Download Spring or string: does tendon elastic action influence wing muscle

Downloaded from http://rspb.royalsocietypublishing.org/ on June 17, 2017
rspb.royalsocietypublishing.org
Research
Cite this article: Konow N, Cheney JA,
Roberts TJ, Waldman JRS, Swartz SM. 2015
Spring or string: does tendon elastic action
influence wing muscle mechanics in bat flight?
Proc. R. Soc. B 282: 20151832.
http://dx.doi.org/10.1098/rspb.2015.1832
Received: 29 July 2015
Accepted: 1 September 2015
Spring or string: does tendon elastic
action influence wing muscle mechanics
in bat flight?
Nicolai Konow1, Jorn A. Cheney1, Thomas J. Roberts1, J. Rhea S. Waldman1
and Sharon M. Swartz1,2
1
2
Department of Ecology and Evolutionary Biology, Providence, RI 02912, USA
School of Engineering, Brown University, Providence, RI 02912, USA
Tendon springs influence locomotor movements in many terrestrial animals,
but their roles in locomotion through fluids as well as in small-bodied mammals are less clear. We measured muscle, tendon and joint mechanics in an
elbow extensor of a small fruit bat during ascending flight. At the end of downstroke, the tendon was stretched by elbow flexion as the wing was folded.
At the end of upstroke, elastic energy was recovered via tendon recoil and
extended the elbow, contributing to unfurling the wing for downstroke.
Compared with a hypothetical ‘string-like’ system lacking series elastic compliance, the tendon spring conferred a 22.5% decrease in muscle fascicle
strain magnitude. Our findings demonstrate tendon elastic action in a small
flying mammal and expand our understanding of the occurrence and action
of series elastic actuator mechanisms in fluid-based locomotion.
Subject Areas:
biomechanics, physiology
Keywords:
biomechanics, elastic energy storage,
biological springs, bioinspiration, flight
Author for correspondence:
Nicolai Konow
e-mail: [email protected]
Electronic supplementary material is available
at http://dx.doi.org/10.1098/rspb.2015.1832 or
via http://rspb.royalsocietypublishing.org.
1. Introduction
When the limbs of terrestrial animals collide with the ground during locomotion, the rapid rise in limb muscle force often leads to stretch of the
tendons that connect muscles to the bones they move [1 –4]. By operating as biological springs, compliant tendons play several well-known and important roles
in terrestrial locomotion, all involving relaxation of the temporal relationship
between muscle contraction and joint movement [2,5] (figure 1). For instance,
the elastic stretch and recoil of series elastic elements allows muscle to operate
at contraction speeds favourable for force and power production [2,5], and
facilitates recovery of mechanical energy with each stride during running
and hopping [3,6 –9]. Elastic mechanisms are widespread among large running
animals [2,4,7,9], but the limb muscles of small mammals (less than 1 kg) may
not generate sufficient force to stretch their tendons, which often are relatively
thick and stiff [1,10–12].
Wings and fins interact with their fluid environments without collisional
impact, preventing impact-driven tendon stretch during flight and swimming.
However, evidence suggests that tendons store and recover energy to decelerate
and reaccelerate limb elements during cyclic movements in aerial and aquatic
locomotion [13–15], and tendon action could contribute up to 60% of the net
mechanical work for wing upstroke in pigeons [16].
In addition to the potential role of tendon in energy storage and recovery,
there may be other benefits to transmitting muscle force via tendons. Elongated
tendons allow muscles to be positioned proximally, and the concentration of
muscle mass near the limb centre of rotation decreases the mass moment of
inertia of the limb [2,17]. This may be a particularly important trait in bats,
which rapidly oscillate heavy wings and thus incur non-negligible inertial
costs during flight [18– 20].
We combined X-ray-based three-dimensional high-speed visualization and
modelling techniques with tissue mechanical testing to study an elbow extensor
muscle, the monoarticular lateral head of triceps brachii, during ascending
flight in the phyllostomid fruit bat Carollia perspicillata (electronic supplementary
& 2015 The Author(s) Published by the Royal Society. All rights reserved.
Downloaded from http://rspb.royalsocietypublishing.org/ on June 17, 2017
(a)
lm
lm
q
(b)
q
lm
lm
lt
t0
t1
t2
q
material, movies S1 and S2). We asked if the tendon of triceps acts
elastically to let joint extension occur at a delay from muscle
shortening, or inelastically to transfer movement directly, like a
servo cable in an aircraft wing. The latter finding would be
consistent with the hypothesis that small mammals have relatively stiff tendons [1,11]. We predicted that if the tendon is
relatively stiff, the muscle fascicles of triceps would lengthen
during elbow flexion, and elbow extension would occur during
muscle fascicle shortening (figure 1a). Alternatively, if the
tendon is relatively compliant, its stretch during muscle contraction would be out of phase with muscle length change
and joint motion (figure 1b). To test between these alternatives,
we compared direct in vivo measurements of muscle fascicle
length change measured by fluoromicrometry [21] to the
strain of the whole muscle tendon unit (MTU), measured from three-dimensional models generated using X-ray
reconstruction of moving morphology (XROMM) [21–23]
(figure 2a–f and electronic supplementary material, movie S3).
