A method is described by which the oxygen uptake of the blow-fly, Lucilia sericata Mg., was measured during flight manometrically in a Warburg and in a Barcroft type of apparatus.

The average oxygen consumption in air for all the flies used was 95·5800.0. per g. wet weight per hour. When flying in pure oxygen the rate of oxygen consumption showed no significant difference ; in oxygen-nitrogen mixtures, containing 10 and 5% oxygen, the rate was considerably less than in air.

In tables representing the rate of oxygen consumption of various animals, calculated per unit weight, by far the highest values are given by flying insects. Hitherto the most reliable determinations have been made with honey bees and several Lepidoptera, and values of 80-310 c.c. oxygen consumption per g. live weight per hour have been recorded. One of the chief difficulties that has been encountered by previous workers on this question has been to obtain insects which will fly continuously in experimental conditions in a confined space. Most previous observers worked with freely flying insects which would not fulfil this condition in a satisfactory manner. To overcome this difficulty we worked with insects flying in a fixed position, making use of the so-called flight reflex (Fraenkel, 1932). If an insect is suspended in mid-air in such a manner that there is no interference with the free movements of the wings, flight starts as a reflex action when the support is withdrawn from the legs and lasts until the legs regain contact with a solid object. In actual practice, however, flight only lasts for a limited period, varying for different species and individuals, coming to an end through fatigue or other causes. There are, nevertheless, several flies which may fly in the suspended condition for periods of up to 1 hr. After trying out several species, including Calliphora erythrocephala Mg., Phormia terranovae R.D., and Sarcophaga falculata Pand., the most satisfactory results were obtained with Lucilia sericata Mg., the sheep blow-fly, which was exclusively used in the experiments now reported. When this fly is suspended in the following manner it will fly continuously for periods of 20 min. or even longer if it is in good condition. A small strip of paper to which is attached a piece of very thin wire bent into a hook is fixed to the dorsal surface of the thorax of the lightly etherized fly with a patent quick-drying adhesive made from amyl acetate and cellulose in such a position that it does not interfere with the movements of the wings. Flies thus prepared lived for the normal period after this treatment if fed on sugar. After each experiment the fly was killed, weighed after removal of the suspension and dried to constancy of weight at 100° C. The water content is usually about 70 %, although figures as low as 60 % and as high as 75 % were occasionally found, these discrepancies most probably being due to different states of nutrition.

The apparatus used in the first series of experiments was a modified Warburg respirometer, and in the second series a modified Barcroft respirometer. Both pieces of apparatus had a graduated side tube attached to them, similar to that described by Dixon (1934, p. 8), so that readings could be made under constant pressure. The bottles had a capacity of about 60 c.c., and the fly was suspended from the end of a tube connecting the interior with the outside air. The bottom of the bottles contained caustic potash as CO2 absorbing agent.

The experiments were performed in a room kept at 27° C. by thermostatic control in which the apparatus and the gas mixtures used in later experiments were continuously kept so that only very short equilibration periods were necessary. Each insect suspended in the apparatus was allowed to fly as long as it would, and the oxygen uptake was measured at intervals timed with a stopwatch. These intervals were intended to be 5 min., but as the flies would sometimes stop and start again the times of actual flight had to be measured accurately.

The consumption of oxygen was calculated both in terms of c.c. 02 per g. dry weight per hour, and in c.c. 08 per g. wet weight per hour. None of these figures for oxygen uptake were reduced to N.T.P. Calculations based on wet weight and on dry weight seemed to have no significant advantage over each other. The flies may have drunk water or fed on sugar just before the experiment, while others would secrete a drop of fluid from the mouth during the experiment or even defaecate. All these factors influence the water content and necessarily account for much of the variation found in different experiments. Only those figures were included in the tables where one fly gave more than one reading, the similarity of the values of several readings being regarded as a check against possible experimental errors.

