The Journal of Experimental Biology 206, 2999-3006 2999

2003 The Company of Biologists Ltd

doi:10.1242/jeb.00502

Into thin air: contributions of aerodynamic and inertial-elastic forces to wing

bending in the hawkmoth Manduca sexta
S. A. Combes* and T. L. Daniel
Department of Biology, University of Washington, Seattle, WA 98195, USA
*Author for correspondence (e-mail: [email protected])

Accepted 18 May 2003

Summary
During flapping flight, insect wings must withstand not contribution of fluid-dynamic forces to wing deformations
only fluid-dynamic forces, but also inertial-elastic forces is significantly reduced. This 85% reduction in air density
generated by the rapid acceleration and deceleration of produced only slight changes in the pattern of Manduca
their own mass. Estimates of overall aerodynamic and wing deformations, suggesting that fluid-dynamic forces
inertial forces vary widely, and the relative importance of have a minimal effect on wing bending. We used a
these forces in determining passive wing deformations simplified finite element model of a wing to show that the
remains unknown. If aeroelastic interactions between a differences observed between wings flapped in air versus
wing and the fluid-dynamic forces it generates are minor helium are most likely due to fluid damping, rather than
compared to the effects of wing inertia, models of insect to aerodynamic forces. This suggests that damped finite
flight that account for passive wing flexibility would be far element models of insect wings (with no fluid-dynamic
simpler to develop. We used an experimental approach to forces included) may be able to predict overall patterns of
examine the contributions of aerodynamic and inertial- wing deformation prior to calculations of aerodynamic
elastic forces to wing bending in the hawkmoth Manduca force production, facilitating integrative models of insect
sexta. We attached fresh Manduca wings to a motor and flight.
flapped them at a realistic wing-beat frequency and stroke
amplitude. We compared wing bending in normal air Key words: insect flight, wing flexibility, wing bending, aerodynamic
versus helium (approx. 15% air density), in which the forces, inertial forces, finite element model, Manduca sexta.