2. Methods
(a) Subjects and experiment design
Four similarly sized male C. perspicillata (Seba’s short-tailed bat;
body mass of 16.1, 18.2, 18.7 and 19.1 g with wingspans ranging
from 293 to 297 mm) were used in this study. All experiments
and animal husbandry complied with a protocol approved by
the Brown University IACUC and with USDA regulations.
Ascending flight trials were elicited within a vertically oriented
thin-walled plastic cylinder (diameter: 400 mm, height: 600 mm),
which helped confine subject movements to the spherical imaging
volume (diameter: 220 mm) of the biplanar high-speed fluoroscopy
system. Following flight training within this arena, subjects were
anaesthetized via facemask (isoflurane, 1–2.5%; O2, 0.8 l min21),
surfaces for skin incisions were depilated (Veet), and radio-opaque
tantalum spheres were implanted during sterile procedures.
(b) Focus muscle and instrumentation
Muscle length was measured in vivo using three-dimensional
motion tracking of 0.5 mm tantalum spheres implanted at the
myotendinous junctions of the lateral head of triceps brachii, a
(c) Post-mortem data acquisition
To obtain XROMM reconstruction of in vivo movement of the
scapula, humerus and radius (figure 2; electronic supplementary
material, movie S2), we obtained mCT scans. One scan came
from the American Museum of Natural History (29 mm isovolumetric voxels, GE phoenix vjtomejx s240) while the other three
bats were scanned at Stony Brook University (36 mm isovolumetric
voxels, Scanco mCT40; n ¼ 3). The scans were segmented in OsiriX
and registered to the digitized three-dimensional marker constellations using a combination of marker-based XROMM [22] and
scientific rotoscoping workflows [23].
The precision for conventional marker-based XROMM has
been calculated to be 0.1 mm based on the standard deviation of
rigid body intermarker distances [22]. To reconstruct scapula and
humerus position in the small bats studied here, we increased the
spatial resolution of the X-ray system to 4.0 lp mm21 (measured
using a line-pair graticule), by using high magnification (Mag II),
at the expense of visualization volume. The error in rigid-body
reconstruction and muscle marker position for a given flight
sequence was estimated using the standard deviation for intermarker distance, which depends on the bone and marker. For the
in-dwelling bone markers, mean precision was 0.076 mm (range:
0.096–0.062 mm); for the twist-tied laser-perforated markers,
mean precision was 0.050 mm (range: 0.500–0.021 mm; n ¼ 4 bats).
Instantaneous MTU length was measured in MAYA (Autodesk)
as each XROMM reconstruction moved through in vivo elbow positions from 48 ascending flight wingbeats in three subjects. MTU
length was measured with locators parented to the triceps
muscle scar on the proximal humerus, and to the sesamoid insertion onto the olecranon, superficial and proximal to the elbow
joint. XROMM models and fluoromicrometry data were then
used to digitally calculate the relationship between triceps MTU
displacement and elbow joint position. For one bat, XROMM
reconstructions were not possible because of marker collinearity
issues (see [22] for further explanation of this phenomenon).
For this bat, the relationship between MTU excursion and joint
position was determined in a tendon travel experiment [25].
Linear regressions of MTU length change (mm) against elbow
joint position (radians) revealed a linear relationship across the
operational range of the muscle for all individuals (electronic
supplementary material, figure S2). We used the slope of the
regression line to calculate MTU length for the 81 wingbeats
that were not reconstructed using XROMM models but for
which elbow flexion–extension kinematics were measured via
three-dimensional motion tracking of bone markers. The vertical
position of the wrist was used to identify reversal points between
upstroke and downstroke for all wingbeats. To explore the
phase relationship between stretch–shortening of the triceps
muscle and flexion –extension of the elbow, we used analyses of
Proc. R. Soc. B 282: 20151832
Figure 1. Schematic of direct position control versus series elastic actuator
control of joint position. (a) The length, lm, of a monoarticular extensor
muscle without significant series elastic compliance, or without tendon,
must increase during a decrease in joint angle, u ( joint flexion), and decrease
during an increase in joint angle ( joint extension). (b) The relationship
between muscle length change and joint motion can be relaxed by
tendon compliance; here, total muscle – tendon unit length is the sum of
muscle fascicle length, lm, and length of a compliant tendon, lt; when the
joint flexes, u decreases, but muscle fascicle length does not increase. The
graphs on the right illustrate predictions for the relationship between
muscle length and joint position in a stiff (a) versus a compliant (b) system.
