The Effect of Carbon Dioxide and Certain Respiratory Inhibitors on the Respiration of Larvae of the Horse Bot Fly (Gastrophilus Intestinalis De Geer)

The respiration of Gastrophilus larvae has been investigated by Kemnitz (1916), Dinulescu (1932) and, in unpublished work to be described below, by Keilin. Kemnitz was principally interested in overall metabolic changes, and his data are difficult to interpret in terms of oxygen uptake per unit weight of larva. In addition, since his experiments were continued over a number of days, it is almost certain that part of the gas changes which he measured was due to bacterial contamination. Dinulescu’s data, on the other hand, are subject to a number of technical criticisms ; these include insufficient attention to adequate temperature control, experiments of only short duration and, most important of all, no measurements in the presence of CO₂. Within the limits of these qualifications, however, he found that the respiratory rate was highest in 2nd instar larvae and lowest during the long quiescent period (diapause) which occurs during the 3rd instar, while keeping the larvae for any length of time ^‘in vitro’produced a progressive decrease in the rate of O₂ uptake.

References to the life cycle, morphology and bionomics of the Gastrophilus larva may be found in the works of Dinulescu (1932), Keilin (1944) and Keilin & Wang (1946), and hence attention need only very briefly be drawn to two peculiar respiratory adaptations in the larva of importance for the present work.

The first of these concerns the mass of giant tracheal cells at the posterior end of the body forming the so-called tracheal organ. These cells are rich in haemoglobin, the properties of which were studied by Keilin & Wang (1946), and are abundantly supplied with intracellular tracheoles. The actual importance of the tracheal organ in respiration is yet to be investigated, but from the available evidence it would appear likely that it functions as a primitive ‘lung’, especially at low oxygen tensions. Secondly, in the 2nd and 3rd instars, the post-abdominal spiracles lie in a depression or pouch formed by transverse dorsal and ventral folds or lips of the larval cuticle. These lips exert a regulatory function in respiration, since they can be expanded as result of muscular pressure transmitted through the larval haemolymph and, in thus completely sealing off the spiracle, they effectively prevent any spiracular gas exchange. It may be noted that a spiracular closing mechanism is relatively rare among Dipterous larvae, and when present is usually indicative of an aquatic or semi-aquatic habitat, e.g. larvae of Culicidae, Bibionidae, Psychodidae, etc.

Mature, 3rd instar G. intestinalis larvae were obtained from a local knackery. As soon as possible after the killing of an infected host, and still attached to the stomach, they were transported to the laboratory in a thermos flask containing water at 38° C. After washing the stomach with warm water, the larvae were carefully picked off the mucosa with forceps, rinsed in warm water, and either used immediately, or kept in a large dish containing slightly acidified tap water at room temperature. Larvae which had been kept warm continuously and were employed for an experiment within an hour after their removal from the horse will be referred to as fresh larvae, while those kept for 1 or 2 days as 24 or 48 hr. larvae respectively.

Except where stated otherwise, all experiments were carried out on diapausing larvae obtained between the beginning of November and the end of May.

The majority of the present experiments were carried out with a modified Haldane manometric apparatus which is shown diagrammatically in Fig. 1.

A circular flat-bottomed glass vessel A, approximately 6 cm. in diameter and 3 cm. in depth, was made with side indentations on which rested a stainless steel gauze B. A small funnel D, blown into one side of the base, was connected to a water suction pump through a piece of wide bore capillary tubing and a glass tap. A small centre well E to contain water was sealed on to the bottom of the vessel. This vessel was closed with a well-fitting paraffin-waxed rubber bung, on to the lower surface of which was cemented a thin Perspex disk to prevent the larvae from attaching themselves to the rubber.

A small thistle funnel C, made of wide bore capillary tubing and fitted with a glass tap, passed through the rubber bung and Perspex disk to within 1–2 mm. of the bottom of the vessel, while a narrow bore capillary tube inserted into the bung connected to a Haldane (1920) constant pressure manometer. The latter was slightly modified in that the normal 1 ml. graduated pipette was replaced by a 5 ml. pipette, H. The other limb of the manometer led to a compensating flask of the same size as the experimental vessel, and contained approximately 9 ml. water. Both vessels were immersed in a thermostatically controlled constant temperature water-bath.

The procedure for carrying out a measurement was as follows. 0·5 ml. water was put into the small centre well, and twenty weighed Gastrophilus larvae were placed on the steel gauze disk. The apparatus having been assembled, both vessels were clamped in position in the water-bath with all taps open. By means of gentle suction at D, a slow stream of air was passed through the experimental flask for 90 min. to allow of temperature equilibration and settling down of the larvae. (This time was empirically determined as being the minimum required in order subsequently to obtain fairly constant readings. The liquid level in the graduated pipette was set at a convenient height by raising or lowering G, the water pump turned off, and the manometer taps were turned so as to connect each side with its corresponding vessel. The fluid level in the U-tube having been levelled by means of F only, the tap at D was closed and a reading taken of the meniscus in the pipette.