Introduction
Flapping wings produce a variety of forces as they accelerate wing bending must be coupled in each time step to calculations
and decelerate through a fluid medium. Some of these forces, of the aerodynamic forces generated by these shapes, a difficult
such as aerodynamic and added-mass forces, are related to the and time-consuming task. However, if inertial-elastic (fluid-
fluid through which the wing moves, while others, such as independent) forces dominate wing bending, the dynamic shape
inertial-elastic forces, are determined solely by the mass of the of flapping wings could be predicted prior to calculations of
wing and its material properties. In insects, these forces bend aerodynamic force production, avoiding the coupled aeroelastic
and twist the wings during flight, resulting in passive shape problem.
changes that may affect many aspects of flight performance, In some insect species, such as Drosophila, wing bending is
from the lift-to-drag ratio of wings (Batchelor, 1967) to thrust limited, and physical or mathematical models that assume the
production and fluid-dynamic efficiency (Combes and Daniel, wings are rigid can provide significant insights into mechanisms
2001; Daniel, 1987; Wu, 1971). Because insect flight muscles of unsteady force production (e.g. Dickinson et al., 1999;
are restricted to the wing base, these passive shape changes are Ramamurti and Sandberg, 2002; Sane and Dickinson, 2002; Sun
controlled primarily by the architecture and material properties and Tang, 2002a). However, the wings of many species, such as
of the wing; in many cases, these design features appear to Manduca, bend and twist dramatically during flight (Dalton,
permit certain beneficial deformations (e.g. lift-enhancing 1975; Wootton, 1990), particularly during slow flight and
torsion; Wootton, 1990), while preventing detrimental bending. hovering (Willmott and Ellington, 1997). Most computational
Being able to predict large, dynamic shape changes is essential models of flight in Manduca have accounted for wing bending
for developing a comprehensive understanding of insect flight, by incorporating simplified shape changes that are specified in
as instantaneous wing shape helps determine the direction and advance (Liu et al., 1998; Liu and Kawachi, 1998); these
magnitude of fluid-dynamic forces generated by the wing approaches have contributed substantially to our understanding
(Batchelor, 1967). If, in turn, these fluid-dynamic forces are of fluid dynamic force generation in specific situations, such
important in determining dynamic wing shape, predictions of as during hovering flight. However, models of insect flight
3000 S. A. Combes and T. L. Daniel
incorporating passive wing deformations could be used to for 5min, then removed one forewing at the base and recorded
address further questions of functional wing morphology and wing mass. We did not include the smaller, overlapping
evolution, as well as to explore the effects of alternative hindwings in this study, as their position relative to the
kinematic patterns on dynamic wing shape and insect flight forewings is variable during flight, and this interaction is
performance. difficult to recreate when the wings are detached from the
Unfortunately, the development of these integrative models animal. We marked three spots on the forewing with a small
has been hindered by uncertainty about the relative importance dot of reflective white paint (weighing 1.5% of total wing
of fluid-dynamic and inertial-elastic forces in determining mass) on both the dorsal and ventral sides: the wing tip, the
dynamic wing shape. Some estimates of overall wing inertia trailing edge (where chord length is maximum, approximately
(averaged spatially and/or temporally) suggest that inertial 50% wing span), and the leading edge (at the same spanwise
forces are generally higher than aerodynamic forces (Ellington, position; Fig.1A). We used cyanoacrylate glue cured with
1984b; Lehmann and Dickinson, 1997; Wilkin and Williams, baking soda to attach the base of the wing to a brass rod that
1993; Zanker and Gotz, 1990), whereas other studies conclude could be rotated by an oscillator constructed from the pen
the opposite (Sun and Tang, 2002b; Wakeling and Ellington, motor and amplifier of a Gould chart recorder (Fig.1A).
1997). A limited number of theoretical studies addressing local The motor was attached to a platform inside a 30cm
bending moments in flapping wings suggest that inertial-elastic Plexiglass box (with 1cm thick walls) and its motion was
forces may play a larger role than aerodynamic forces in
determining instantaneous wing shape. For example, Ennos Camera
(1989) estimated that spanwise bending moments due to the A
inertia of flapping wings are at least twice as large as those due
to aerodynamic forces, and showed that wing inertia alone
could cause the tip-to-base torsional wave seen in many insect
wings during supination (Ennos, 1988). Daniel and Combes Helium in
(2002) showed that chordwise bending moments generated by
elastic wave propagation in flapping insect wings (inertial-
elastic effects) are significantly larger than the moments
exerted on wings by the surrounding fluid. ld
wt
In this study, we used an experimental approach to examine
the relative contributions of inertial-elastic and fluid-dynamic tr y
forces to passive wing bending. We attached fresh Manduca
Air out Motor x
sexta wings to a motor and flapped them around the z
dorsalventral axis of the wing hinge at a realistic wing-beat
frequency and stroke amplitude, mimicking the large- Function
amplitude motions of freely flying moths. We used high-speed generator
video recording to compare instantaneous wing deformations
of wings flapped in normal air versus helium (approx. 15% air
density). The lower density of helium substantially reduces the
B
contribution of fluid-dynamic forces to the observed wing
deformations, allowing us to determine the relative importance
of these forces in passive wing bending.
At the same time, however, this lower fluid density also wt
ld
reduces external damping of the wings motions. We used a
simplified finite element model based on a Manduca wing to tr x
explore how damping alone (in the absence of fluid-dynamic z
forces) affects wing motions. Because the finite element y
analysis does not include fluid-dynamic forces, the motions of
the model wing depend solely on structural features and inertial- Fig.1. Apparatus used to visualize Manduca sexta wing bending in
elastic effects. We subjected the model to the same motions as normal air and helium. (A) Each wing was marked with dots of
real wings, and compared bending patterns in the undamped reflective paint at the wing tip (wt), leading edge (ld) and trailing
edge (tr) and filmed from orthogonal views while flapping around the
model wing to those of the model with damping added.
dorsalventral axis of the wing hinge (the y-axis). After filming, air
was repeatedly removed from the box and replaced with helium until
