Wingbeat Amplitude Research Articles

Course-Control, Metabolism and Wing Interference During Ultralong Tethered Flight in Drosophila Melanogaster

ABSTRACT Tethered flight in a 3-day-old female Drosophila was sustained for 32·2 h with only short interruptions during uptake of sucrose solution. The course-control reactions derived from the difference of the wingbeat amplitudes on either side have been used to simulate the rotatory displacement of the surrounding landmarks during a comparable turn in free flight. Stabilization of a target in the preferred area of the visual field requires continuous visual attention. A rate of about 5 course-correcting manoeuvres per second was maintained throughout the experiment. Drosophila seems to be able to cover long distances in search of a favourable habitat. Flight-specific carbohydrate consumption is equivalent to a metabolic power input per body weight of about 18 W N−1. The tethered fly produces about 40 % of the lift required to sustain hovering flight. The resulting mechanochemical efficiency of about 0·04-0·07 is within the expected order of magnitude for flying insects. Expenditure of reserve substances may account for the difference between the comparatively low power input of about 7 W N−1, derived from carbohydrate uptake in the first hours of flight (Wigglesworth, 1949), and the actual metabolic turnover of about 21 WN−1, derived from oxygen consumption during this period (Laurie-Ahlberg et al. 1985). Weis-Fogh’s ‘clap and fling’, a widespread lift-generating process exploiting the aerodynamic wing interference at the dorsal end of the wingbeat, was in action throughout the flight. However, there were two significant modifications (as first conceived by Ellington, 1980) : (1) during ‘clap’, there is a progress of wing contact from the leading to the trailing edge, which is likely to ‘squeeze’ a thrust-generating jet of air to the rear; (2) during ‘fling’, there is a progress of wing separation in the same direction, which is described as a ‘peel’ resembling the progressive separation of two plastic foils pulled apart against forces of mutual attraction. The wings of the test fly survived about 23 million such peels without damage. Increasing airspeed decreases the intensity of ‘clap and fling’ in Drosophila : results obtained in the wind tunnel show the transition to a ‘near clap and fling’, lacking mutual wing contact.

Compensatory head roll in the blowflyCalliphoraduring flight

Video records were made of the blowflyCalliphora erythrocephalaL. mainly during tethered flight in a wind-tunnel, to study its movements about the longitudinal body axis (roll). During undisturbed flight, flies hold their head on average aligned with the body but may turn it about all three body axes. Pitch and yaw turns of the head are comparatively small (20°), whereas roll turns can be large (90°), and fast (1200° s-1). When passively rolled, flies produce compensatory head movements during walking or flight; at rest this reflex is turned off. Flies perceive a static misalignment relative to the vertical, as well as roll motion up to 10000° s-1. Within this range flies counteract an imposed roll with maximal gain at about 1000° s-1. Compensatory head movements are made with very low latency (down to ∆t ≈ 5ms), and with considerable speed (up to ω = 1000° s-1). Flies may ‘disregard’ an apparent deviation from their correct orientation, and may superimpose spontaneous head movements on those elicited by a stimulus. Compensatory head movements generally undercompensate the imposed misalignment. Simultaneously, however, flies modify their wing pitch and wingbeat amplitude to produce a compensatory roll torque. Since head and body roll act simultaneously and in the same direction, the overall speed and degree of head realignment, relative to external coordinates, increase considerably. This is certainly an advantage for flight in turbulent air. In still air, without need to correct an imposed misalignment, flies nevertheless produce spontaneous fluctuations of their flight torque, and head roll movements in the opposite direction. This is to be expected if flies intend to keep their eyes aligned with the coordinates of the environment while spontaneously performing banked turns. The limits of fly vision and the advantages of compensatory head movements for different visually guided behaviour are discussed. Compensatory head roll movements give flies greater manoeuvrability when cruising than the visual system would allow, without such a stabilizing reflex.