The larvae were now allowed to respire for a known period of time, generally 1 hr., after which the level of the liquid in the U-tube was again levelled by means of F, and a second reading taken on the pipette. A known volume—3 ml. was a convenient amount—of 2N-K0H previously warmed to 37° C. was then run into the bottom of the vessel through C, leaving that part of the funnel between the tap and the end full of alkali so as to prevent the entry of air. The volume of air introduced initially in running in the KOH was negligible. The experimental vessel was then very gently shaken for exactly 5 min., this time having been found sufficient in pilot experiments for the absorption of the respiratory carbon dioxide, and a third reading taken on the pipette.

If the KOH were allowed to remain in the flask, the subsequent rate of O₂ uptake in the absence of CO₂ could now be measured directly. Alternatively, by opening the tap at C and applying suction at D, the alkali could be removed, any remaining traces being neutralized by running through a little dilute acid from the thistle funnel. The acid was followed by distilled water and a current of air. After a short period of re-equilibration, further measurements could again be made in the presence of CO₂, without the larvae having in any way been disturbed due to handling or shaking throughout the procedure.

In order to calculate the results, it is necessary to take into account the change in reading on the pipette due to the introduction of a volume of liquid equal to that of the KOH, and also the oxygen utilized during the 5 min. during which the CO₂ was absorbed. The assumption is made that the rate of O₂ uptake during these 5 min. is the same as for the preceding hour. The equations for the conversion of the pipette readings into values for the O₂ uptake and CO₂ production were derived as follows.

Adopting the usual convention that all gas evolved produces a positive, and all gas absorbed a negative reading, then if

x = amount of O₂ absorbed in time t_1,

y = amount of CO₂ evolved in time t₁,

and h₀= initial pipette reading,

h₁ = pipette reading at time t₁,

h₂ = pipette reading at time i₂ after absorption of CO₂,

The barometric pressure was noted and these values, corrected to N.T.P. and t₁ = 60 min., were divided by the weight of the larvae to give the required figures for the and expressed throughout this paper as μl. of gas per mg. wet weight per hour. This procedure will be referred to below as the Haldane manometric method.

The protocol of an actual experiment may be cited as an example.

Wet weight of twenty Gastrophilus larvae = 8·597 g.

A single weighed larva and 0·05 ml. water were placed in the centre well of a normal type Warburg flask of c. 16 ml. capacity. The main compartment contained 2 ml. M/5 potassium permanganate in M/500 H₂SO₄, the side bulb 0·2 ml. of 30% sodium iodide (acidified to N/500 H₂SO₄ just before use). The larva was allowed to respire for a known period of time, after which the two solutions were mixed, resulting in the production of a highly alkaline mixture which on shaking the manometer at 100 oscillations per min. rapidly absorbed the respiratory CO₂ (Krebs, 1930).

In calculating the results for this modified Dickens & Šimer technique a correction had to be applied for the small positive pressure produced in the manometer as a result of mixing the solutions; the value for this was obtained by using a control manometer containing only these two solutions. The assumption is made that the rate of O₂ uptake during the time allowed for absorption of CO₂—generally 5 min.— is the same as for the precedirig period. The results are calculated as follows ; if

x is the O₂ absorbed in time t_1,

y is the CO₂ evolved in time t_1,

h_o is the initial manometer reading,

h₁ is the manometer reading at time i_1;

h₂ is the manometer reading after mixing the solutions at time t₂, corrected for the control,

After most of the present experiments had been completed, Laser & Rothschild (1949) described a manometric apparatus for the measurement of respiration in the presence of CO₂. A number of additional experiments were therefore carried out using this method according to the details described in the paper of these authors.

Unless otherwise stated, all measurements were made at 37° C.

Before the respiration of Gastrophilus larvae could be measured by manometric methods, it was necessary to ascertain whether significant quantities of any volatile acid or base were liberated, especially since free ammonia is known to be excreted by other species of Dipterous larvae. The method adopted was to bubble air passed over a batch of respiring larvae into either N/100 acid or alkali. After some hours the acid or alkali (the effect of respiratory CO₂ had to be considered here) was titrated and the titre compared with a suitable control. These experiments produced no evidence for the formation by the larvae of any volatile substances which could be detected by this procedure, and hence it was concluded that manometric methods could safely be employed.