Materials and methods the box was filled with>95% helium. Wings were then filmed while
flapping at the same amplitude and frequency. (B) Coordinates of the
Dynamic bending experiments marked points were digitized and converted into angular position (),
We anaesthetized hawkmoths Manduca sexta (Linnaeus with the origin at the wing base and position (viewed from the
1763) from a colony at the University of Washington at 0C leading edge) measured from the center of rotation.
 Aerodynamic and inertial forces in wing bending 3001
controlled with a function generator. Wings were flapped ld
sinusoidally around the dorsalventral axis of the wing hinge
at room temperature. Total amplitude varied between 107 and
110, corresponding to intermediate stroke amplitudes in free- wt
flying hawkmoths (Willmott and Ellington, 1997). Wing
motions were recorded by two high-speed video cameras
(Redlake Inc., San Diego, CA, USA) at 1000 framess1, one y
viewing the wing from its leading edge and the other from its
tip (Fig.1A). tr x
1 cm
z
Each wing was filmed while flapping at 0.5Hz, as a control
for the shape and position of the wing when no dynamic Fig.2. Finite element model based on a Manduca sexta wing. The
bending occurs, and at 26Hz, a typical wing-beat frequency model approximates the planform geometry, vein configuration and
for Manduca sexta (Willmott and Ellington, 1997). The density spatial variation in flexural stiffness of a real wing. Declining
of fluid inside the chamber was then reduced by repeatedly material stiffness (E) of membrane and vein elements results in an
exponential decline in flexural stiffness (EI), as measured in
removing air through an opening near the bottom of the box
Manduca wings (Combes, 2002; Combes and Daniel, 2003b). Each
and adding helium through an opening near the top (Fig.1A).
color represents a different value of material stiffness, which varies
The wing was filmed flapping at 26Hz in a mixture of no from 4.7107 N m2 to 4.5109 N m2 in membrane elements, and
less than 95% helium, which has a density of 0.164gl1 from 1.91011 N m2 to 1.81013 N m2 in vein elements. The wing
(approximately 14% of normal air density; CRC, 2001). was rotated at its base around the y-axis, and bending was analyzed
Finally, the box was opened to release the helium and the wing by tracking the positions of nodes at the wing tip (wt), leading edge
was again filmed at 26Hz in normal air, to check for potential (ld) and trailing edge (tr).
wing damage. All filming was completed within 1h, during
which time the flexural stiffness of the wing does not change
appreciably (Combes, 2002). The procedure was repeated on Finite element modeling
four different wings from three individuals. As a wing is flapped through the air and deformed by inertial
and/or aerodynamic forces, its motions are damped by some
Wing bending analysis combination of internal (e.g. elastic or structural) and external
We analyzed frames from three complete flaps in the (fluid) mechanisms. When the surrounding fluid is less dense
middle of each filming sequence, to avoid bending artefacts (as is the case with the Manduca wings flapped in helium) the
at the onset of motion. A custom Matlab program (developed wing experiences less external damping. We explored how
by M. S. Tu) identified the coordinates of the wing tip, damping affects wing bending by constructing a simplified
leading edge and trailing edge in each frame. These three- finite element model (FEM) based on a male Manduca
dimensional Cartesian coordinates were converted to forewing, and comparing bending in the undamped wing to
spherical coordinates, using the wing base as the origin and bending in the model wing with damping added. The model
measuring the position of the wing (viewed from the leading was created in MSC Marc/Mentat and is composed of thin
edge) in degrees, with 0 at the center of rotation (Fig.1B). shell elements of uniform thickness, recreating the planform
Flapping frequency was found by dividing the number of configuration of veins and membranes in a real wing (but
complete flaps by the total number of frames and multiplying omitting details of three-dimensional wing structure; see
by 1000. Amplitude applied by the motor was measured at Combes and Daniel, 2003b). We applied declining values of
the leading edge in the control sequence (0.5Hz) to avoid material stiffness to the model wing in 12 strips, oriented
wing bending, using the maximum excursion of the leading diagonally (Fig.2); these strips are perpendicular to most of
edge marker to define the sides of a right triangle. To the wing veins, which decrease in diameter towards the wing
determine if amplitude applied at the base changes edge and thus are likely to decrease in stiffness along this axis.
significantly with flapping frequency, a brass rod of the same This configuration results in an exponential decline in flexural
length and mass as a Manduca wing was attached to the stiffness EI (the product of Youngs modulus E and the second
motor and filmed at 0.5Hz and 26Hz. moment of area I) in both the spanwise and chordwise
To examine temporal patterns of bending at each wing directions of the wing, approximating patterns of flexural
location, we compared the trajectory of a wing flapping at stiffness measured in real wings (Combes, 2002; Combes and
26Hz and at 0.5Hz (where no dynamic bending occurs), Daniel, 2003b). Within each strip, vein elements have a higher
adjusting the time base of the control sequence to match that material stiffness than membrane elements, mimicking the
of the experimental sequence and splining data to equal time increased flexural stiffness of tubular veins. We used an
intervals in Matlab. We then calculated the difference in element density of 1200kgm3 (as measured in insect wings;
position at each time point and performed a Fourier analysis Wainwright et al., 1982), a thickness of 45m, and a Poissons
on this wing bending data to determine the dominant ratio of 0.49 (consistent with measured values of biological
frequencies of wing motion and the amplitude coefficient at materials; Wainwright et al., 1982). To determine the
each frequency. minimum number of elements necessary to capture the bending
3002 S. A. Combes and T. L. Daniel
behavior of wings, we performed a sensitivity analysis with initial conditions of zero displacement and zero velocity at all
models composed of 200, 350, 865 and 2300 total elements, nodes, and gradually increased the rotation at the wing hinge
and found that 865 elements are sufficient to ensure asymptotic to a sinusoidal motion with the following function:
performance of the model.
(t) = (1et/)sin(t), (1)
We applied boundary conditions to the nodes at the wing
hinge so that they could not translate in any direction and could where is rotation at the base nodes, t is time in s, is the time
rotate only along the dorsalventral axis, as in experiments on constant and is the angular frequency (2f, where f is the
real wings (Fig.2, red arrows). We began the simulation with flapping frequency). We flapped the wing at 26Hz, and found