Table I represents the results of a series of experiments carried out with the modified Warburg apparatus (only one chamber, manometer open at other end). It can be seen that the oxygen consumption of an individual fly remains fairly constant during subsequent periods. The variation of the values for different individuals are not considered to be unduly high in view of the unavoidable variations in dry weight and wet weight on which the calculations are based, and the individual differences in oxygen consumption which are well-known phenomena, particularly in mobile poikilothermic animals.

Table I.

Oxygen consumption of flies during flight in air

Oxygen consumption of flies during flight in air
Oxygen consumption of flies during flight in air

In the second series of experiments (Table II) a modified Barcroft-type apparatus was used, because it has a compensating vessel which corrects equilibrium effects caused by temperature changes. In cases 1-6 the respiration was measured first in air and then in pure oxygen. In three cases (1, 2, 3) the oxygen consumption was practically the same in air and oxygen; in three other cases (4, 5, 6) it was only slightly higher in oxygen than in air. It therefore seems that an oxygen partial pressure of 21 % is sufficient to cover the considerable requirements during flight.

Table II.

Oxygen consumption of flies during flight in air, in pure oxygen, and in mixtures containing 10 and 5 % oxygen

Oxygen consumption of flies during flight in air, in pure oxygen, and in mixtures containing 10 and 5 % oxygen
Oxygen consumption of flies during flight in air, in pure oxygen, and in mixtures containing 10 and 5 % oxygen

Finally, an attempt was made to determine whether and to what extent lowering of the oxygen tension would render the oxygen consumption during flight dependent on the oxygen tension. In an oxygen-nitrogen mixture, containing 10% oxygen, the oxygen consumption becomes considerably less than in air (fly 7-9). In a mixture containing only 5 % oxygen only a few flies would fly. The oxygen consumption is still less than in 10% oxygen (fly 10, 11).

Table III contains a summary of the previous work done on the respiration of insects during flight. As can be seen from column 2, the principal method applied was to keep flying insects in a confined space and analyse the air for changes in oxygen and/or CO2 content. Parhon (1909) and Tauchert (1930) were not dealing with the condition of continuous flight, and the work of Raffy & Portier (1931) was concerned with butterflies injected with nicotine which did not fly but vibrated the wings convulsively. In the only other work in which a manometric method was used, the authors (Kosmin et al. 1932) give no indication of how the experiments were arranged, and it is therefore impossible to discuss the results, which, compared with others, seem to give much too high values. Considering the differences of methods and material and the inherent difficulties of the experiment, the results of different authors seem remarkably similar. The average figures given for the bee by Jongbloed & Wiersma (1934) and for the fly by us are almost identical. The comparatively low values given for the hawk moth (Kalmus, 1929) can be explained by the much larger size and the smaller frequency of the wing beat of this moth, compared with bees and flies. The rate of the wing beat of two similar Sphingids, Acherontia átropos and Macroglossa bombyliformis, is 22 and 80 per sec. respectively; bees beat their wings at a frequency of 250 per sec. approximately (Magnan, 1934). We determined the rate of the wing beat of Lucilia sericata to be 160 per sec. approximately (measured acoustically with the aid of a microphone, an oscillograph and an oscillator).

Table III.

Oxygen consumption of flying insects

Oxygen consumption of flying insects
Oxygen consumption of flying insects

It can therefore be stated that an insect of the size of a fly or a bee, which vibrates the wings rapidly, consumes oxygen at a rate of approximately 100 c.c. per g. live weight per hour. It is difficult to compare this value with that for the basal metabolism. This has never been measured in active adult insects like bees and flies. The basal metabolism of blow-fly larvae is about 0-5 c.c. per g. live weight per hour (Fraenkel & Herford, 1938), and the oxygen uptake of adult blow-flies, which were not at rest, was determined as 2-3 c.c. per g. live weight per hour. It therefore seems that the ratio between the rest and the flight metabolism is in the region of 1 : 100. Flying insects maintain these extremely high rates for some time ; a blow-fly may sometimes fly for more than 30 min. In man, during extreme muscular activity, the rate of metabolism increases by ten to fourteen times the normal resting exchange, but this cannot be maintained for more than a few minutes (Starling, 1936).

