According to the classical investigation of Boettiger & Furshpan (1952), the wing beat of flies is accompanied by a bistable mechanism, the 'click mechanism'.
J. exp. Biol. 12S, 463^t68 (1987) Printed in Great Britain @ The Company of Biologists Limited 1987
463
SHORT COMMUNICATION CRITICAL COMMENTS ON A 'NOVEL MECHANICAL MODEL OF DIPTERAN FLIGHT' (MIYAN & EWING, 1985) BY HANS K. PFAU Institut fur Zoologie der Universitat, 6500 Mainz, FRG Accepted 2 October 1986 According to the classical investigation of Boettiger & Furshpan (1952), the wing beat of flies is accompanied by a bistable mechanism, the 'click mechanism'. This mechanism is controlled by the pleurosternal and tergopleural muscles (see Heide, 1971; Pfau, 1978; Pfau, Schroeter & Steitz, 1977). Recently doubts have been raised concerning the existence of the click mechanism during flight (Miyan & Ewing, 1985a,b). The authors suspect that the click mechanism is an artefact of CC14 anaesthesia and propose a novel mechanical model of dipteran flight. In the following I analyse this model critically. The classical model The click mechanism was illustrated by Boettiger & Furshpan (1952, fig. 3) by diagrammatic cross-section drawings. Pringle (1957) presented a simplified threedimensional working model which made the mechanism easier to understand, demonstrating that the thorax of flies is, in principle, a four-bar crank chain (Alexander, 1968, p. 52). According to my studies this classical model is essentially correct, with the exception of the mechanism of pronation and supination: Boettiger & Furshpan and Pringle state that these movements of the wing are achieved directly ('automatically') by the scutellar lever; according to Pfau (1977) pronation and supination are controlled by the muscles of a special mechanical system, mainly consisting of axillaries 3 and 4. A different model of a wing actuation lever system, which can be traced back to 1822 (Chabrier, 1822, cited in Pringle, 1957, p. 10), has been named 'Deckel-TopfPrinzip' (Nachtigall, 1968). This popular model describes well the wing lever system of Odonata (Pfau, 1986), but it is not compatible with the four-bar crank chain of the Diptera. The 'novel mechanical model' According to Miyan & Ewing (1985a), a 'lid-saucepan-mechanism' exists in a certain part of the wing beat of flies. Here, the side part of the scutum (parascutum) is used to form a unit with the axillary 1 lever: the authors state that in the middle of Keywords: Diptera,flight,mechanism.
464
H. K. PFAU
the downstroke the axillary 1 'locks onto the parascutum'; during the rest of the downstroke this unified parascutum-axillary 1 lever will force the scutum upwards, stretching the dorsoventral muscles directly. The locking of axillary 1 onto the parascutum is interpreted as a structural impediment to the 'click' (for a second one see below). I, however, found that there is no wing position in which the axillary 1 can lock onto the parascutum during the downstroke. If one examines cross-sectional preparations, the medially protruding process of axillary 1 (onto which the muscles that are able to adjust the downstroke velocity and amplitude are attached; Pfau, 1977) may seem to produce such a locking. However, during the whole downstroke this process moves freely in a lateral membraneous indentation of the scutellar lever and the parascutum. In addition, the motion of the parascutum can easily be observed to take place in a manner quite different from Miyan & Ewing's fig. 6: the parascutum is rotated clearly upwards around its proximal joint by the upward movement of the scutellar lever. According to Miyan & Ewing, shortly after the alleged locking of axillary 1 onto the parascutum ('a few degrees below the horizontal position on the downstroke') another event takes place. A process on the underside of the wing ('radial stop') crashes against the crest of the pleural wing process ('wing base stop mechanism'). From then on the mechanical principle of the downstroke is said to change again: the muscles of axillary 3 now bend the wing downwards distally of its stroke hinge ('elastic energy storage'), performing the final part of the downstroke and leading to a 'dramatic acceleration' at the start of the upstroke. The 'wing base stop mechanism' should, according to fig. 6, block the wing base completely. This would also make a further stretching of the dorsoventral muscles impossible. However, the authors seem to assume that the dorsoventral muscles are still stretched in the second half of the downstroke. To solve this problem they postulate that 'the actual pivot for wing movement switches from the base of axillary 2 to the crest of the pleural wing process' (see legend to fig. 6). The proposed 'wing base stop mechanism' will be disengaged when the pleurosternal muscles are 'superstimulated'. Only in this 'abnormal' situation