90 Ai 16 Aii
Wing tip position/bending (degrees)

Position, normal air

14 Position, helium
60 * Position, slow rotation
12 Bending, normal air
30
10 Bending, helium
0 8

30 6
4
60
2
90 0
Leading edge position/bending (degrees)

90 Bi 16 Bii
14
Bending amplitude coefficient

60
12
30
10
0 8
*
6
30
4
60 2

90 0
Trailing edge position/bending (degrees)

90 Ci 16 Cii
*
60 14
12
30
10
0 8

30 6
4
60
2
90 0

0 20 40 60 80 100 120 0 50 100 150 200 250

Time (ms) Frequency (Hz)
Fig.3. (AiCi) Angular position and bending at the wing tip (A), leading edge (B) and trailing edge (C) of a Manduca wing flapped in normal
air versus helium. The time base of 0.5Hz control sequences (slow rotation) was adjusted to match 26Hz experiments for comparison. At each
of the wing locations, angular position during the control sequence (slow rotation; black line) was subtracted from position during the 26Hz
sequences in normal air or helium (green or orange lines) to quantify temporal patterns of wing bending (blue or red lines). (AiiCii) Amplitude
coefficients from Fourier analyses of wing bending in normal air and helium are shown on the right, with the driving frequency of 26Hz
indicated by asterisks.
 Aerodynamic and inertial forces in wing bending 3003
that a time constant () of 1/20s avoids transient artefacts of experimental sequences is not significantly different from that
rapid initial acceleration and allows the wing to reach its full applied during control sequences. Wing tip, leading edge and
stroke amplitude of 108 (the mean amplitude of experiments trailing edge trajectories from sequences recorded in normal air
on real wings) after 6.5 flaps. at the end of the experiment were indistinguishable from those
We performed one simulation with no damping added to the recorded at the beginning (Combes, 2002), confirming that the
computational analysis, and another in which we added mass experimental procedure does not damage wings. Data from only
damping, adjusting the level of damping to best represent the the initial sequence in normal air were used for further analysis.
observed changes in motion between real wings flapped in In control sequences, the angular positions of the wing tip,
normal air versus helium. We ran each simulation for 19200 leading edge and trailing edge were nearly identical (black
time steps, recreating 12.5 flaps in 0.48s, and measured lines, Fig.3) and total amplitude was equivalent to the
displacement of nodes at the wing tip, leading edge and trailing amplitude applied at the base, demonstrating that no dynamic
edge (in the same locations as on real wings; Fig.2), through bending occurs at these low frequencies. When flapping
six flaps after the wing motion had reached full amplitude. To frequency was increased during experimental sequences, peak
provide a control with no dynamic bending (analogous to the amplitudes at the wing tip and trailing edge increased,
slow rotation in the experiments), we created a stiff wing indicating that wings bent considerably at the end of each
by changing the Youngs modulus of all elements to stroke (Fig.3AiCi), while amplitude at the leading edge
11016N m2, and subjected this wing to the same motions as changed only slightly (Fig.3Bi).
the flexible wings. We quantified temporal patterns of wing Wings flapped in helium displayed slightly higher peak
bending at each of the wing locations by finding the difference amplitudes than those flapped in normal air, but patterns of
in position between the stiff wing and the flexible wing (with bending were similar (Fig.3AiCi; see also http://faculty.
and without damping), and performed a Fourier analysis on the washington.edu/danielt/movies for movies of flapping wings).
resulting data. Fourier analysis reveals that the dominant frequencies of wing
bending were the same in both helium and normal air, and that
only the amplitude of some higher harmonics differed
Results (Fig.3AiiCii; Table1). Amplitude coefficients were similar in
Wing bending experiments normal air and helium at the driving frequency (26Hz), but were
The frequency at which Manduca wings were flapped varied often larger in helium at higher harmonics, particularly at the
from 25.5 to 26.9Hz (x =26.2Hz) in experimental sequences, second harmonic (78Hz; Table1).
and from 0.56 to 0.57Hz (x =0.565) in control sequences.
Flapping amplitude during control sequences varied from Finite element modeling
107.1 to 109.9 (x =108.0). Amplitude of the brass rod In the stiff FEM wing, the angular positions of the wing tip,
changed by only 1.7% when flapped at 0.57Hz and 26.3Hz, leading edge and trailing edge were identical (Fig.4, black
indicating that amplitude applied at the base during lines) and equivalent to amplitude at the base; thus, as in

