The effect of fast and slow motor unit activation on whole-muscle mechanical performance: the size principle may not pose a mechanical paradox.

The output of skeletal muscle can be varied by selectively recruiting different motor units. However, our knowledge of muscle function is largely derived from muscle in which all motor units are activated. This discrepancy may limit our understanding of in vivo muscle function. Hence, this study aimed to characterize the mechanical properties of muscle with different motor unit activation. We determined the isometric properties and isotonic force-velocity relationship of rat plantaris muscles in situ with all of the muscle active, 30% of the muscle containing predominately slower motor units active or 20% of the muscle containing predominately faster motor units active. There was a significant effect of active motor unit type on isometric force rise time (p < 0.001) and the force-velocity relationship (p < 0.001). Surprisingly, force rise time was longer and maximum shortening velocity higher when all motor units were active than when either fast or slow motor units were selectively activated. We propose this is due to the greater relative effects of factors such as series compliance and muscle resistance to shortening during sub-maximal contractions. The findings presented here suggest that recruitment according to the size principle, where slow motor units are activated first and faster ones recruited as demand increases, may not pose a mechanical paradox, as has been previously suggested.

Modeling muscle function using experimentally determined subject-specific muscle properties.

Muscle models are commonly based on intrinsic properties pooled across a number of individuals, often from a different species, and rarely validated against directly measured muscle forces. Here we use a rich data set of rat medial gastrocnemius muscle forces recorded during in-situ and in-vivo isometric, isotonic, and cyclic contractions to test the accuracy of forces predicted using Hill-type muscle models. We identified force-length and force-velocity parameters for each individual, and used either these subject-specific intrinsic properties, or population-averaged properties within the models. The modeled forces for cyclic in-vivo and in-situ contractions matched with measured muscle-tendon forces with r2 between 0.70 and 0.86, and root-mean square errors (RMSE) of 0.10 to 0.13 (values normalized to the maximum isometric force). The modeled forces were least accurate at the highest movement and cycle frequencies and did not show an improvement in r2 when subject-specific intrinsic properties were used; however, there was a reduction in the RMSE with fewer predictions having higher errors. We additionally recorded and tested muscle models specific to proximal and distal regions of the muscle and compared them to measures and models from the whole muscle belly: there was no improvement in model performance when using data from specific anatomical regions. These results show that Hill-type muscle models can yield very good performance for cyclic contractions typical of locomotion, with small reductions in errors when subject-specific intrinsic properties are used.

Biomechanics of mammalian terrestrial locomotion.

Mammalian skeletons experience peak locomotor stresses (force per area) that are 25 to 50% of their failure strength, indicating a safety factor of between two and four. The mechanism by which animals achieve a constant safety factor varies depending on the size of the animal. Over much of their size (0.1 to 300 kilograms), larger mammals maintain uniform skeletal stress primarily by having a more upright posture, which decreases mass-specific muscle force by increasing muscle mechanical advantage. At greater sizes, increased skeletal allometry and decreased locomotor performance likely maintain stresses constant. At smaller sizes, skeletal stiffness may be more critical than strength. The decrease in mass-specific muscle force in mammals weighing 0.1 to 300 kilogram indicates that peak muscle stresses are also constant and correlates with a decrease in mass-specific energy cost of locomotion. The consistent pattern of locomotor stresses developed in long bones at different speeds and gaits within a species may have important implications for how bones adaptively remodel to changes in stress.

Scaling body support in mammals: limb posture and muscle mechanics.

The scaling of bone and muscle geometry in mammals suggests that peak stresses (ratio of force to cross-sectional area) acting in these two support elements increase with increasing body size. Observations of stresses acting in the limb bones of different sized mammals during strenuous activity, however, indicate that peak bone stress is independent of size (maintaining a safety factor of between 2 and 4). It appears that similar peak bone stresses and muscle stresses in large and small mammals are achieved primarily by a size-dependent change in locomotor limb posture: small animals run with crouched postures, whereas larger species run more upright. By adopting an upright posture, large animals align their limbs more closely with the ground reaction force, substantially reducing the forces that their muscles must exert (proportional to body mass) and hence, the forces that their bones must resist, to counteract joint moments. This change in limb posture to maintain locomotor stresses within safe limits, however, likely limits the maneuverability and accelerative capability of large animals.