is the click mechanism said to be possible. Thus Miyan & Ewing conclude that the 'click' is an abnormal effect which might be caused by an anaesthesia artefact. A switch of the wing pivot as postulated by Miyan & Ewing can in fact only occur if the axillary 2 and pleurum could be separated. The axillary 2 would then have to move upwards. However, this is impossible since the axillary 2 does not have any clearance in the upward direction. In addition, it is difficult to understand why such drastic changes in wing hinge forces as must occur in Miyan & Ewing's model should not appear as speed changes in their high speed film. This is even more illogical since the authors postulate that a lack of drastic speed changes in the wing movement - as well as the observed asymmetry between up- and downstroke - is proof of the nonexistence of the click mechanism. The click forces, however, are certainly not the only forces acting on the wing. Apart from the inertial forces, the aerodynamic forces are of prime importance. They
Dipteran
flight
465
can be influenced decisively by the angle of attack of the wing, which is reported to be adjustable for up- and downstroke independently (Pfau, 1977). The forces acting through contraction and elasticity of muscles must also influence the wing motion independently on the up- and downstroke (Pfau, 1977: muscles of axillary 1, 3 and 4). Additionally, a resilin element in the apodeme of the axillary 3 muscles will lead to a certain asymmetry in the wing motion, since its strongest force is produced at the lower turning point of the wing beat (see below). All these forces may be so strong that the outside effects of the 'click' are blurred. Moreover, the strength of the pleurosternal and tergopleural muscles is variable during flight. Any experiment trying to prove or disprove the click mechanism should therefore monitor the state of contraction of these muscles. The belief that the click mechanism is an artefact ('superstimulation' of the pleurosternal muscles) can be disproved easily: even if the pleurosternal muscles are completely cut, in most cases of CC^-killed animals a vestige of the click effect can be demonstrated. This is caused by the tergopleural muscles. These have a much weaker effect on the 'click' and are certainly not capable of forcing the system into an 'abnormal state by excessive tension'. If the tergopleural muscles are also cut, the 'click' disappears (Pfau, 1978; Pfau et al. 1977). The axillary 3 muscles are interpreted by Miyan & Ewing to be synergists of the downstroke muscles (Miyan & Ewing, 1985a, and 19856, p. 289). However, the tendon of these muscles is actually stretched during the downstroke on account of the arc of movement of its distal attachment point. The axillary 3 muscles are, in fact, the only existing muscles of pronation, pronating the wing relative to the wing base, i.e. without rotating the radial stop itself (see below). Concerning the pronation— supination movements, Miyan & Ewing suppose erroneously, that — besides a pronation-supination via the scutellar lever plus axillary 4 - the two muscles of the axillary 1 are responsible (p. 287). The 'gear-change mechanism' (Pfau, 1973, 1985) A different interpretation of a possible contact between the 'radial stop' (structure 'A') and the pleural wing process (structure 'B', with one or two grooves in the upper side) has been described by Pfau (1973; 'gear-change mechanism'); this hypothesis has been extended as a means for producing effective pronation movements at the upper turning point of the wing beat (Pfau, 1985). According to these results structure A contacts B only if the two muscles adjusting the click mechanism (pleurosternal and tergopleural muscle) are contracted forcefully. In this case A slides, at the upper end of the upstroke, briefly into one of the two grooves of B and leaves it at the onset of the downstroke. In this brief moment of contact, activity of the axillary 3 muscles may produce a very efficient wing pronation since further function of these muscles (backward folding of the wing) is blocked by A and B. The tergopleural muscle is able to 'shift' between the two grooves (Bl, B2) of the fulcrum as soon as A has left B. Since this muscle has a simultaneous effect on the strength of |he 'click', the click mechanism is increased while the gear is shifted from no contact
466
H, K. PFAU
to Bl contact ('first gear') to B2 contact ('second gear'). Thus, certain flies are able to produce efficient pronation movements at the upper turning point of the wing beat at high 'click' strength and to select between two distinct 'gear' positions with different
Fig. 1. The 'axilla' of the wing of Calliphora erythmcephala ('first gear'). A, process of the base of subcosta + radius; B, 'free' fulcrum; ax2, axillary 2, articulating ventrocaudally of the fulcrum. Scale bar, 0' 1 mm.