Variability of motor output during flight steering in locusts

1. A flight balance has been used to demonstrate the generation of torques by a flying locust,Locusta migratoria, in response to roll movements of an artificial horizon. 2. The movements of the horizon can be perceived by either the compound eyes or the ocelli alone and lead to roll responses from the animal. 3. Amputation experiments show that either fore- or hindwings alone are able to produce a roll response of appropriate sign. 4. EMGs have been recorded from the flight muscles. Shifts in the relative latency between homologous pairs of depressor and elevator muscles are frequently correlated with roll responses. The directions of these shifts are opposite in the fore- and hindwings. The relative latencies of both elevators and depressors of a given wing shift in parallel. 5. No systematic changes in burst length correlated with roll responses are seen. This may be a consequence of the experimental conditions. 6. Roll torques can be developed without changes in the relative firing times of basalar (pronator) and subalar (supinator) muscles. 7. Changes in wingbeat amplitude are not always systematically correlated with roll responses. 8. Both fore- and hindwings beat out of phase during roll responses, a wing being advanced or retarded in its cycle in comparison to its contralateral homologue. As seen for latency shifts, the direction of these phase shifts is reversed between fore- and hindwings. The phase shift of the wing is probably the main result of the latency shifts of the flight muscles. 9. The motor output alterations which occur during roll responses of flying locusts are variable. The magnitude of the latency shifts can vary, changes in burst length are not indispensable, and neither rudderlike movements of the abdomen and the hindlegs nor changes in the wingbeat amplitude are necessarily correlated with torques. The flying locust is in every case producing a roll response.

Movement as a specific stimulus for prey catching behaviour in rhinolophid and hipposiderid bats

1. The echolocating ‘long CF/FM-bat’Rhinolophus rouxi and the ‘short CF/FM-bats’Hipposideros bicolor andHipposideros speoris were tested for catching responses to moving and non-moving targets. 2. Under our experimental conditions (freshly caught caged bats in a natural environment)Rhinolophus rouxi andHipposideros speoris only responded to insects of any sort that were beating their wings. The bats showed no reactions whatsoever to nonmoving insects or those walking on the floor or the sides of the cage. 3. Hipposideros bicolor responded in the same way as the above species to wingbeating insects but in addition also attacked walking insects. In 27 presentations 15 walking insects were caught (Fig. 2). 4. Rhinolophus rouxi, Hipposideros speoris andHipposideros bicolor also detected, approached and seized tethered cockroaches hanging from the ceiling when these were vibrating up and down (Fig. 3). This indicates that any oscillating movement and not specific aspects of wing beating were the key releasers for catching behaviour in all three species. However, a wing beating insect is strongly preferred over a vibrating one in all three species (Fig. 4). 5. Rhinolophus rouxi, Hipposideros speoris andHipposideros bicolor attacked and seized a dead bait when it was associated with a wing beating device (Fig. 1). All three species responded effectively to beat frequencies as low as 10 beats/s (peak-to-peak amplitude of the wing excursion 20 mm). For lower frequencies the response rates rapidly deteriorated (Fig. 5). 6. Horseshoe bats no longer responded to wing beats of 5 beats/s when the wing beat amplitude was 2 to 1 mm or to wing beats of 2 to 1 beats/s when the amplitude was 3 mm or lower (Fig. 6). This suggests that the speed of the wing is a critical parameter. From these data we infer that the threshold for the catching responses is at a wing speed of about 2 to 1 cm/s. 7. In horseshoe bats (experimental tests) and the two hipposiderid species (behavioural observations) one single wing beat was enough to elicit a catching response (Fig. 8). 8. It is concluded that ‘long’ and ‘short’ CF/ FM-bats feature a similar responsiveness to fluttering targets. The sensitivity to oscillating movements is considered as an effective detection mechanism for any sort of potential prey.