From our observations and those of other authors it seems reasonable to assume that the oxygen consumed during flight derives from carbohydrate metabolism. From the equation C6H12O6 + 6O2 = 6CO2 + 6H2O it follows that 6×22·41. of oxygen are required for 1 g.mol. carbohydrate, i.e. 180 g. Hence, a fly with an average wet weight of 31 mg. consuming oxygen at the rate of 95-58 c.c. per g. per hour would consume in 1 hr. 4 mg. of sugar. It was not found practicable to weigh flies before and after the experiment for determining the loss of weight during flight, because flies very often vomit or defaecate during the experiment, and the periods of flight are often too short for accurate determination of loss of weight to be made. A series of determinations was, however, made to find out the amount of sugar which one fly may take up in a single meal. Flies were placed on a piece of cane sugar of known weight and the weight again determined after feeding. The results in four cases were 4·8, 5·1, 4·4 and 2·3 mg. It therefore seems that a well-fed fly carries sufficient sugar to allow flight for a period of 1 hr.

It was observed in some of the determinations that the oxygen consumption in subsequent readings fell off. It is difficult to say whether this was caused by depletion of carbohydrate stores in the flies, which were obviously in different experiments in a different state of feeding, or by gradual decrease in oxygen tension in the small vessel. This falling off in the oxygen consumption was often very marked between the first and second reading. This phenomenon, which has often been noticed in the measurement of the respiration of cold-blooded animals, is due to the fact that animals are stimulated at first after being placed in a respiratory chamber and later “settle down”. This effect in our experiments was certainly not a temperature equilibration phenomenon for the following reasons: rapid wing vibration warms the body of the fly and thus the surrounding air also; the air in the vessel would then expand and the reading would be lower instead of higher. After a time a new equilibrium would be reached and the effect would die away. Again, since the readings were regular and consistent on the whole, it seems unlikely that temperature changes influenced them. Also, the Warburg method seemed just as reliable as the Barcroft method in which there is a compensating vessel.

One may make the comment on flying insects in general that nearly all those which normally vibrate their wings rapidly like bees, wasps, flies, mosquitoes, Sphingids, and other Lepidoptera, feed on carbohydrates in the form of nectar. A notable exception to this rule are tse-tse flies which feed exclusively on blood. While the amount of blood sugar is certainly too low for sustaining enduring flight (approx, 0·1 g. %), it can be assumed that blood contains other food substances in an easily assimilable form. Humming birds, which are very insect-like in their habits and which vibrate their wings very rapidly and on account of their small sizç certainly have a very high metabolism, also feed mainly on carbohydrates.

Dixon
,
M.
(
1934
).
Manometric Methods
.
Cambridge
.
Fraenkel
,
G.
(
1932
).
Z. vergl. Physiol
.
16
,
371
93
.
Fraenkel
,
G.
&
Herford
,
G. V. B.
(
1938
).
J. exp. Biol
.
15
,
266
80
.
Jongbloed
,
J.
&
Wiersma
,
C. A. G.
(
1934
).
Z. vergl. Physiol
.
21
,
519
33
.
Kalmus
,
H.
(
1929
).
Z. vergl. Physiol
.
10
,
445
55
.
Kosmin
,
N. P.
,
Alpatov
,
W. W.
&
Resnitschenko
,
M. S.
(
1932
).
Z. vergl. Physiol
.
17
,
408
22
.
Krogh
,
A.
(
1919
).
Z. allg. Physiol
.
16
,
9
15
.
Magnan
,
A.
(
1934
).
Le vol des insectes
.
Paris
.
Parhon
,
M.
(
1909
).
Ann. Sci. nat. (Zool.)
,
9
,
1
58
.
Raffy
,
A.
&
Portier
,
P.
(
1931
).
C. R. Soc. Biol., Paris
,
108
,
1062
3
.
Starling
(
1936
).
Principles of Human Physiology
, 7th ed.
London
.
Tauchert
,
F.
(
1930
).
Z. Biol
.
89
,
541
6
.