Table1. Amplitude coefficients from Fourier analyses of wing bending at the tip, leading edge and trailing edge of
Manduca wings
Amplitude coefficient
26 Hz* 52 Hz 78 Hz 104 Hz
Wing Normal Helium Normal Helium Normal Helium Normal Helium
Wing tip 1 12.675 10.641 1.540 3.999 3.747 9.012 1.347 1.190
2 12.196 10.706 1.289 3.496 3.050 9.083 1.437 3.367
3 13.235 11.856 2.783 2.744 2.710 10.049 1.514 3.739
4 12.545 13.097 3.613 3.041 2.465 7.077 1.372 4.025
Leading edge 1 4.620 4.323 0.906 0.798 1.524 2.564 0.119 0.614
2 3.486 4.452 1.026 0.911 1.126 3.014 0.231 0.436
3 3.801 5.280 0.901 0.687 0.896 2.906 0.469 0.858
4 5.236 6.817 1.561 1.175 1.092 2.045 0.595 1.120
Trailing edge 1 11.386 9.904 1.157 2.677 3.626 10.295 1.227 5.581
2 12.589 9.794 1.341 2.536 4.467 9.553 1.519 4.666
3 14.034 10.564 0.870 1.629 5.228 10.714 1.377 4.206
4 14.965 13.062 1.222 2.856 5.680 9.485 2.779 4.470

Coefficients at the driving frequency (26Hz; asterisk) and first three harmonics are shown for each wing in normal air and in helium.
Cases where the coefficient in helium varied by more than 100% from the coefficient in normal air are shown in bold.
3004 S. A. Combes and T. L. Daniel
control sequences on real wings, the stiff FEM wing displayed damping and the wing with mass damping were similar to
no dynamic bending. In the flexible FEM wings, maximum tip those seen between real wings flapped in helium and in normal
and trailing edge amplitudes were higher than the amplitude air. The undamped model wing showed slightly higher peaks
applied at the base, while the leading edge amplitude changed in wing bending (Fig.4AiCi), but the same overall bending
only slightly (Fig.4AiCi). Although bending amplitude at the pattern as the damped wing. Fourier analysis revealed that the
trailing edge of the FEM wings was lower than in real dominant frequencies of bending were the same in the two
Manduca wings, temporal patterns of wing bending were simulations, and that amplitude coefficients were similar at the
similar (Figs3AiCi, 4AiCi). driving frequency and larger in the undamped model at higher
In addition, the differences between the FEM wing with no frequencies (Fig.4AiiCii).

90 Ai Position, damped FEM

Aii
Wing tip position/bending (degrees)

16 Position, undamped FEM

60 14 * Position, stiff FEM
Bending, damped FEM
12
30 Bending, undamped FEM
10
0 8

30 6
4
60 2

90 0
Trailing edge position/bending (degrees) Leading edge position/bending (degrees)

90
Bi 16 Bii
Bending amplitude coefficient

60 14
12
30
10
0 8
*
30 6
4
60
2
90 0

90 Ci 16 Cii
60 14
12
30 *
10
0 8

30 6
4
60
2
90 0

0 20 40 60 80 100 120 0 50 100 150 200 250

Time (ms) Frequency (Hz)