Hind limb scaling of kangaroos and wallabies (superfamily Macropodoidea): implications for hopping performance, safety factor and elastic savings.

The aim of this study was to examine hind limb scaling of the musculoskeletal system in the Macropodoidea, the superfamily containing wallabies and kangaroos, to re-examine the effect of size on the locomotor mechanics and physiology of marsupial hopping. Morphometric musculoskeletal analyses were conducted of 15 species and skeletal specimens of 21 species spanning a size range from 0.8 to 80 kg that included representatives of 12 of the 16 extant genera of macropodoids. We found that unlike other groups, macropodoids are able to match force demands associated with increasing body size primarily through a combination of positive allometry in muscle area and muscle moment arms. Isometric scaling of primary hind limb bones suggests, however, that larger species experience relatively greater bone stresses. Muscle to tendon area ratios of the ankle extensors scale with strong positive allometry, indicating that peak tendon stresses also increase with increasing body size but to a lesser degree than previously reported. Consistent with previous morphological and experimental studies, large macropodoids are therefore better suited for elastic strain energy recovery but operate at lower safety factors, which likely poses an upper limit to body size. Scaling patterns for extant macropodoids suggest that extinct giant kangaroos (approximately 250 kg) were likely limited in locomotor capacity.

Differential design for hopping in two species of wallabies.

Hindlimb musculoskeletal anatomy and steady speed over ground hopping mechanics were compared in two species of macropod marsupials, tammar wallabies and yellow-footed rock wallabies (YFRW). These two species are relatively closely related and are of similar size and general body plan, yet they inhabit different environments with presumably different musculoskeletal demands. Tammar wallabies live in relatively flat, open habitat whereas yellow-footed rock wallabies inhabit steep cliff faces. The goal of this study was to explore musculoskeletal differences between tammar wallabies and yellow-footed rock wallabies and determine how these differences influence each species' hopping mechanics. We found the cross-sectional area of the combined ankle extensor tendons of yellow-footed rock wallabies was 13% greater than that of tammar wallabies. Both species experienced similar ankle joint moments during steady-speed hopping, however due to a lower mechanical advantage at this joint, tammar wallabies produced 26% more muscle force. Thus, during moderate speed hopping, yellow-footed rock wallabies operated with 38% higher tendon safety factors, while tammar wallabies were able to store 73% more elastic strain energy (2.18 J per leg vs. 1.26 J in YFRW). This likely reflects the differing demands of the environments inhabited by these two species, where selection for non-steady locomotor performance in rocky terrain likely requires trade-offs in locomotor economy.

Different Segments within Vertebrate Muscles Can Operate on Different Regions of Their Force-Length Relationships.

To relate in vivo behavior of fascicle segments within a muscle to their in vitro force-length relationships, we examined the strain behavior of paired segments within each of three vertebrate muscles. After determining in vivo muscle activity patterns and length changes of in-series segments within the semimembranosus muscle (SM) in the American Toad (Bufo americanus) during hopping and within the sternohyoid (SH) muscle in the rat (Rattus rattus) during swallowing, and of spatially separated fascicles within the medial gastrocnemius (MG) muscle in the rat during trotting, we measured their corresponding in vitro (toad) or in situ (rat) force-length relationships (FLRs). For all three muscles, in vivo strain heterogeneity lasted for about 36-57% of the behavior cycle, during which one segment or fascicle shortened while the other segment or fascicle simultaneously lengthened. In the toad SM, the proximal segment shortened from the descending limb across the plateau of its FLR from 1.12 to 0.91 of its optimal length (Lo), while the distal segment lengthened (by 0.04 ± 0.04 Lo) before shortening down the ascending limb from 0.94 to 0.83 Lo. In the rat SH muscle, the proximal segment tended to shorten on its ascending limb from 0.90 to 0.85 Lo while the distal segment tended to lengthen across Lo (0.96-1.12 Lo). In the rat MG muscle, in vivo strains of proximal fascicles ranged from 0.72 to 1.02 Lo, while the distal fascicles ranged from 0.88 to 1.11 Lo. Even though the timing of muscle activation patterns were similar between segments, the heterogeneous strain patterns of fascicle segments measured in vivo coincided with different operating ranges across their FLRs simultaneously, implying differences in force-velocity behavior as well. The three vertebrate skeletal muscles represent a diversity of fiber architectures and functions and suggest that patterns of in vivo contractile strain and the operating range over the FLR in one muscle region does not necessarily represent other regions within the same muscle.