2
rspb.royalsocietypublishing.org
q
monoarticular elbow extensor with near-parallel-fibred architecture.
Tantalum spheres (0.4 mm) were also implanted into the humerus
and radius to reconstruct wingbeat kinematics and muscle–
tendon unit length. The medial border and acromion process of
the scapula were marked by piercing the approximately 0.4 mm
thick bone with 26G hypodermic needles, threading 30G silver
wires through the bone perforation and twist-tying 1.0 mm laserperforated tantalum spheres tightly to the bone (electronic
supplementary material, figure S1). During ascending flight, biplanar
recordings of marker positions were taken using two high-resolution
(1760 1760 pixel), high-speed (330 fps) cameras recording at
500 ms shutter speed through two X-ray systems (X-ray technique:
53–55 kV; 250 mA; 145 cm SID). Muscle activity was measured
from the triceps lateral head using fine-wire electromyography
[24]. Elbow kinematics were calculated from tracked wing bone
markers, and muscle length change trajectory was calculated via
fluoromicrometry [21] from the myotendinous junction markers.
Downloaded from http://rspb.royalsocietypublishing.org/ on June 17, 2017
(a)
3
(f )
rspb.royalsocietypublishing.org
q
(g)
60
(b)
qelbow(°)
10 mm
80
100
140
Lmtu (mm)
(c)
Lmuscle (mm)
(i)
(e)
( j)
speed (Lm,r s–1)
(d)
25.6
16.8
25.2
16.4
24.8
16.0
Ltendon (mm)
17.2
26.0
12
111
10
0.6
9
0
EMG (mV)
(h)
6
3
shortening
0
–3
lengthening
–6
0.20
0.25
0.30
0.35
0.40
time (s)
Figure 2. XROMM reconstruction of bat wing movements, and measurements of elbow joint motion and triceps muscle tendon mechanical function during flight.
(a – e) Frames from orthogonal high-speed videofluoroscopy (left column, dorsal; right column, lateral; see also the electronic supplementary material, movie S2).
Dark circles are radio-opaque markers, implanted in bone and at myotendinous junctions; mCT bone scans are registered to marker positions in X-ray images (semitransparent in a). Cable carries electromyographic signals. (a) Mid-upstroke. (b) Early elbow extension. (c) Late elbow extension. (d ) Early elbow flexion. (e) Late
elbow flexion. (f ) Dorsal view of XROMM reconstruction (a,c,d,e), with scapula held in fixed position. (g) Elbow joint angle. (h) MTU length (black line, from circle
markers in f ) and tendon length (blue line). (i) Muscle fascicle length (red line) and rectified EMG. ( j ) Speed of length change (negative values are lengthening) for
tendon (blue) and MTU (black), expressed in resting muscle lengths (Lm,r) per second. Grey bars indicate downstroke.
cross-covariance on muscle length and elbow position, cropped to
individual wingbeats (n ¼ 66).
(d) Tendon mechanical testing
We performed tendon mechanical testing to compare mechanical
properties of the tendons with their in vivo mechanical behaviour; methodology followed a published study [26]. Briefly,
MTUs (n ¼ 15) were excised from bats that had been frozen
shortly after death. The three head muscle complex was dissected
in isotonic saline and pennation angle, muscle fascicle length
and muscle mass were measured for each head to calculate
muscle physiological cross-sectional area (PCSA) [27]. Theoretical
peak isometric muscle force, achievable if the muscle complex was
simultaneously and maximally activated during an isometric contraction, was calculated as the product of PCSA (0.093 +
0.009 cm2; n ¼ 15 muscle complexes) and a reasonable estimate
of specific tension for mammalian skeletal muscle of 230 kPa
[27]. This value was chosen from the wide range of published
values because it came from fast-fibred cat muscle, and bat
flight muscles are composed mainly of fast fibres [28].
During mechanical testing, the proximal tendon sheet was
secured in a fixed clamp and a stiff Kevlar line anchored the
distal sesamoid to the lever of an ergometer (Aurora Scientific,
305B-LR). Markers for motion tracking were painted on the free
tendon using a fine ink pen. Each tendon was subjected to 19
stretch–shortening cycles (90 s cycle duration, 12.5% L0). Figure 5
shows the penultimate period of stretch for seven tendon specimens for which testing was successful. Each trial was video
recorded (Photron Fastcam 1024 PCI). Surface marker positions
were digitized frame by frame in XRAYPROJECT [22] to calculate
instantaneous tendon strain during ramp lengthening, free from
artefact owing to tendon slippage or end effects. After testing, tendons were embedded in cryohistological medium, and 15 mm thick
cross sections were taken on a cryotome and photographed.