Fig.4. (AiCi) Angular position and bending at the wing tip (A), leading edge (B) and trailing edge (C) of a finite element model based on a
Manduca wing. Wing bending was calculated as in Fig.3, by finding the difference between the angular position of a stiff wing (black line;
analogous to the 0.5Hz control sequence in real wings) and that of a flexible model wing, with or without mass damping (green or orange
lines). (AiiCii) Amplitude coefficients from Fourier analyses of wing bending in the damped versus undamped model are shown on the right,
with the driving frequency of 26Hz indicated by asterisks.
 Aerodynamic and inertial forces in wing bending 3005
Discussion may involve increased aerodynamic forces, as well as rapid
Aerodynamic versus inertial-elastic forces in Manduca wing accelerations and decelerations that could augment inertial-
bending elastic forces. The extent to which more detailed kinematics
Our measurements of regional wing bending show that might alter our findings about the relative contributions of
flapping Manduca wings undergo significantly more dynamic aerodynamic and inertial-elastic forces to wing bending
bending at the wing tip and trailing edge than along the leading remains a subject of future study.
edge, confirming previous static measurements of regional It is also important to note that the relative contributions of
aerodynamic and inertial-elastic forces to wing bending are
flexural stiffness (Combes and Daniel, 2003b). Wings flapped
likely to vary along a continuum, from hovering, where
in helium displayed similar spatial and temporal bending
inertial-elastic forces appear to dominate in Manduca, to the
patterns and the same dominant frequencies of motion as wings
extreme case of steady, forward flight with no flapping, where
flapped in normal air, despite an 85% reduction in fluid density.
inertial forces are negligible and any wing bending would be
This demonstrates that the contribution of aerodynamic
due solely to aerodynamic forces. In many insects, however,
loading to instantaneous wing shape in Manduca is minor
the most pronounced wing bending and twisting occurs during
compared to the contribution of wing inertia.
slow flight or hovering (e.g. in Manduca; Willmott and
Although overall patterns of bending were remarkably
Ellington, 1997), so passive deformations may in fact decrease
similar, high-frequency components of bending motion were
as aerodynamic forces begin to dominate.
more pronounced in wings flapped in helium (Fig.3AiiCii);
this was manifested visually as rapid oscillations in the more Insect size and wing design
flexible regions of the wing, particularly as the wing slowed
Because the Manduca wings used in this study are relatively
and began to move in the opposite direction. Simulations of
large and heavy, it is possible that inertial-elastic effects are
wing bending in the finite element model suggest that reduced
more important in determining wing bending in this species
damping may explain this difference. Adding damping to the than in other species with smaller, lighter wings. A simple
finite element analysis reduced higher-frequency components analysis of average bending moments can be used to assess
of motion in the model wing, just as increasing the density of the relative magnitudes of inertial-elastic and aerodynamic
the fluid (by using normal air as opposed to helium) reduced moments on the flapping wings of different species (Daniel and
higher-frequency components of motion in real wings Combes, 2002):
(Figs3AiiCii, 4AiiCii).
These results suggest that a damped finite element model R = (mw/mb)42L / 3g, (2)
(with realistic, three-dimensional forces applied at the base) where R is the ratio of inertial-elastic to aerodynamic bending
could be successful in predicting the overall pattern and moments, mw is mass of one wing, mb is mass of the body,
magnitude of Manduca wing deformations during flight, is angular stroke amplitude, is angular frequency, L is wing
independent of aerodynamic calculations. The finite element span and g is earths gravitational acceleration. The ratio of
model used in this study contains several simplifications in wing to body mass in insects has been shown to vary from
three-dimensional geometry that may limit its ability to predict 0.5% in bees to 6% in hawkmoths (Ellington, 1984a), and wing
wing motions precisely. In addition, we did not incorporate an span varies widely. However, because the frequency term in
accurate distribution of wing mass, which declines sharply the above equation is squared, wing-beat frequency has a large
towards the tip and trailing edges (although preliminary effect on the moment ratio. Thus, many small insects (with