How Diptera move their wings: a re-examination of the wing base articulation and muscle systems concerned with flight

The accepted mechanical model for dipteran wing movements is not compatible with the observations of unanaesthetized flies in tethered flight. In this paper a new model is proposed together with previously undescribed morphological features in support of it. A direct coupling is demonstrated between the indirect flight muscles, consisting of a wing base stop mechanism and a locking of the first axillary sclerite onto the parascutal shelf during the downstroke. We propose that wingbeat amplitude is controlled by axillary muscles which alter the downward force on the wing. We also describe the anatomical basis of the automatic changes in the wing stroke during the wingbeat. These are shown to be a direct consequence of the attachment between the fourth axillary sclerite and the scutellar lever arm. By virtue of their attachments to the sclerite, the fourth axillary muscles are shown to have potential control over the tonic characteristics of these automatic changes. The calypters are connected to the third and fourth axillary sclerites and this suggests an aerodynamic function for these structures which has not previously been proposed. A detailed analysis of the anatomy of the non-fibrillar muscles has allowed a complete functional characterization to be made which is consistent with our new model. The functions given to many of the muscles in this analysis are supported by physiological evidence from other studies. Muscular control of a switch mechanism comprising the radial stop and pleural wing process is discussed, as is the control of a wing base proprioceptor.

Optomotor control of the force of flight in Drosophila and Musca

Drift of the retinal images of the surroundings elicits optomotor responses of flight control in the fruitfly, Drosophila melanogaster, and in the housefly, Musca domestica. The present investigation deals with the responses of tethered flies in still air. The responses were elicited by continuous movement of striped patterns in front of the eyes, and characterized by the magnitude and elevation of the resulting force of flight which is the average of the forces produced during a wingbeat cycle. The force of flight is resolved into the upward directed lift and the forward directed thrust. In either species, pattern movement acts upon the magnitude, but not upon the elevation of the force of flight. The elevation relative to the longitudinal body axis is almost invariably 24° in Drosophila, and 29° in Musca. The lift/thrust ratio in still air is fixed accordingly, and can be changed only by variation of the body angle. Keeping an angle of minimum body drag does not contribute significantly to the efficiency of insect flight at very low Reynolds numbers (Re). Control of the lift/thrust ratio by variation of the body angle is, therefore, less surprising in Drosophila where Re is in the order of 102, than in Musca, where Re is in the order of 103. Control of this ratio without variation of the body angle is actually established in insects flying at even higher Re. Covariance of lift and thrust in the investigated flies is achieved by control of wingbeat amplitude or wingbeat frequency, but not by control of wing pitch or stroke plane. A change in the latter parameters would have deflected the force of flight and is, therefore, inconsistent with the constant elevation found in the present experiments. The results obtained, so far, do not exclude active deflections of the force vector during occasional bouts of aerobatics, or passive deflections of this vector during flight at non-zero airspeed.

Correlations between stationary measurable parameters of wing movement and aerodynamic force production in the blowfly (Calliphora vicina R.-D.)

With the aid of a mechanical aerodynamic two component balance, stroboscopes and photographic projections, the following dynamic and kinematic parameters were determined from short periods of stationary flight recorded in tethered blowflies which flew for several hours in a wind tunnel: weight, lift, flight speed, wing-beat frequency, nine angles characteristic for right and left wing-beat positions and a value characterizing the angle of the stroke plane. The data were subjected to a correlation analysis; the total correlation coefficients and the first order partial correlation coefficients were calculated with a significance of 5% and 1% respectively. The correlation between wing-beat frequency and lift was found to be strongly positive. Frequency, lift and flight speed are positively correlated with the sum of the wing-beat angles. Although under the given experimental conditions a significant tendency towards compatible adjustment has been demonstrated lift and thrust can be controlled independently in individual cases. Negative correlations of wing-beat angles of the right and left wing are generally caused by shifting the lower turning points and indicate a neuromotoric counter action of the two wings during turning reactions. A decreasing wing-beat amplitude is often coupled with an increase of the mean stroke plane angle. Numerous other correlations and detailed descriptions are given and compared to functional morphological and neuromotoric data.