Tendon cross-sectional area (0.34 + 0.04 mm2) was measured in
IMAGEJ as the average of seven measurements from tendon sections
where rupture did not occur (n ¼ 15). The agreement between
in vivo estimates of tendon mechanical behaviour and tendon behaviour predicted from mechanical testing and estimated muscle stress
was then assessed. This was done by comparing in vitro measurements of tendon stress versus strain with in vivo estimates of
Proc. R. Soc. B 282: 20151832
120
Downloaded from http://rspb.royalsocietypublishing.org/ on June 17, 2017
tendon strain from the XROMM model versus peak isometric
muscle force for the three muscle heads of triceps.
(a)
4
1.22
4. Discussion
The main goal of our study was to determine the mechanical
function of a bat wing tendon: bat wing bones are highly
elongated [29], and wing muscles such as triceps insert via
tendons that appear elongated compared with those of
other small non-volant mammals. The influence of tendon
compliance is apparent in the phase lag between muscle
fascicle strain and joint angle (illustrated schematically in
figure 1b), which is 17% of the wingbeat cycle on average
(figure 4). The pattern of strain is also different for the triceps
muscle fascicles and the MTU, indicating that much of the
elbow joint motion is owing to tendon action rather than due
to muscle fascicle length change (figure 3). We suggest that
stretch of the tendon owing to elbow flexion may cause a
eØ = 29%
Lm/Lr
1.00
0.89
i
iii
iv
n = 19
0.78
(b)
Proc. R. Soc. B 282: 20151832
1.22
1.00
eØ = 23%
Lm/Lr
1.11
0.89
n = 18
0.78
(c)
1.22
1.00
0.89
eØ = 24%
Lm/Lr
1.11
n = 27
0.78
(d)
1.22
1.00
eØ = 14%
1.11
Lm/Lr
XROMM modelling revealed the expected stretch– shortening
cycle pattern for a monoarticular extensor MTU: lengthening
during elbow flexion and shortening during elbow extension
(figure 2g,h). The elbow underwent excursion of 59 + 148
(mean + s.d.), and the length change of the MTU was
1.7 + 0.3 mm (6.5 + 0.8% strain). The observed change in
muscle fascicle length 1.2 + 0.2 mm (16.7 + 0.45% strain;
figure 2i) was significantly less than the change in length of
the whole MTU (repeated-measures ANOVA; F3,118 ¼ 45.52;
p , 0.001). The pattern of muscle fascicle length change differed
substantially from the patterns of MTU stretch–shortening
and elbow flexion–extension (figure 2g–i).
The relationship between muscle fascicle length and
elbow joint position showed distinct phases for all bats
(figure 3). In three bats (figure 3a–c), four distinct phases
were apparent: (i) initiation of elbow flexion, which coincided
with stretch of the muscle fascicles; (ii) the primary phase of
elbow flexion, during which the muscle fascicles remained
near-isometric and the tendon underwent the majority of
stretch; (iii) initiation of elbow extension, during which the
muscle fascicles shortened; and (iv) the primary phase of
elbow extension, coincident with near-constant or slightly
increasing muscle fascicle length, the main period of tendon
recoil. In one bat (figure 3d), phase (iii) was defined by
muscle shortening, not by joint extension and phase
(iv) was prolonged, so that the majority of elbow extension
occurred with the muscle fascicles remaining near-isometric.
Electromyographic recordings (figure 2i) showed that the triceps was activated during phase (ii) (mid-upstroke) and
remained active throughout phase (iv) (downstroke –
upstroke transition). A cross-covariance analysis of the
phase-relationship between muscle length change and joint
motion showed that elbow flexion–extension lagged stretch–
shortening of the triceps muscle fascicles by 17 + 7% of the
wingbeat duration (n ¼ 66 wingbeats; figure 4).
During elbow flexion, the tendon stretched 9.1 + 1.7%
(figure 2h). The lengthening strains observed in the tendon
in vivo fell within the range predicted based on ex vivo
measurements of tendon stress–strain behaviour and calculations of peak isometric force based on muscle PCSA
(2.1 + 0.31 N; mean + s.d., n ¼ 15 muscles; figure 5).
ii
0.89
n = 22
0.78
40
rspb.royalsocietypublishing.org
1.11
3. Results
60
80 100
qelbow(°)
120
140
Figure 3. Relationship between triceps muscle length and elbow angle
during ascending flight. Data from four C. perspicillata individuals (a – d)
for a total of 86 wingbeats. Muscle length is normalized by its resting
length. Coloured dashed lines indicate the robust means regression relationship between triceps MTU length change and elbow angle (electronic
supplementary material, figure S2). This regression also predicts the relationship between muscle length and elbow angle for an MTU with a
non-compliant tendon (i.e. a stiff system). Grey projections to the y-axes
(and dashed arrows in a) indicate the muscle fascicle length change required
in a stiff system to actuate the average amount of elbow joint motion
observed in vivo. Vertical double-arrows indicate the mean shortening of
triceps muscle fascicles across wingbeats. The magnitude of muscle shortening relative to the predicted value for the condition of no tendon compliance,
epsilon, was 22.5% lower, averaged over all individuals and trials.