Effects of load carrying on metabolic cost and hindlimb muscle dynamics in guinea fowl (Numida meleagris).

The goal of this study was to test whether the contractile patterns of two major hindlimb extensors of guinea fowl are altered by load-carrying exercise. We hypothesized that changes in contractile pattern, specifically a decrease in muscle shortening velocity or enhanced stretch activation, would result in a reduction in locomotor energy cost relative to the load carried. We also anticipated that changes in kinematics would reflect underlying changes in muscle strain. Oxygen consumption, muscle activation intensity, and fascicle strain rate were measured over a range of speeds while animals ran unloaded vs. when they carried a trunk load equal to 22% of their body mass. Our results showed that loading produced no significant (P > 0.05) changes in kinematic patterns at any speed. In vivo muscle contractile strain patterns in the iliotibialis lateralis pars postacetabularis and the medial head of the gastrocnemius showed a significant increase in active stretch early in stance (P < 0.01), but muscle fascicle shortening velocity was not significantly affected by load carrying. The rate of oxygen consumption increased by 17% (P < 0.01) during loaded conditions, equivalent to 77% of the relative increase in mass. Additionally, relative increases in EMG intensity (quantified as mean spike amplitude) indicated less than proportional recruitment, consistent with force enhancement via stretch activation, in the proximal iliotibialis lateralis pars postacetabularis; however, a greater than proportional increase in the medial gastrocnemius was observed. As a result, when averaged for the two muscles, EMG intensity increased in direct proportion to the fractional increase in load carried.

Adaptive changes in trabecular architecture in relation to functional strain patterns and disuse.

Principal strains and their orientation, determined from in vivo and in situ strains recorded from the lateral cortical surface of the calcaneus of potoroos (a small marsupial) during treadmill exercise and tension applied via the Achilles tendon, were compared with the underlying trabecular architecture and its alignment to test Wolff's "trajectorial theory" of trabecular alignment. In vivo and in situ principal compressive strains (-800 to -2000 mu e) were found to be aligned (mean 161 +/- 7 degrees) close to the preferred alignment (160 degrees) of underlying trabeculae within the calcaneal metaphysis [a second trabecular arcade was closely aligned (70 degrees) with the direction (71 degrees) of principal tensile strain]. This finding represents quantitative verification of Wolff's trajectorial theory of trabecular alignment. These trabecular alignments, as measured by trabecular anisotropy (TbAn, the ratio of horizontal: vertical intercepts), remained unchanged (p > 0.05) after 8 weeks of disuse. However, trabecular bone volume fraction (BV/TV, -35%), trabecular thickness (TbTh, -25%), and trabecular number (TbN, -16%) were reduced for the tenotomized calcaneii relative to their contralateral controls (p < 0.001 to < 0.003). The reduction in trabecular number was associated with a corresponding increase in trabecular spacing (TbSp, +30%). Together, these results suggest that once trabecular alignment is established during growth (along the directions of principal strain during locomotion), it is not altered when functional strains are removed.

Structural response of growing bone to exercise and disuse.

The limbs of growing chicks (2-12 wk of age) were subjected to differing conditions of mechanical use to examine the effect of extrinsic loading on bone modeling early in postnatal growth. One group of chicks was subjected to intensive exercise by running on a treadmill 5 days/wk at 60% maximum speed while carrying on their trunk a load equal to 20% of body weight (EXER). In a second group, weight-bearing function was eliminated by sciatic denervation of one hindlimb at 2 wk of age (DNV). A third group grew under sedentary conditions (SED). Comparisons among groups were made on the basis of bone mass and length, cortical cross-sectional area and second moment of area, cortical thickness, longitudinal curvature, and % ash. After normalizing for growth-related differences in body mass among the three groups, we found that exercise led to an overall 16 +/- 13% increase in cortical cross-sectional area and a 26 +/- 21% increase in second moment of area measured at proximal, midshaft, and distal levels of the bone compared with values of SED animals. These increases in cortical geometry corresponded to a 10% increase in total bone mass and were generally established by 8 wk of age (6 wk of training) and maintained to 12 wk of age. When deprived of functional use, the growing bones of DNV animals were reduced in mass (-19%), cortical area (-8 +/- 7%), and second moment of area (-11 +/- 9%) compared with SED animals. DNV tibiotarsi were also significantly shorter (7% at 8 wk and 14% at 12 wk); however, the contralateral load-bearing tibiotarsus of the DNV animals was similarly reduced compared with SED and EXER animals, suggesting a general reduction of growth in the DNV group. Even more pronounced than the reductions in bone mass and area, however, were the loss of normal longitudinal curvature and an increase in the variability of cross-sectional shape and cortical thickness of the DNV tibiotarsi compared with SED and EXER animals.