The O₂ uptake of Gastrophilus larvae as measured over a period of hours by Warburg’s direct method, i.e. by allowing 10 –15 min. for temperature equilibration and absorption of CO₂ with KOH, showed that the rate of O₂ uptake declines in an irregular, but nevertheless continuous, manner. A typical series of values is shown in Table 1. This decline might be due to either or both of two causes, viz. acclimatization of the larvae to the experimental conditions imposed by the technique (i.e. the so-called settling down effect well known to occur with insect material), or to the absence of carbon dioxide.

There is little doubt that the former phenomenon is of considerable importance in the case of Gastrophilus-, at 37° C. the larvae are very active and move about attempting to attach themselves to any material which they can pierce with their mandibles. However, of greater significance is the fact that after some hours in the manometer flasks larvae are occasionally found to be moribund. Such larvae can readily be distinguished by their red colour; this is due to the leakage of haemoglobin from the tracheal organ cells into the blood of the larvae, which is normally of a light amber colour.

From many experiments using the direct manometric method, it appeared very probable that this lytic effect and the progressive decrease in the was due to absence of CO₂; Keilin in earlier unpublished experiments had obtained similar results which also led to this hypothesis, and his technique was adopted to obtain conclusive evidence.

A batch of larvae was divided into three groups; larvae of the first group were enclosed in a loose-fitting stainless steel cage, those of the second had their spiracular lips kept open by the insertion of a small brass paper fastener with sharpened ends bent outwards at c. 45 °, and will be referred to below as larvae with opened spiracles, while in the third group the spiracular lips were ligatured, thus preventing any gas exchange through the spiracle. Attached to either the cage, the clip or the ligature thread, was a length of stiff wire the top end of which was fastened with a piece of split rubber pressure tubing inside the hollow neck of a Thunberg tube stopper. Thus, on inserting the stopper, the larvae were suspended in mid-air inside the Thunberg tube. The tubes were evacuated and filled with a variety of gases ; except when carbon dioxide was a constituent of the gas phase, tubes were set up in duplicate, a filter-paper roll soaked with 2N-K0H being inserted near the bottom of one of the tubes. Larvae contained in the latter tubes, therefore, were in an atmosphere completely devoid of CO₂. A little water was placed inside the hollow stopper to prevent desiccation, the tubes were incubated at 38° and the larvae examined at hourly intervals. The results of three such series of experiments are shown in Table 2.

It will be seen that in normal larvae, or those with opened spiracles, leakage of haemoglobin from the tracheal cells into the blood occurred only in the absence of CO₂; when the posterior spiracles were ligatured, thereby considerably reducing the loss of respiratory CO₂, the larvae remained normal. Results similar to those shown in Table 2 were obtained on a fourth experiment the duration of which was extended to 18 hr.

In order to examine whether the absence of carbon dioxide affected the tracheal cells directly, the tracheal organ was dissected out and, suspended by the spiracular plate inside a Thunberg tube, was immersed in the insects’ own haemolymph. In all other respects the experiment was the same as for whole larvae, except that the roll of KOH-moistened filter-paper was placed half way up the tube with a layer of grease immediately below to prevent any alkali reaching the blood or cells. The results were in agreement with the former series of experiments ; in the absence of carbon dioxide haemoglobin leaked out from the cells into the surrounding blood, thus demonstrating that the cause of the impaired permeability of the tracheal cells is most probably loss of CO₂ from the haemolymph.

Evidence that a certain amount of CO₂ is indeed lost from the blood of larvae with lysed tracheal cells was obtained by pH measurement with the glass electrode. A sample of haemolymph taken from larvae with opened spiracles kept over KOH an atmosphere of N₂ had a pH of 7 ·14, which is 0 ·3 –0 ·4 pH unit more alkaline than normal blood (cf. Levenbook, 1950 a).

The first series of experiments with the Haldane manometric method was carried out on 3rd instar Calliphora erythrocephala larvae at the stage when they had cleared their gut. The insects were obtained from a culture which had been in-bred in this laboratory for many generations. Fifty larvae were placed in the manometer vessel, 90 min. allowed for equilibration and the respiration measured at 25° C. To test whether carbon dioxide had any effect on the respiratory rate, the measurements were made in the presence or absence of KOH during each alternate hour as described above. The result of a typical experiment is shown in Table 3.

It will be seen that, apart from the atypical 1st hour’s reading, the presence or absence of CO₂ was without effect on the respiration of Calliphora larvae. After the 1st hour the values were reasonably steady, with an average value of 0 ·52. This is considerably lower than the corresponding figures obtained, among others, by Fraenkel & Herford (1938), Agrell (1947) and Hurst (1949), who found the to be in the region of 1 ·0. This may be explained, at least in part, by the longer initial equilibration period in the present measurements. As evidence for this, another experiment may be cited in which the equilibration period was reduced to 45 min., when the for the 1st hour in the presence of CO₂ was 0 ·92, and 0 ·87 during the 2nd hour in the absence of CO₂.