The tethered flight performance of a laboratory population of Triatoma infestans (Klug) (Hemiptera: Reduviidae)

AbstractThe flight performance of a laboratory population of Triatoma infestans (Klug) was tested on a flight balance. Bugs adopted a typical flight posture which is described. They were capable of steady flight and produced reasonable amounts of lift. Flight durations were generally short, but capacity for flight rose to a peak in the third week after adult eclosion with the longest recorded flight being one of 2 h 40 min by a male. Sexual differences were slight; males had a slightly higher mean wing-beat frequency (57·8 Hz against 55·6 Hz), and a few more females than males made longer flights. Differences were noted between short and long fliers; the latter producing significantly more lift and showing signs of a flight pattern divisible into rising, steady and falling phases. The short fliers showed only rising and falling phases. Lift and wing-beat frequency were correlated, but it is evident that lift also depends on other variables such as stroke-plane angle, body angle and wing-beat amplitude, which are discussed.

Steering responses of adult and nymphal crickets to light, with special reference to the head rolling movement

Most tethered adult crickets ( Gryllus bimaculatus) assumed flight postures with or without flapping their wings in a windstream. Nymphal crickets (sixth and seventh, i.e. final, instars) also displayed the flight posture in spite of the incompleteness of wing development. These adult nymphal crickets rolled their heads towards the light source in response to unequal illumination of the compound eyes only while maintaining the flight posture. The amphtude of the head rolling movements was proportional to the change of light position up to 120°C, and independent of the light intensity if the duration was longer than 1 sec. The unequal illumination could also induce a transient increase in discharge frequency of the wing muscles on both sides, a decrease in wing beat amplitude of the ipsilateral wing on the illuminated side, and bending movements of the legs and abdomen towards the light. Cutting either of the nerve connectives at any level between the subosophageal and metathoracic ganglia did not affect the response of either the head or the abdomen to illumination. These results are discussed in relation to the steering mechanism associated with the dorsal light reaction.

Wing-beat pattern in Rhodnius prolixus Stål (Heteroptera: Reduviidae) during exhaustive flight

Stationary, tethered Rhodnius prolixus males were stimulated to beat their wings for periods averaging 2–3 h. Wing-beat frequency averaged 70 Hz and was quite variable. Wing-beat amplitude declined as the insects approached exhaustion, but no consistent change was observed in beat frequency. Male R. prolixus assumes a characteristic flight posture and exhibits the typical reflexes and anatomical adaptations observed in other flying insects. The long periods of wing beating under laboratory conditions suggest they may be capable of flights of considerable duration in the wild.

The metabolic cost of flight in unrestrained birds.

Oxygen consumption and carbon dioxide production were measured during flight in unrestrained starlings by a new method. Mean RQ after the first 30 min of flight was 0.69 +/- 0.08 (+/- S.D.). Mean rate of carbon dioxide production was 19.7 +/- 2.2 ml CO2/min, which corresponds to a metabolic rate of 8.9 +/- 1 W. Metabolic rate during flight did not change significantly over a range of air speeds from 8 to 18 m/s and birds would not fly at speeds outside of this range. Current theories of bird flight predict a large change in metabolic rate over the same range of speeds. Wingbeat frequency was constant at 12 +/- 0.5 Hz. Wingbeat amplitude reached a minimum at a speed of 14 m/s and increased at both higher and lower speeds. Angle between the body and horizontal was least at high speeds and increased at low speeds. As existing theories do not take into account the change of drag resulting from changes in body attitude, this may be a cause of the discrepancies between theory and observation.

A visually-evoked roll response in the housefly

1. A flying, tethered housefly exhibits a turning response about its long axis when presented with a visual stimulus consisting of a rotating radial grating. This roll response is isometrically measured as a torque (Fig. 2). 2. Comparison of the dynamic characteristics of the roll response with those of the classical optomotor response reveals that similar neuronal mechanisms mediate visual stabilization of roll and yaw motions. 3. The changes of wingbeat amplitude that are elicited by a rotating radial grating are photographically measured and compared to those elicited by a vertically-moving, linear grating (Fig. 4). This data reveals that the flight mechanism generates a roll torque by creating an asymmetry in the lift-forces produced by the two wings. 4. The manner in which a fly detects and responds to roll is consistent with a scheme in which the lift produced by a given wing is controlled by the vertical component of motion seen by the ipsilateral eye (Table 1). 5. Under closed-loop conditions the flying, tethered fly exhibits a strong tendency to orient visual patterns consisting of a stripe or a linear grating along a direction perpendicular to the equatorial plane of the eyes (Figs. 5, 6).