Evaluation of a bone's in vivo 24-hour loading history for physical exercise compared with background loading.

The present study sought to answer two research questions. First, how distinctive, as a potential osteogenic stimulus, are short-duration bouts of treadmill exercise relative to sedentary background activity? Second, how well does daily effective strain stimulus relate the loading history for one such exercise program, in comparison with other experimental loading programs, to bone formation? In vivo cortical strains were measured in the tibiotarsus of White Leghorn chickens at a late stage of skeletal growth (14-34 weeks old) under the conditions of a previous investigation of bone formation in response to an exercise program (15 min/day, treadmill gait at 60% maximum speed while carrying 20% body mass) that included sedentary background activity. These strain data were compiled into 24-hour loading histories of peak cyclic strain, demonstrating that strains were statistically different for exercise and background activities (p < 0.0001), with both the magnitude and number of cyclic strain events being greater during exercise (generally greater than 500 microstrain, 2,500 cycles/day) than during background activity (generally less than 500 microstrain, mean: 775 cycles/day). Strains during exercise accounted for more than 97% of the daily effective strain stimulus for bone adaptation, despite the fact that exercise comprised only 1% of the daily period (15 min/day). The levels of the daily effective strain stimulus were similar to those calculated for strains engendered by artificial loading of functionally isolated avian ulnae, which either maintained bone mass or resulted in a 15% increase of cortical cross-sectional area in both sets of studies. These results indicate that short-duration bouts of treadmill exercise and sedentary background activity can represent distinct osteogenic stimuli for adaptive bone modeling. They also provide experimental support for the use of a daily effective strain stimulus to quantify skeletal loading histories for differing programs of physical exercise, although the relative importance of other mechanical and nonmechanical factors requires further investigation.

Musculoskeletal design in relation to body size.

Irrespective of body size and phylogenetic diversity, the skeletal systems of terrestrial mammals are built of tissue components having similar mechanical properties and material organization. Because of scale effects on skeletal form, therefore, larger mammals increase the effective mechanical advantage of their limbs to decrease mass-specific forces associated with the support of gravitational loads imposed during locomotion to maintain a similar safety factor. Larger animals accomplish this by adopting a more upright posture while running, which aligns their limb joints more closely with the resultant ground reaction force, thereby decreasing the mass-specific force that their muscles must generate to support externally applied joint moments. As a result, peak (compressive) bone stresses determined from in vivo bone strain recordings and force platform and kinematic analyses of the limb generally range from -40 to -80 MPa (mean: -55 +/- 23 MPa), corresponding to a safety factor to compressive bone failure of about three to four. The decrease in mass-specific muscle force indicates that the maximum stresses developed in limb muscles of different sized species are also similar at equivalent levels of performance. Stresses developed in the midshafts of most long bones are primarily the result of bending, often engendered by axial forces transmitted about the bone's longitudinal curvature. The consistency of bending-induced skeletal strain over a range of physical activity and the associated expense of increased strain magnitude that this form of loading incurs suggest that functional strain patterns developed through bending may be a desirable architectural objective of most long bones. Alteration of a bone's normal functional strain distribution, therefore, is likely a key factor underlying adaptive remodeling in response to changes in mechanical loading.

The functional morphology of xenarthrous vertebrae in the armadillo Dasypus novemcinctus (Mammalia, Xenarthra).