Downloaded from http://rspb.royalsocietypublishing.org/ on June 17, 2017
elbow extension speed (Kdeg s–1)
22
111
3
50
2
25
1
n = 48
0
0
0
0.025 0.050 0.075 0.100
tendon strain (DL – Lr /Lr)
peak isometric force (N)
75
n = 15
tendon stress (MPa)
Figure 4. Frequency distribution (expressed as % of wingbeats) of phase lags
(given in % of wingbeat duration) measured between triceps muscle fascicle
length change and elbow joint position waveforms in 66 wingbeats across n ¼
4 bats. Vertical dashed line indicates zero lag as expected for a stiff system. Data
to the right of this line represent muscle shortening prior to joint extension.
0.125
Figure 5. Quasi-static mechanical testing of tendons. In C. perspicillata, the triceps complex comprises three muscle heads that insert via a common tendon
and stress–strain curves are shown for seven such tendons (identified by
colour). Circles indicate tendon strain corresponding to the estimated peak isometric force for the muscle complex. Grey projection to y-axis indicates the peak
isometric force range for the set of seven muscle complexes. The vertical box plot
provides peak isometric force values for an additional sample of 15 triceps
muscle complexes where tendon stress–strain data were not obtained. Grey
projection to x-axis indicates the predicted range of tendon strains based on
this set of values. Horizontal box plot shows in vivo tendon strains, as calculated
from XROMM models of 48 ascending flight wingbeats.
reduction in the magnitude of muscle-shortening strain to generate elbow extension by 22.5% on average compared with a
hypothetical stiff system lacking tendon compliance (figure 3).
Because of the relaxation of the temporal link between muscle
contraction and joint motion, the rate of elbow extension
remains high (1511 + 3178 s21), regardless of the mode of
muscle contraction during elbow extension (figure 6).
(a) Mechanical action of tendon in flight
Our measurements support the hypothesis that elastic action
of the triceps tendon in C. perspicillata recycles mechanical
energy across stroke transitions during ascending flight.
The observed patterns of elbow motion and tendon length
change indicate that energy is loaded into the tendon by
forces associated with deceleration of the wing during elbow
flexion late in downstroke. Although the patterns appear
consistent with tendon loading by inertial forces, we cannot
discriminate between contributions from inertial, aerodynamic
5
2.0
1.5
1.0
0.5
0
–10
lengthening
shortening
–5
5
0
muscle speed (Lm s–1)
10
Figure 6. Tendon elastic action relaxes the relationship between speed of
elbow extension and triceps muscle contraction during upstroke of ascending
flight. Mean elbow extension speed plotted against mean muscle fascicle
contraction speed (n ¼ 107 wingbeats, circles). The vertical dotted line indicates isometric muscle behaviour, and the dashed line shows the relationship
for non-compliant tendon, where the speed of muscle shortening is directly
linked to elbow extension speed.
and antagonist muscle force. The muscle fascicles generate
tension during the transition between downstroke and
upstroke, which allows energy to be stored in the tendon,
and muscle contraction may contribute further energy storage.
Energy is released via tendon recoil to power elbow extension
during late upstroke, a critical element in positioning the
wing for generating aerodynamic force production during
downstroke.
Bats may not be the only animals to use elastic energy
storage in wing muscles. A study of the pigeon humerotriceps,
the evolutionary homologue of the muscle studied here, provided similar evidence of a phase lag between elbow motion
and muscle fascicle length change, indicating that elbow flexion occurs with near-isometric muscle fascicles [15]. This
phase lag suggests an influence of tendon elasticity, as hypothesized previously [30,31]. An important role for tendon
elasticity in the primary elevator of the wing, the supracoracoideus, has also been shown in pigeons. Shortening of
supracoracoideus muscle fascicles precedes the initiation of
upstroke by greater than 10% of the wingbeat cycle, indicating
that stretch and recoil of the slender tendon helps power
upstroke [16].