Temperature regulation of the sphinx moth, Manduca sexta. I. Flight energetics and body temperature during free and tethered flight.

ABSTRACT The sphinx moth, Manduca sexta, maintained an average thoracic temperature of 40–42°C during free flight in ambient temperatures (TA) of about 16·33°C. In the extremes, the excess of thoracic temperature TTh over TA varied from a mean of 25°C at 12·5°C, to a mean of 8°C at a TA of 35°C. During tethered flight TTh increased directly with TA, and the excess of TTh over TA varied from about 11-4°C. The oxygen consumption was about 45-50 c.c. Oa/g h during free flight from ambient temperatures of 15 –30°C. During captive flight the oxygen consumption was about 21 c.c. O2/g h. The wing-beat frequency and amplitude during both free flight and captive flight did not vary significantly with TA. The wing-beat frequency was about the same during free flight and captive flight but the wing-beat amplitude was significantly less in the latter. The moths showed little variation of flight speed with respect to TA on the flight mill. The difference between TTh and TA was strongly correlated with flight speed at low, but not at high, TA. The cooling rate of dead moths was only slightly correlated with air speeds from 2 to 5 m/s. The cooling rate of thoraces without scales was 2·4 times as great as with scales intact at an air flow 2 m/s, but the cooling rate of the abdomen was only slightly increased after the removal of its scales. The data suggest that the rate of metabolism during flight is altered with regard to the flight effort, but not with regard to temperature-regulation. Heat is actively dissipated from the thorax during flight at high TA, or during fast flight when TTh reaches 40°C or above.

Neuro-muscular control of dipteran flight.

ABSTRACT Electrical activity from the indirect, myogenic muscles of calliphorid flies was recorded during flight. The animals were suspended from an aerodynamic balance in the laminar air-stream from a wind-tunnel. Muscle action potentials, recorded with 25 μ wire, were 5−7 msec, in duration, up to 10 mV. in amplitude and positive in sign. Frequencies were mostly under 20/sec. Frequencies in all the indirect muscles were similar, but these varied together with changes in aerodynamic power. Frequencies in the indirect muscles of the two sides varied by no more than ±10% during extreme turns to right or left (only left or only right wing beating). Electrical records from the non-myogenic direct muscles were made during tethered flight. The potentials were 2−4 msec, in duration, up to 2 mV. positive and had frequencies up to 180/sec. A nearly linear positive correlation exists between impulse frequency in the musculus latus (pleurostemal muscle), the inward movement of the pleural wall, and the wingbeat frequency, suggesting that this muscle is the basic frequency determiner. Strong turning behaviour is associated with opposed frequency changes in the pairs of antagonistic adductor and abductor muscles of the wings on the two sides of the body. The musculus dorsoventralis IV (tergo-trochanteral) is activated by a short impulse burst at the beginning of flight. It apparently acts as an oscillation starter. Flight initiation normally requires 30−60 msec. Usually activity begins in the musculus latus, which stiffens the thorax. Then simultaneously the myogenic muscles are activated and the ‘starter’ muscle causes a jump and the beginning of oscillation of the thorax. Then the wings are drawn gradually forward and full wingbeat amplitude develops within the first six wingbeats. Flight begins with maximal lift and wingbeat frequency and a nearly synchronous burst discharge in all the indirect muscles. Power production and the transmission and distribution of power are under separate control. The myogenic indirect motor varies only in total power output, this being influenced by its own state of excitation and by a muscle-controlling wingbeat frequency. Steering is accomplished by non-myogenic direct muscles which are capable of differentially engaging the two wings with the motor.