Measurement of the respiration of Gastrophilus larvae under similar conditions gave quite different results. A typical experiment using twenty fresh larvae is shown in Table 4.

The mean of eight series of experiments of this type have been combined in the histogram shown in Fig. 2. The average initial of fresh larvae in the presence of CO₂ is close to 0 ·27, or about 0 ·13 ml. O₂ uptake per larva per hour at 37° C.

During each hour’s respiration in the absence of CO₂ the was lower than for a preceding or subsequent hour when CO₂ was present. However, the increased during the subsequent hour as compared with the lower value in the absence of CO₂ was insufficient to bring the rate of O₂ uptake to the pre-existing level before the CO₂ was removed. Hence, under the described experimental conditions, the declined in a stepwise manner during 8 hr. The respiratory quotient (R.Q.) during the same period increased from the original value of c. 0 ·9 to 1 ·1 –1 ·2. After some 8 –10 hr. the final was about 0 ·04 –0 ·08, and once this low rate of O₂ uptake had been attained, removal of CO₂ from the atmosphere had only a very slight or even no effect on the respiration.

A number of measurements were made on larvae which had remained in the manometer vessel at 37° C. for 18 hr. During this period either a very slow current of air was sucked through the apparatus, or KOH was present and all the taps were left open. It may be noted that the O₂ content of the closed manometer vessel was sufficient to maintain the normal respiration rate of twenty larvae for 4 –5 hr. ; unless air was sucked through, it is uncertain to what extent gaseous diffusion through the taps was sufficient to supply the O₂ requirement of the larvae. However, Gastrophilus larvae are metabolically well adapted to survive anaerobically for considerable periods (Kemnitz, 1916; Dinulescu, 1932).

The experiments showed that the initial of these larvae kept in the presence of CO₂ was only slightly lower than that of fresh larvae, ranging from 0 ·20 to 0 ·24, and the effect of removing CO₂ during each alternate hour was entirely similar to that described for fresh larvae. This is demonstrated by a typical experiment shown in Fig. 3.

The of the larvae equilibrated for 18 hr. in the absence of CO₂ was very low, of the order of 0 ·04, and not infrequently some of these larvae were found to be moribund with lysed tracheal cells. This low was scarcely altered in the presence of CO₂ during the next 2 hr.

These latter experiments, therefore, would appear to show that the inhibitory effects on respiration ascribed to the absence of CO₂ can in no way be due to insufficient time allowed for settling down.

If the decline in the larval is attributed to the absence of CO₂, it might be expected that the would not diminish during a period of hours in the continuous presence of carbon dioxide; this was in fact found to be the case, as shown in Table 5.

The values obtained by the Haldane manometric method were confirmed using the Laser-Rothschild (1949) apparatus; the results of one such experiment are shown in Table 6.

From the beginning of June-July, the diapause is broken and the larvae prepare to leave the alimentary tract and subsequently to pupate in the soil. This is correlated with a change in metabolism and, in particular, with an increase in the (Table 7).

A similar increase in respiration is found in most insects on their emergence from diapause (see Wigglesworth, 1939). The larvae continue to be sensitive to the presence of CO₂, since all measurements made at this time showed that in its absence the progressively declined as described above for diapausing larvae (cf. Fig. 2).

During June-July Gastrophilus pupae may be obtained in the laboratory by placing the larvae into a dish containing damp sawdust, when pupation occurs in 2 –4 days at 24° C. In Table 8 data are presented for the respiration of twelve pupae 3 days old; the was found to be 0 ·23, and this was not affected by the absence of CO₂ from the gas phase ; the latter finding was also confirmed using older pupae.

Normal larvae. To obtain an indication of the variation in O₂ uptake between different larvae, the modified Dickens & Šimer method was employed. The during the first 3 hr. or so decreased with time since, as already mentioned, the equifibration period was insufficient to allow for the effect of settling down. Furthermore, once the solutions had been mixed so as to form the alkaline absorbing fluid, there was no means of re-introducing CO₂; hence determinations of the respiratory rate during the 2nd and 3rd hour were made by transferring larvae whose respiration had already been measured during the 1st hour as rapidly as possible to fresh manometer flasks previously warmed to 37° C.

The data obtained by this technique are presented in Table 9.