In other flight muscles, there appears to be little role for
tendon elastic action. In the avian pectoralis, the relationship
between muscle fascicle length and joint position is relatively
direct, a finding that is consistent with limited series elastic
compliance [15,16,32,33]. In this stiff system, near-direct
transmission of muscle strain to joint motion means that the
pectoralis acts primarily as a controller of wing element position
(figure 1a). By contrast, our results suggest that the bat triceps
acts as a force controller (figure 1b), where joint motion emerges
from the mechanical action of compliant tendon [34].
Although many other wing muscles in bats and birds also
insert via long tendons, it is important to note that tendon
presence alone is not sufficient to establish that significant
elastic energy storage and recovery occurs. Many tendons
are too stiff relative to the force generation capacity of their
muscle to lead to significant recovery of mechanical energy
[11]. For example, the mallard gastrocnemius tendon shows
only slight evidence of elastic energy recovery during swimming and running, because the gastrocnemius musculature
Proc. R. Soc. B 282: 20151832
0
–30 –20 –10 0 10 20 30 40 50 60
% wingbeat duration
2.5
rspb.royalsocietypublishing.org
% of wingbeats
33
Downloaded from http://rspb.royalsocietypublishing.org/ on June 17, 2017
(b) Force- versus position-based control of jointed
systems
Ethics. All experiments and animal husbandry complied with a proto-
Our finding in bats of a phase lag between muscle contraction
and joint movement of 17% wingbeat duration (figure 4)
indicates that the mechanical function is more likely to be
force- than position-based. Human-engineered jointed systems
typically rely on direct position control (figure 1a), because
designs that avoid compliant elements in series with the
actuator tend to simplify the control algorithms. However, an
actuator in series with a spring (figure 1b) allows for a
force-based control strategy that provides several potential
advantages over direct position control [2,34]. These advantages have only recently been harnessed in designs of
ground-locomoting robots [40]. The possible advantages of
series elastic compliance could be valuable for bioinspiration
Data accessibility. Data used in this study are available from the XMA
5. Conclusion
A tendon spring in the bat wing relaxes the temporal
link between triceps muscle fascicle shortening and elbow
joint motion, resulting in storage of inertial energy associated
with wing element deceleration. Tendon elastic energy is then
recovered to aid in elbow extension. The series elastic compliance in this system also reduces the muscle-shortening strain
required to extend the elbow. These findings broaden our
knowledge about the occurrence and function of biological springs by providing a robust demonstration of the
importance of recovery of elastic energy from a limb muscle
tendon to effect fluid-based locomotion in a small mammal.
Detailed modelling could be employed in future studies to
quantitatively estimate energy stored and released in bat
wing muscle tendons.
col approved by the Brown University IACUC and with USDA
regulations.
portal at http://tinyurl.com/nqg9jzz
Competing interests. We declare that we have no competing interests.
Funding. Supported by AFOSR (FA9550-12-1-0301, monitored by
P. Bradshaw to S.M.S. and T.J.R.) and the National Science
Foundation (IOS 1145549 to S.M.S.).
Acknowledgements. E. Tavares, E. Longman, M. L. Hedberg and the
Brown University Animal Care Facility staff assisted with animal husbandry and experiments. H. Walker, M. O’Connor, D. Podlisny,
E. Longman and L. Wyatt digitized videos. D. Boerma and R. Brown
assisted with rotoscoping. I. Wallace, Stony Brook University, and
N. Simmons, American Museum of Natural History, generously
provided mCT scans. We thank C. Anderson, C. Arellano, J. Bahlman,
D. Boerma, R. Carney, D. Sustaita and two anonymous reviewers for
their comments on a draft of the manuscript.
References
1.
2.
3.
4.
5.
6.
Pollock CM, Shadwick RE. 1994 Relationship
between body mass and biomechanical properties
of limb tendons in adult mammals. Am. J. Physiol.
266, R1016–R1021.
Alexander RM. 2002 Tendon elasticity and muscle
function. Comp. Biochem. Physiol. A, Mol. Integr.
Physiol. 133, 1001–1011. (doi:10.1016/S10956433(02)00143-5)
Biewener A, Konieczynski D, Baudinette R. 1998
In vivo muscle force-length behavior during steadyspeed hopping in tammar wallabies. J. Exp. Biol.
201, 1681 –1694.
Alexander RM, Bennet-Clark HC. 1977 Storage of elastic
strain energy in muscle and other tissues. Nature 265,
114–117. (doi:10.1038/265114a0)
Roberts TJ, Azizi E. 2011 Flexible mechanisms: the
diverse roles of biological springs in vertebrate
movement. J. Exp. Biol. 214, 353 –361. (doi:10.
1242/jeb.038588)
Cavagna GA, Heglund NC, Taylor CR. 1977
Mechanical work in terrestrial locomotion: two basic
mechanisms for minimizing energy expenditure.
Am. J. Physiol. Regul. Integr. Comp. Physiol. 233,
R243 –R261.
