such estimates are always low though within a given batch of flies they are surprisingly consistent. Results obtained from four batches of flies are shown in Table 4. The mean body weights of newly emerged males and females of the same batch are usually significantly different and they have therefore been listed separately. However, provided it is expressed as a percentage of the total body weight, the loss due to removal of the blood does not differ between the sexes and the results have therefore been combined. The values obtained for the percentage contribution of blood to total body weight are high, being of the order of 30 %, but as has been mentioned even this figure is probably an underestimate.

The blood volume of a newly emerged blowfly is at least as high as the highest value recorded for soft-bodied larvae (cf. Beard, 1953), and this is surely a reflexion of the fact that it is partly dependent on a hydrostatic skeleton. Essentially this is a mechanism whereby different contractile elements (in this case those of the head and abdomen) can be antagonized (Chapman, 1958). In the newly emerged blowfly there are no compressible air sacs so that the haemolymph can transmit pressure efficiently in all directions. During ‘muscular efforts’ both the head and abdomen contract simultaneously so that there is an increase in pressure but little movement of fluid (except perhaps into the appendages). During digging the head and abdomen contract out of phase (except at the time of the pressure spike) and there is bulk movement of fluid first into the head and then into the abdomen. In this respect the system resembles that of the tube feet of starfish (Smith, 1946) and differs from that of the earthworm where there is little or no bulk movement (Newell, 1950).

Although a hydrostatic mechanism is essential for digging and for expansion, such systems are undesirable in an adult insect where activities such as flight depend on the buckling characteristics of parts of the exoskeleton—characteristics which would be considerably altered by changes in internal pressure. In the adult blowfly the possibility Of transmitting pressures from one part of the body to another is greatly diminished by the filling of the large air sacs. In addition, the haemolymph volume is very greatly reduced shortly after expansion. Table 5 shows that the mean body weight of samples of eight specimens of Sarcophaga fell by 22 · 0 mg. from 85 · 1 mg. to 63 · 1 mg. in the 30 hr. following emergence. At the same time the weight of blood which it was possible to remove by gentle compression and blotting fell by 19 · 8 mg. from 25 · 5 to 5 · 7 mg. Part of this diminution in blood volume is no doubt due to loss of water by transpiration through the cuticle, but in the first few hours following expansion a considerable amount of clear fluid can be seen to be eliminated through the anus. In an attemptto estimate what part of the total loss was due to this elimination through the anus, the changes in weight shown by starved normal Sarcophaga over the first 25 hr. of adult life were compared with those shown by starved individuals in which the anus had been blocked shortly after emergence. The results are shown in Table 6. It can be seen that from expansion (i.e. 1 hr. after emergence) to 10 hr. after emergence the normal flies lost weight at a much higher rate than those in which the anus was blocked. Dissection showed that in individuals of the latter group the rectum was abnormally distended by comparison with those of the former. The two groups began to lose weight at approximately the same rate after the untreated flies had lost an average of about 7 % more of their original body weight than the treated flies. These observations suggest that, in this rearing batch, fluid amounting to some 7 % of the body weight was eliminated through the anus in order to bring the blood volume down to a level suitable for adult life.

Expansion in Sarcophaga. (a) and (b). Aberrant expansion of wings in male with denervated abdomen, (c) and (d). Normal female at approximately the same stages in the pumping cycle as (a) and (b).