simulations suggest that mass distribution affects primarily the higher wing beat frequencies; Dudley, 2000) may actually have
magnitude, not the pattern of wing bending). Yet even this higher ratios of inertial-elastic to aerodynamic bending
simplified model was able to simulate temporal and regional moments, despite having smaller, lighter wings. Our estimates
wing bending patterns relatively well, suggesting that a slightly suggest that this ratio is quite large in insects over a broad size
more detailed finite element model could provide very accurate range (R=7 in Manduca and R=6 in Drosophila). Although the
results. magnitude of passive wing bending that actually occurs during
To recreate Manduca wing motions during flight precisely, flight depends on additional factors (such as the scaling of wing
the boundary conditions at the base of the model wing would stiffness; Combes and Daniel, 2003a), these results indicate
also need to be altered. The experimental work and dynamic that the spatial and temporal patterns of whatever passive
modeling in this study were based on a relatively simple bending does occur are likely to be determined primarily by
kinematic pattern, in which the wing was rotated around only inertial-elastic effects in many species.
the dorsalventral axis of the wing hinge. In most insects, In addition to large variations in size, insect wings display
muscular forces transmitted to the wing base not only propel tremendous variability in design features, such as planform
the wing with large amplitude motions such as these, but also wing shape and the arrangement of supporting veins, which
rotate the wing around its leading edge, controlling the angle could affect how their wings respond to aerodynamic and
of attack of the wing and, in some cases, causing significant inertial-elastic forces. Interestingly, despite dramatic visual
spanwise twisting. The rapid wing rotations evident in some differences in wing design, overall wing stiffness appears to
species during stroke reversal (e.g. Dickinson et al., 1999) scale strongly with wing size in a broad range of species
3006 S. A. Combes and T. L. Daniel
(Combes and Daniel, 2003a). Patterns of regional stiffness References
variation in insect wings may in fact be affected by wing shape Batchelor, G. K. (1967). An Introduction to Fluid Dynamics. Cambridge:
Cambridge University Press.
and venation; however, we have measured very similar Combes, S. A. (2002). Wing flexibility and design for animal flight. PhD
patterns of stiffness variation in the wings of hawkmoths and thesis, University of Washington.
aeshnid dragonflies, insects with strikingly different wing Combes, S. A. and Daniel, T. L. (2001). Shape, flapping and flexion: wing
and fin design for forward flight. J. Exp. Biol. 204, 2073-2085.
designs (Combes and Daniel, 2003b). These results suggest Combes, S. A. and Daniel, T. L. (2003a). Flexural stiffness in insect
that large differences in insect wing design do not necessarily wings: I. Scaling and the influence of wing venation. J. Exp. Biol. 206,
lead to equivalent differences in wing stiffness and bending 2979-2987.
Combes, S. A. and Daniel, T. L. (2003b). Flexural stiffness in insect
behavior. wings: II. Spatial distribution and dynamic wing bending. J. Exp. Biol. 206,
2989-2997.
Concluding remarks CRC (2001). Handbook of Chemistry and Physics, 81st edition. Cleveland,
Ohio: CRC Press.
Because aeroelastic effects appear to be relatively Dalton, S. (1975). Borne on the Wind: The Extraordinary World of Insects in
unimportant in determining dynamic wing shape, an Flight. New York: Readers Digest Press.
integrative model of insect flight that incorporates passive wing Daniel, T. L. (1987). Forward flapping flight from flexible fins. Can J. Zool.
66, 630-638.
flexibility may be easier to develop than previously thought. Daniel, T. L. and Combes, S. A. (2002). Flexing wings and fins: bending by
Although the experimental work presented in this study inertial or fluid-dynamic forces? Int. Comp. Biol. 42, 1044-1049.
addresses the relative contributions of aerodynamic and Dickinson, M. H., Lehman, F.-O. and Sane, S. P. (1999). Wing rotation and
the aerodynamic basis of insect flight. Science 284, 1954-1960.
inertial-elastic forces to wing bending in only one species using Dudley, R. (2000). The Biomechanics of Insect Flight: Form, Function,
a particular kinematic pattern, these results verify recent Evolution. Princeton, NJ: Princeton University Press.
theoretical studies (Daniel and Combes, 2002) suggesting that Ellington, C. P. (1984a). The aerodynamics of hovering insect flight. Part
II: Morphological parameters. Phil. Trans. R. Soc. Lond. B 305, 17-40.