The for fresh larvae during the 1st hour is considerably higher than the corresponding values found with the Haldane manometric apparatus. This may be attributed almost certainly to the much shorter equilibration period used for the single larvae. If the O₂ uptake after similar equilibration times be compared, i.e. during the 1st hour of the Haldane manometric method and the 3rd hour for the single larvae, the respective values — 0 ·27 and 0 ·29 are in good agreement. It may also be seen that the variation in the for the various larvae obtained from different batches is no greater than might be expected from such heterogeneous populations.

The respiratory function of the post-abdominal spiracle in Gastrophilus was investigated by measuring the respiration of a normal larva and comparing this value with that obtained using the same insect with either opened or ligatured spiracles.

Ligaturing the posterior spiracle decreased the by about 95%, the values obtained ranging from 0 ·02 to 0 ·06. The carbon dioxide production of such larvae was not decreased to a similar extent, and consequently R.Q. values of 4 –9 were obtained. These results are similar to those obtained by Fraenkel & Herford (1938) for Calliphora larvae with ligatured post-abdominal spiracles.

A ligature tied between the brain and the anterior spiracles, which are morpho logically functional (Keilin, 1944), reduced the O₂ uptake in four experiments by 14 –3°%. However, it is improbable that this percentage represents the fraction of the total respiration passing through the anterior spiracles because of the following observations.

Larvae with ligatured posterior spiracles were suspended in Thunberg tubes and the latter evacuated, when the larvae became inflated to 2 –3 times their normal size due to expansion of the enclosed air. After 18 hr. there was only a very slight diminution in volume, and the larvae, none of which died, returned to their normal dimensions when the vacuum was released. If gas were free to diffuse through the anterior spiracles, it seems unlikely that under the extreme conditions of these experiments the larvae would have remained in their inflated condition.

The O₂ uptake of larvae with opened spiracles was slightly, but consistently, increased. The average for nine such larvae measured by the Dickens & Šimer method was increased by < 10 % above that of normal larvae, while the R.Q. increased from 0 ·72 to 0 ·86.

The respiration of four larvae with opened spiracles was also measured in the Laser-Rothschild manometer; allowing an hour for equilibration, the following values were obtained for the subsequent 30 min. period: ⁠, R.Q. = 0 ·84. Comparing these figures with those for normal larvae (Table 6), it appears that opening the spiracles resulted in an increase of almost 50% in both O₂ uptake and CO₂ evolution. This increase in O₂ uptake is much larger than found for the corresponding value using individual larvae, and the explanation probably lies in the fact that in the latter case measurements were made during the 1st hour with only a short equilibration period; as already shown under these conditions the respiratory rate is higher than normal, and the spiracles would be more frequently, if not continuously, open to allow of the greater gas exchange. It might be expected, therefore, that forcing open the spiracles of such larvae would result in a relatively smaller increase in the ⁠.

Some preliminary experiments were carried out employing respiratory inhibitors in order to assess the functional importance of the cytochrome system in the diapausing larvae.

Cyanide is known to inhibit the respiration of almost all aerobic forms of life by combining in a reversible manner with heavy metal enzymes and, in particular, with the oxidized form of cytochrome oxidase.

By means of a fine hypodermic syringe with a micrometer-controlled plunger, known amounts of neutralized KCN were injected into Gastrophilus larvae through a ligature at the level of the 2nd segment. Controls were similarly injected with M/15 pH 6 ·8 phosphate buffer. The final hydrocyanic acid concentration in the larvae was calculated to vary from M/500 to M/1000. At these concentrations cyanide produced a rapid and profound decrease in respiration and killed the larvae.

For the more interesting results obtained by the use of gaseous cyanide, I am indebted to Prof. D. Keilin for permission to quote some of his unpublished experiments. Briefly, these showed that, using the direct manometric method, gaseous cyanide inhibited the respiration of normal 3rd instar larvae almost completely, but the same concentration of cyanide inhibited to the extent of only 30 –50% when the spiracles were forcibly opened. In a further series of experiments in which the larvae were allowed to respire in a closed vessel from which samples of air were withdrawn to determine the respiratory rate by actual gas analysis, the results were quite different. Under these conditions cyanide inhibited the respiration of normal larvae by about 30%, while with opened spiracles the inhibition was >75%. The difference in these results is almost certainly due to the presence or absence of carbon dioxide in the various experiments.

The particular interest attached to this respiratory inhibitor lies in the fact that Levenbook & Wang (1948) found a high concentration of succinic acid to be a normal constituent of Gastrophilus larva blood, and that the tracheal organ cells possess a very active succin-oxidase system. The question therefore arises as to whether the succinate in the blood is metabolized. Since malonate is a specific competitive inhibitor of the succin-oxidase system (Quastel & Wooldridge, 1928), any decrease in the respiration of larvae injected with this substance should be a measure of the O₂ uptake due to the oxidation of succinate.