Dipteran
flight
467
'click' strength. If the tergopleural muscle is more relaxed, A passes frontolaterally to B, without having any contact at the upper turning point (stroboscopic observation). In my study of 1973 I suspected that the structures A and B might contribute to the click mechanism: A would be pressed against B, leading to a larger outward movement of the pleural ridge. It is now supposed that A, under natural conditions, slides smoothly out of the Bl or B2 groove at the beginning of the downstroke. This was demonstrated by a mechanical model (Pfau, 1985, and in preparation). According to Miyan & Ewing (19856) the 'radial stop' (= A) 'locates' into one of the grooves of the pleural wing process approximately in the middle of the downstroke (see also above). In order for this to work, A would first have to hit Bl or B2 precisely. This would, however, be extremely difficult at the high wing speed attained in the middle of the downstroke, particularly when one considers the interaction of muscular and aerodynamic forces, which would render a precise insertion most unlikely. Furthermore, if the wing did not locate exactly, unpredictable mechanical effects would arise. In my own model, A and B are brought together automatically at the upper turning point (at low speed!) — as if into the end of a funnel - but only if the strength of the click muscles is high (Fig. 1). Miyan & Ewing (19856) incorrectly state (p. 283) that according to Pfau (1973) the 'second basalar muscle' is responsible for the change between the grooves of the fulcrum (possibly misunderstanding a presumed function of this muscle which was described by Wisser & Nachtigall, 1984-table 2: ' b l ' - a s Pfau's view). [In this table the functions of the axillary muscles (Pfau, 1977) are not correctly cited.] The authors deny the possibility that the pleural wing process might move cranially, suspecting that the tergopleural muscle instead shifts the scutum with respect to B, thus determining the position of A via axillary 1 'by control of the twisting of the wing base' (pp. 283, 295). This is incorrect: the scutum, the frontal part of which is connected to the prothorax and mesopleurum, cannot shift. Instead a membranous vertical cleft in front of the pleural ridge gives way to a frontal bending of the pleural ridge and wing process (see also Boettiger & Furshpan, 1952, p. 203). Concerning a supposed 'natural' separation between A and B, Miyan & Ewing (19856, p. 282) refer to their fig. 9 (plate 2). However, A in this picture seems to lie mainly frontolaterally to B, as is the case in flies in which the click muscles are not strongly contracted. So this does not really correspond to fig. 4 (p. 279) and is by no means a proof for the 'natural' separation of A and B. In the light of the results of physiological and mainly functional anatomical research many of the conclusions of Miyan & Ewing must therefore be questioned, others may be disproved by simple demonstrations. This contribution in my opinion demonstrates that physiological research in this field must be combined with most profound functional anatomical research to produce meaningful results. I would like to thank Dr U. Koch, Dr B. Schroeter and Dr M. Tritsch for much Ibelp in translating the text.
468
H. K. PFAU
REFERENCES ALEXANDER, R. M C N . (1968). Animal Mechanics. London: Sidgwick & Jackson. BOETTIGER, E. G. & FURSHPAN, E. (1952). The mechanics of flight movements in Diptera. Biol. Bull. mar. biol. Lab., Woods Hole 102, 200-211. CHABRIER, J. (1822). Essai sur le vol des insectes. Mem. Mus. natn. Hist, nat., Paris 7, 297-372. HEIDE, G. (1971). Die Funktion der nicht-fibrillaren Flugmuskeln von Calliphora. Teil II: Muskulare Mechanismen der Flugsteuerung und ihre nervose Kontrolle. Zool.Jb. (Physiol.) 76, 99-138. MIYAN, J. A. & EwiNG, A. W. (1985a). Is the 'click' mechanism of dipteran flight an artefact of CC14 anaesthesia? J . exp. Biol. 116, 313-322. MIYAN, J. A. & EWING, A. W. (19856). How diptera move their wings: A re-examination of the wing base articulation and muscle systems concerned with flight. Phil. Trans. R. Soc. Ser. B 311, 271-302. NACHTIGALL, W. (1968). Gldserne Schwingen. Munchen: Moos Verlag. PFAU, H. K. (1973). Fliegt unsere SchmeiBfliege mit Gangschaltung? Naturwissenschaften 60, 160. PFAU, H. K. (1977). Funktion einiger direkter, tonischer Flugelmuskeln von Calliphora erythrocephala Meig. Verh. dt. Zool. Ges. 70, 275. PFAU, H. K. (1978). Funktionsanatomische Aspekte des Insektenflugs. Zool. jfb. (Anat.) 99, 99-108. PFAU, H. K. (1985). Zur funktionellen und phylogenetischen Bedeutung der "Gangschaltung" der Fliegen. Verh. dt. Zool. Ges. 78, 168. PFAU, H. K. (1986). Untersuchungen zur Konstruktion, Funktion und Evolution des Flugapparates der Libellen (Insecta, Odonata). Tijdschr. Ent. (in press). PFAU, H. K., ScHROETER, B. & STEITZ, E. (1977). Das Experiment: Der Klickmechanismus der Fliege - ein Demonstrationsmodell. Biologie in unserer Zeit 7, 188-190. PRINGLE, J. W. S. (1957). Insect Flight. London: Cambridge University Press. WlSSER, A. & NACHTIGALL, W. (1984). Functional-morphological investigations on the flight muscles and their insertion points in the blowfly Calliphora erythrocephala (Insecta, Diptera). Zoomorphology 104, 188-195.