7. Ker RF, Bennett MB, Bibby SR, Kester RC, Alexander
RM. 1987 The spring in the arch of the human foot.
Nature 325, 147–149. (doi:10.1038/325147a0)
8. Roberts TJ, Marsh RL, Weyand PG, Taylor CR. 1997
Muscular force in running turkeys: the economy of
minimizing work. Science 275, 1113 –1115. (doi:10.
1126/science.275.5303.1113)
9. Wilson AM, Watson JC, Lichtwark GA. 2003
Biomechanics: a catapult action for rapid limb
protraction. Nature 421, 35–36. (doi:10.1038/421035a)
10. Bullimore SR, Burn JF. 2005 Scaling of elastic energy
storage in mammalian limb tendons: do small
mammals really lose out? Biol. Lett. 1, 57 –59.
(doi:10.1098/rsbl.2004.0243)
11. Ker RF, Alexander RM, Bennett MB. 1988 Why are
mammalian tendons so thick? J. Zool. 216, 309 –
324. (doi:10.1111/j.1469-7998.1988.tb02432.x)
12. Biewener AA, Blickhan R. 1988 Kangaroo rat
locomotion: design for elastic energy storage or
acceleration? J. Exp. Biol. 140, 243–255.
13. Askew GN, Marsh RL. 2002 Muscle
designed for maximum short-term power
output: quail flight muscle. J. Exp. Biol. 205,
2153– 2160.
14. Blickhan R, Cheng JY. 1994 Energy storage by elastic
mechanisms in the tail of large swimmers: a
reevaluation. J. Theor. Biol. 168, 315–321. (doi:10.
1006/jtbi.1994.1112)
15. Robertson AMB, Biewener AA. 2012 Muscle function
during takeoff and landing flight in the pigeon
(Columba livia). J. Exp. Biol. 215, 4104–4114.
(doi:10.1242/jeb.075275)
16. Tobalske BW, Biewener AA. 2008 Contractile
properties of the pigeon supracoracoideus during
different modes of flight. J. Exp. Biol. 211,
170–179. (doi:10.1242/jeb.007476)
17. Hildebrand M. 1988 Form and function in vertebrate
feeding and locomotion. Am. Zool. 28, 727 –738.
(doi:10.2307/3883300)
18. Van den Berg C, Rayner JMV. 1995 The moment of
inertia of bird wings and the inertial power requirement
for flapping flight. J. Exp. Biol. 198, 1655–1664.
6
Proc. R. Soc. B 282: 20151832
with the accelerating development of small flying robots and
autonomous flapping micro air-vehicles.
rspb.royalsocietypublishing.org
does not generate high enough forces to load energy into the
tendon [35]. More broadly, it has been suggested that the
limb tendons of small mammals tend to be mechanically
‘overbuilt’ because of scaling patterns and that the associated
low tendon strains limits the role of elastic energy recovery in
small animals [11,12,36,37]. However, this suggestion is not
supported by our results for the bat triceps tendon. In
addition, long tendons decrease the limb mass moment of
inertia [2,17] and can contribute to reducing inertial costs,
which may be important during flight [18–20]. Hence, the
presence of long tendons, compliant or not, probably benefits
the mechanical function of vertebrate wings.
In undulating swimmers, a role for tendon action in energy
recycling across stroke transitions has been hypothesized
for tail tendons. However, direct force measurements from
tail tendons in thunniform swimmers suggest a function as
inextensible force transmitters [38], and analyses of cetacean
fluke tendons have yielded conflicting conclusions [14,39].
Downloaded from http://rspb.royalsocietypublishing.org/ on June 17, 2017
27.
28.
30.
31.
32.
33.
34. Pratt GA, Williamson MM, Dillworth P, Pratt J, Wright
A. 1995 Stiffness isn’t everything. In Experimental
robotics IV (eds O Khatib, JK Salisbury), pp. 253–262.
Berlin, Germany: Springer-Verlag.
35. Biewener AA, Corning WR. 2001 Dynamics of
mallard (Anas platyrynchos) gastrocnemius function
during swimming versus terrestrial locomotion.
J. Exp. Biol. 204, 1745–1756.
36. Biewener A, Alexander RM, Heglund NC. 1981
Elastic energy storage in the hopping of
kangaroo rats (Dipodomys spectabilis). J. Zool.
195, 369–383. (doi:10.1111/j.1469-7998.1981.
tb03471.x)
37. Pollock CM, Shadwick RE. 1994 Allometry of muscle,
tendon, and elastic energy storage capacity in
mammals. Am. J. Physiol. Regul. Integr. Comp.
Physiol. 266, R1022–R1031.