10 μl. of M/5 sodium malonate were injected into the larvae by the method described above, the final concentration of malonate being approximately equal to the succinate already present. According to Keilin & Hartree (1949) this should be sufficient to inhibit almost completely the succin-oxidase system. Control larvae were injected with the same amount of phosphate buffer. The results of these experiments, one of which is shown in Table 10, indicated that injection of malonate did not inhibit respiration to a greater extent than did a similar concentration of phosphate buffer. This result was confirmed in the following manner.

Two lots of five larvae were placed on damp filter-paper in rubber-stoppered 50 ml. flasks kept at 37° C. Samples of air were then withdrawn from the flasks by piercing the stoppers with a hypodermic syringe moistened with salt solution, and the oxygen and carbon dioxide content of the samples determined by the Roughton & Scholander (1943) technique. The larvae in one flask were then injected with phosphate buffer, the others with sodium malonate, after which the new respiratory rate was again similarly determined. The data so obtained are shown in Table 11. On account of the fact that no time was allowed for settling down, the values shown in Table 11 are rather higher than would otherwise have been expected; nevertheless, it was again found that the injection of malonate did not inhibit the O₂ uptake as compared with the control larvae. It may be concluded, therefore, that in vivo the haemolymph succinate is not oxidized at any appreciable rate.

CO inhibits respiration due to its property of combining with the reduced form of heavy metals forming the prosthetic group of respiratory enzymes and carriers. Keilin & Wang (1946) found that the haemoglobin of the tracheal organ cells was peculiar in that it had a higher affinity for O₂ than for CO. On the assumption that the effect of CO could be demonstrated only when the O₂ tension was reduced to such an extent that any further reduction produced a significant decrease in the ⁠, Keilin (unpublished) found by the direct manometric method that reducing the O₂ tension from 21 to 1 % decreased the O₂ uptake by 50%, but even at this low tension 50% CO had no effect on the rate of O₂ uptake.

In extending Keilin’s experiments, the effect of CO was examined in an atmosphere containing carbon dioxide, using the Laser-Rothschild manometer (Table 12).

The data show that a CO/O₂ ratio of 0 ·5 was sufficient to inhibit reversibly the O₂ uptake by some 60 %. All the larvae were alive at the end of the experiment. Carbon dioxide production was maintained at almost the normal rate, and the R.Q. approached 2. Higher values for the respiratory quotient, and inhibition of the by some 80% were obtained by increasing the CO/O₂ ratio to 9; at this concentration the effect was still reversible and none of the larvae died.

Attempts were made to see whether the inhibition produced by CO was light-reversible; in the two experiments carried out the O₂ uptake did not increase on illumination, whereas CO₂ production was greatly stimulated. It was found subsequently that larvae when brightly illuminated showed an augmented CO₂ production even in the absence of CO (Table 13).

It was also noted that the illumination caused increased larval activity which, unexpectedly, was unaccompanied by a concomitant increase in Q'_o2. The CO₂ production might therefore be due to aerobic ‘physiological’ glycolysis, i.e. the formation of gaseous CO₂ from the haemolymph bicarbonate (Levenbook, 1950 a) by lactic acid liberated as a result of muscular activity.

In the extensive literature on insect respiration, Heller (1928) appears to be the only author to have reported that the respiration of an insect—the pupa of the moth Deilephila euphorbiae—was higher in the presence of carbon dioxide than in its absence. Unfortunately, no details were given. In the case of Gastrophilus the present experiments show that, to sustain the rate of O₂ uptake at its original level, the presence of CO₂ in the gas phase is essential for the 3rd instar larva, but is without effect on the respiration of the pupa. The explanation for this is apparently connected with the necessity for a high concentration of CO₂ in the larval haemolymph to maintain the tracheal organ cells (and possibly the cells of some or all of the other tissues) in a physiologically normal condition. The haemolymph CO₂ content is in turn a function of the CO₂ tension in the environment (Levenbook, 1950 a); hence, when CO₂ is removed from the atmosphere, there is of necessity a concomitant loss of CO₂ from the blood and cellular damage results, which is manifested by lysis, a decline in the and final death.

The fact that Gastrophilus pupae are not sensitive to CO₂ indicates a profound physiological change accompanying the transition to an aerobic, free-living mode of life. Levenbook (unpublished) has found that at the onset of pupation the CO₂ content of the haemolymph is greatly decreased, the intracellular tracheoles are withdrawn from inside the tracheal organ cells, and these show a marked increase in viscosity and are less readily lysed. Superimposed on this are the normal changes e.g. phagocytosis, which accompany metamorphosis, but the nature of the phenomenon occurring at pupation which renders this instar insensitive to CO₂ yet remains to be determined.