fluid-dynamic forces have only a minor effect on passive Ellington, C. P. (1984b). The aerodynamics of hovering insect flight. Part
bending when flexible structures are flapped in air (versus VI: Lift and power requirements. Phil. Trans. R. Soc. Lond. B 305, 145-
water, where fluid forces dominate). In addition, while detailed 181.
numerical methods (e.g. Liu et al., 1998; Ramamurti and Ennos, A. R. (1988). The inertial cause of wing rotation in Diptera. J. Exp.
Biol. 140, 161-169.
Sandberg, 2002; Sun and Tang, 2002a; Wang, 2000) will Ennos, A. R. (1989). Inertial and aerodynamic torques on the wings of Diptera
undoubtedly continue to contribute to our understanding of in flight. J. Exp. Biol. 142, 87-95.
three-dimensional fluid flow around flapping wings, recent Lehmann, F.-O. and Dickinson, M. H. (1997). The changes in power
requirements and muscle efficiency during elevated force production in the
work suggests that far simpler analytical methods are able to fruit fly Drosophila melanogaster. J. Exp. Biol. 200, 1133-1143.
predict temporal patterns of unsteady force production Liu, H., Ellington, C. P., Kawachi, K., Van den Berg, C., and Willmott,
A. P. (1998). A computational fluid dynamic study of hawkmoth hovering.
remarkably well (Sane and Dickinson, 2002). Inserting the J. Exp. Biol. 201, 461-477.
dynamic shape of wing sections (as determined by finite Liu, H. and Kawachi, K. (1998). A numerical study of insect flight. J. Comp.
element analysis or other inertia-based methods) into quasi- Physics 146, 124-156.
Ramamurti, R. and Sandberg, W. C. (2002). A three-dimensional study
steady models of flight that account for unsteady effects may of the aerodynamic mechanisms of insect flight. J. Exp. Biol. 205, 1507-
yield a tractable modeling tool that could be used to explore 1518.
the effects of wing flexibility on unsteady force production. Sane, S. P. and Dickinson, M. H. (2002). The aerodynamic effects of wing
rotation and a revised quasi-steady model of flapping flight. J. Exp. Biol.
Models of this type could be used to determine when and how 205, 1087-1096.
wing flexibility affects aerodynamic force generation, and Sun, M. and Tang, J. (2002a). Unsteady aerodynamic force generation by a
ultimately contribute to an integrative model of insect flight model fruit fly wing in flapping motion. J. Exp. Biol. 205, 55-70.
Sun, M. and Tang, J. (2002b). Lift and power requirements of hovering flight
linking sensory feedback and patterns of muscle force in Drosophila virilis. J. Exp. Biol. 205, 2413-2427.
production to dynamic wing motions, force generation and Timoshenko, S., Young, D. H. and Weaver, W., Jr (1974). Vibration
insect flight performance. Problems in Engineering. New York: John Wiley & Sons.
Wainwright, S. A., Biggs, W. D., Currey, J. D. and Gosline, J. M.
(1982). Mechanical Design in Organisms. Princeton: Princeton University
The authors would like to thank M. Dillon for his assistance Press.
in constructing the vacuum chamber and calibrating the high- Wakeling, J. M. and Ellington, C. P. (1997). Dragonfly flight. III. Lift and
speed video cameras, and J. Longnion for his work in power requirements. J. Exp. Biol. 200, 583-600.
Wang, Z. J. (2000). Vortex shedding and frequency selection in flapping
salvaging and altering chart recorder motors. E. Goldman flight. J. Fluid Mech. 410, 323-341.
provided valuable suggestions regarding data analysis, and M. Wilkin, P. J. and Williams, M. H. (1993). Comparison of the aerodynamic
Tu provided Matlab code for digitizing wing coordinates. K. forces on a flying sphingid moth with those predicted by quasi-steady
theory. Physiol. Zool. 66, 1015-1044.
Flick was instrumental in troubleshooting the high-speed Willmott, A. P. and Ellington, C. P. (1997). The mechanics of flight in the
video cameras and J. Dierberger at MSC Software provided hawkmoth Manduca sexta. I. Kinematics of hovering and forward flight. J.
crucial assistance with the FEM model. S. Sane, M. Tu and Exp. Biol. 200, 2705-2722.
Wootton, R. J. (1990). The mechanical design of insect wings. Sci. Am.
two anonymous reviewers provided helpful comments on November, 114-120.
drafts of the paper. This work was supported by NSF grant Wu, T. Y. (1971). Hydromechanics of swimming propulsion. Part 1.
F094801 to T.D., the John D. and Catherine T. MacArthur Swimming of a two dimensional flexible plate at variable forward speeds in
an inviscid fluid. J. Fluid Mech. 46, 337-355.
Foundation, an ONR-MURI grant to T.D., an NSF graduate Zanker, J. M. and Gotz, K. G. (1990). The wing beat of Drosophila
fellowship to S.C. and an ARCS fellowship to S.C. melanogaster. II. Dynamics. Phil. Trans. R. Soc. Lond. B 327, 19-44.