38. Knower AT. 1998 Biomechanics of thunniform
swimming: electromyography, kinematics, and caudal
tendon function in the yellowfin tuna Thunnus
albacares and the skipjack tuna Katsuwonus
pelamis. San Diego, CA: UC.
39. Bennett MB, Ker RF, Alexander RM. 1987 Elastic
properties of structures in the tails of cetaceans
(Phocaena and Lagenorhynchus) and their
effect on the energy cost of swimming. J. Zool.
211, 177–192. (doi:10.1111/j.1469-7998.1987.
tb07461.x)
40. Scarfogliero U, Stefanini C, Dario P. 2009 The use
of compliant joints and elastic energy storage in
bio-inspired legged robots. Mech. Mach. Theor.
44, 580 –590. (doi:10.1016/j.mechmachtheory.
2008.08.010)
7
Proc. R. Soc. B 282: 20151832
29.
J. Exp. Biol. 215, 3552 –3558. (doi:10.1242/jeb.
072728)
Spector SA, Gardiner PF, Zernicke RF, Roy RR,
Edgerton VR. 1980 Muscle architecture and forcevelocity characteristics of cat soleus and medial
gastrocnemius: implications for motor control. J.
Neurophysiol. 44, 951–960.
Hermanson JW, Ryan JH, Cobb MA, Bentley J,
Schutt WA. 1998 Histochemical and electrophoretic
analysis of the primary flight muscle of several
phyllostomid bats. Can. J. Zool. 76, 1983–1992.
(doi:10.1139/cjz-76-11-1983)
Swartz SM. 1997 Allometric patterning in the limb
skeleton of bats: implications for the mechanics and
energetics of powered flight. J. Morphol. 234,
277 –294. (doi:10.1002/(sici)1097-4687(199712)
234:3,277::aid-jmor6.3.0.co;2-6)
Dial KP, Jenkins FA. 1991 The functional anatomy of
the shoulder in the European starling (Sturnus
vulgaris). J. Morphol. 207, 327–344. (doi:10.1002/
jmor.1052070309)
Dial KP. 1992 Activity patterns of the wing muscles
of the pigeon (Columba livia) during different
modes of flight. J. Exp. Zool. 262, 357–373.
(doi:10.1002/jez.1402620402)
Hedrick TL, Tobalske BW, Biewener AA. 2003 How
cockatiels (Nymphicus hollandicus) modulate
pectoralis power output across flight speeds. J. Exp.
Biol. 206, 1363 –1378. (doi:10.1242/jeb.00272)
Tobalske BW, Biewener AA, Warrick DR, Hedrick TL,
Powers DR. 2010 Effects of flight speed upon
muscle activity in hummingbirds. J. Exp. Biol. 213,
2515 –2523. (doi:10.1242/jeb.043844)
rspb.royalsocietypublishing.org
19. Iriarte-Diaz J, Riskin DK, Willis DJ, Breuer KS, Swartz
SM. 2011 Whole-body kinematics of a fruit bat
reveal the influence of wing inertia on body
accelerations. J. Exp. Biol. 214, 1546 –1553.
(doi:10.1242/jeb.037804)
20. Riskin DK, Bergou A, Breuer KS, Swartz SM. 2012
Upstroke wing flexion and the inertial cost of bat
flight. Proc. R. Soc. B 279, 2945– 2950. (doi:10.
1098/rspb.2012.0346)
21. Astley HC, Roberts TJ. 2012 Evidence for a vertebrate
catapult: elastic energy storage in the plantaris
tendon during frog jumping. Biol. Lett. 8, 386–389.
(doi:10.1098/rsbl.2011.0982)
22. Brainerd EL, Baier DB, Gatesy SM, Hedrick TL,
Metzger KA, Gilbert SL, Crisco JJ. 2010 X-ray
reconstruction of moving morphology (XROMM):
precision, accuracy and applications in comparative
biomechanics research. J. Exp. Zool. A Ecol. Genet.
Physiol. 313, 262–279. (doi:10.1002/jez.589)
23. Gatesy SM, Baier DB, Jenkins FA, Dial KP. 2010
Scientific rotoscoping: a morphology-based
method of 3-D motion analysis and visualization. J. Exp.
Zool. A Ecol. Genet. Physiol. 313, 244–261. (doi:10.
1002/jez.588)
24. Basmajian J, Stecko G. 1962 A new bipolar electrode
for electromyography. J. Appl. Physiol. 17, 849.
25. Lutz GJ, Rome LC. 1996 Muscle function during
jumping in frogs. I. Sarcomere length change, EMG
pattern, and jumping performance. Am. J. Physiol.
271, C563–C570.
26. Matson A, Konow N, Miller S, Konow PP, Roberts TJ.
2012 Tendon material properties vary and are
interdependent among turkey hindlimb muscles.