The reason why CO₂ should be required to maintain cellular impermeability is at present unknown ; it cannot be due to a decrease in hydrogen-ion concentration, since the pH of the bright red blood of larvae with lysed tracheal cells was scarcely outside the range encountered in normal larvae. Jacobs (1922) showed that exposing a variety of organisms, e.g. Paramecium, Colpidium, Arbacia eggs etc., to an excess of CO₂ decreased protoplasmic viscosity after a short exposure, but increased the viscosity following a longer (1 hr.) period. A similar phenomenon affecting cell permeability may occur in Gastrophilus. It would be of interest to investigate the influence of different, known concentrations of carbon dioxide on lysis of the tracheal cells and respiration.

It has been shown for various insects (e.g. Hazelhoff, 1927; Stahn, 1928; Wiggles-worth, 1935) that an increase in atmospheric CO₂ results in opening of the spiracles and hyperpnoea, while increasing the CO₂ tension beyond a certain threshold produces anaesthesia (Williams, 1946; Beadle & Beadle, 1949). This appears to be true only for species which are not adapted to high CO₂ tensions in their normal habitats. Thus, Kupka & Schaerffenberg (1947) found that several insect species (e.g. Melolontha spp., wireworm larvae) which live deep in the soil where the CO₂ concentration is high, were very resistant to gaseous CO₂ in which they could live for days, and showed little tendency to incur an O₂ debt; other species, which lived in the surface layers of the soil, reacted to higher CO₂ concentrations like the unadapted forms described above. The Gastrophilus larvae behave like the deep soil dwelling species in that they also survive for long periods in pure CO₂, and this gas does not induce either opening of the spiracles or anaesthesia. There can be little doubt, therefore, that the Gastrophilus larva is well adapted to the very high CO₂ tension in its environment (cf. Levenbook, 1950 a).

The respiratory rate of Gastrophilus measured over short (10 min.) periods shows considerable fluctuation even in the presence of CO₂. This is in agreement with the observations of Punt (1943, 1950), who found that for those insects which possess a spiracular closing mechanism CO₂ was emitted at irregular intervals of 15 –30 min., whereas Calliphora larvae which lack a closing mechanism gave off CO₂ continuously. It appears, although further work is necessary definitely to establish this fact, that Gastrophilus larvae can open the lips of their spiracular pouch and, by the play of muscular pressure transmitted through the haemolymph, fill the spacious tracheal trunks adjoining the spiracle with the external gas phase. The spiracular pouch is then closed, the oxygen in the enclosed air utilized and respiratory CO₂ given off, probably to a significant extent through the cuticle. This would account for the observations that when measured over short periods the respiratory quotient varies widely, frequently exceeding 1, and that larvae immersed in water will open their spiracles and give off bubbles of gas.

The physiological function of the spiracles in respiration is not well defined by the present experiments. As regards the paired anterior spiracles, each of these is situated in a deep furrow and leads into a long and dense felt chamber (Keilin, 1944). It may be, therefore, that the reason why these spiracles in the inflated larvae placed in a vacuum were air-tight was due to compression of the felt chamber by the excessive internal pressure. If gas were unable to penetrate the felt chamber, very little or no gas could pass through the spiracle. Furthermore, the O₂ uptake of larvae with ligatured posterior spiracles was very low, yet their CO₂ output was still considerable. This indicates that CO₂ was lost either through the anterior spiracles, or the cuticle, or through both. Nevertheless, when such larvae were placed in the vacuum they apparently did not lose CO₂ since they remained inflated ; this may be due either to the bulk of the dissolved tissue CO₂ having passed out through the cuticle at the time of evacuation, or perhaps to the molecular structure of the tightly stretched cuticle having become disorientated in such a way as to diminish its permeability to gases.

From the observation that (1) the respiration of inactive larvae with opened spiracles was considerably increased, whereas under similar conditions that of active larvae was only very slightly increased, and (2) that very high tensions of CO₂ do not induce opening of the spiracular pouches in normal larvae, it seems probable that these are opened in response to. O₂ lack and not to CO₂ excess.

The wide fluctuation in the respiratory quotient calls for comment. The R.Q. is generally assumed to be an indication of the predominant substrate being metabolized (e.g. Richardson, 1929) but, as pointed out, among others, by Soskine & Levine (1946), the R.Q. of a whole animal cannot per se be regarded as sufficient evidence for the predominant type of foodstuff being ‘burnt’.

The R.Q. in Gastrophilus is a combination of effects at physiological as well as biochemical levels. A range of experimental factors, which include removal of gaseous carbon dioxide, stimulation, the influence of bright illumination or various artificial treatments may, within a few hours, increase the R.Q. of fresh, diapausing larvae from the normal value of 0 ·85 –0 ·9 to over 1 ; subsequently, by changing the conditions, values approaching the original may once more be attained. Such variation can scarcely be attributed to qualitative differences in metabolism, but to the effect of these factors on the permeability of the cuticle to CO₂, on the frequency of opening or closing of the spiracular pouch or the liberation of CO₂ from the blood and tissues due to a decrease in pH. In connexion with the last-mentioned phenomenon it has been found (Levenbook, 1950 a, b)that bicarbonate was approximately equally distributed between the blood and tissues, and that the buffering capacity of the haemolymph was least at its normal pH. Furthermore, the blood lactic acid of larvae kept at 38° C. was much higher than of inactive larvae at o ° C. (Levenbook, 1950c). The conditions are therefore favourable for the formation by active larva of gaseous CO₂ from bicarbonate which would yield high values for the ‘respiratory’ CO₂ and hence the R.Q. At present it has not proved possible to distinguish such inorganic CO₂ from true respiratory CO₂.

A genuine change in metabolism as shown by an increase in the ‘metabolic’ R.Q. does, however, occur: this is seen when the larvae emerge from diapause, or when they are kept in vitro for a number of days. Kemnitz (1916) has shown that under the latter conditions glycogen is partially converted to fat—a reaction for which the R.Q. is > 1.

The action of inhibitors. In a detailed study of the effects of inhibitors on the eggs and embryos of the grasshopper Melanoplus differentialis, Bodine and co-workers (Bodine, 1934; Robbie, Boell & Bodine, 1938; Robbie, 1941) found that eggs in diapause were relatively cyanide insensitive compared with pre- or post-diapause eggs. Further, CO stimulated the respiration of diapausing eggs while inhibiting that of the developing egg (Bodine & Boell, 1934), and the latter inhibition was not light-reversible. It was concluded that the low respiration of eggs in diapause was not due to an enzyme system containing a heavy metal protein, whereas the higher respiratory rate during active development did involve a heavy metal protein, probably the cytochrome system.

The preliminary experiments reported in this paper indicate that, contrary to the diapausing Melanoplus egg, the respiration of the Gastrophilus larva during diapause does pass through a cyanide and CO sensitive heavy metal protein, probably, but not necessarily, cytochrome oxidase. The irreversibility of the CO inhibition by light may be due to the opacity of the larvae, since in a single experiment the respiration of a tracheal cell homogenate was found to be light-reversible.

There remains finally the interpretation of the lack of any inhibitory effects by sodium malonate. There is no doubt that in vivo the succinate in the blood is not oxidized to any appreciable extent by the succin-oxidase system of the tracheal organ cells, but in vitro, when the same cells are suspended in the insect’s haemolymph, the succinate is rapidly metabolized.^* However, even in vitro, malonate normally has hardly any effect, but there is a very pronounced inhibition when the cells are crushed. It is probable, therefore, that the deciding factor is the permeability of the tracheal cells to succinate and malonate. In this connexion, Keilin (unpublished) found that malonate injected into larvae with damaged tracheal cells (as evinced by lysis), did in fact inhibit respiration. This aspect of the work is being further investigated.

Due to its parasitic habitat, the physiology of the Gastrophilus larva bears certain resemblances to that of various helminths also parasitic in the mammalian alimentary tract. For the purpose of such a comparison, a discussion on the respiration of these parasites may be found in the works of Laser (1944), von Brand (1946), Hobson (1948) and Moulder (1950).

1. The respiration of the Gastrophilus larva has been investigated using three different manometric methods by which measurements may be made in the presence of carbon dioxide.

2. A new manometric apparatus of simple construction, designed for measurement of insect respiration in the presence or absence of CO₂, is described.

3. In the absence of CO₂ the respiration of Gastrophilus larvae progressively decreases and they eventually die. A manifestation of the CO₂ deficiency is an impaired cell permeability, resulting in leakage of haemoglobin from the tracheal cells into the blood.

4. The respiration of Gastrophilus pupae and Calliphora larvae was not affected by the absence of C0₂.

5. Both cyanide and carbon monoxide, but not sodium malonate, inhibited the respiration of Gastrophilus larvae during diapause.

6. The importance of allowing sufficient time for equilibration and settling down in the measurement of insect respiration is demonstrated.

I should like to thank Prof. D. Keilin, F.R.S., for his interest and advice in the present work. I am indebted to Dr H. Laser for carrying out all the experiments involving the use of the Laser-Rothschild manometer, and to the Agricultural Research Council for a maintenance grant.