In order to assess the mechanical properties of xenarthrous vertebrae, and to evaluate the role of xenarthrae as fossorial adaptations, in vitro bending tests were performed on posterior thoracic and lumbar vertebral segments excised from specimens of the armadillo Dasypus novemcinctus and the opossum Didelphis virginiana, the latter being used to represent the primitive mammalian condition. The columns of the two species were subjected to dorsal, ventral, and lateral bending, as well as torsion, in order to determine their stiffness in each of these directions. During these tests, bone strains in the centra of selected vertebrae were determined using rosette strain gages. Overall stiffness of the armadillo backbone at physiologically relevant displacement levels was significantly higher than that of the opossum for both dorsal and lateral bending. The two species also exhibited significant differences in angular displacement of individual vertebrae and in vertebral strain magnitudes and orientations in these two directions. No significant differences were observed when the columns of the two species were subjected to torsion or to ventral bending. Our results suggest that some, but not all, of the mechanical differences between the two species are due to the presence of xenarthrae. For example, removal of the xenarthrae from selected vertebrae (L2-L4) changes strain orientation and shear, but not strain magnitudes. Comparisons with functional data from other digging mammals indicate that the modified mechanical properties of the Dasypus column are consistent with an interpretation of xenarthrae as digging adaptations and lend support to the idea that the order Xenarthra represents an early offshoot of placental mammals specialized for fossoriality.

Experimental alteration of limb posture in the chicken (Gallus gallus) and its bearing on the use of birds as analogs for dinosaur locomotion.

Extant birds represent the only diverse living bipeds, and can be informative for investigations into the life-history parameters of their extinct dinosaurian relatives. However, morphological changes that occurred during early avian evolution, including the unique adoption of a nearly horizontal femoral orientation associated with a shift in center of mass (CM), suggest that caution is warranted in the use of birds as analogs for nonavian dinosaur locomotion. In this study, we fitted a group of white leghorn chickens (Gallus gallus) with a weight suspended posterior to the hip in order to examine the effects on loading and morphology. This caused a CM shift that necessitated a change in femoral posture (by 35 degrees towards the horizontal, P < 0.001), and resulted in reorientation of the ground reaction force (GRF) vector relative to the femur (from 41 degrees to 82 degrees, P < 0.001). Despite similar strain magnitudes, an overall increase in torsion relative to bending (from 1.70 to 1.95 times bending, P < 0.001) was observed, which was weakly associated with a tendency for increased femoral cross-sectional dimensions (P = 0.1). We suggest that a relative increase in torsion is consistent with a change in femoral posture towards the horizontal, since this change increases the degree to which the bone axis and the GRF vector produce mediolateral long-axis rotation of the bone. These results support the hypothesis that a postural change during early avian evolution could underlie the allometric differences seen between bird and nonavian dinosaur femora by requiring more robust femoral dimensions in birds due to an increase in torsion.

Differential scaling of the long bones in the terrestrial carnivora and other mammals.

We measured the lengths and diameters of four long bones from 118 terrestrial carnivoran species using museum specimens. Though intrafamilial regressions scaled linearly, nearly all intraordinal regressions scaled non-linearly. The observed non-linear scaling of bone dimensions within this order results from a systematic decrease in intrafamilial allometric slope with increasing body size. A change in limb posture (more upright in larger species) to maintain similar peak bone stresses may allow the nearly isometric scaling of skeletal dimensions observed in smaller sized mammals (below about 100 kg). However, strong positive allometry is consistently observed in a number of large terrestrial mammals (the largest Carnivora, the large Bovidae, and the Ceratomorpha). This suggests that the capacity to compensate for size increases through alteration of limb posture is limited in extremely large-sized mammals, such that radical changes in bone shape are required to maintain similar levels of peak bone stress.

Bone stress in the horse forelimb during locomotion at different gaits: a comparison of two experimental methods.

Longitudinal stresses acting in the cranial and caudal cortices of the radius and the dorsal and palmar cortices of the metacarpus in the horse were determined using two independent methods simultaneously. One approach involved the use of rosette strain gauges to record in vivo bone strain; the other involved filming the position of the horse's forelimb as it passed over a force plate. Agreement between the two analyses was better for the radius than for the metacarpus. Both methods showed the radius to be loaded primarily in sagittal bending, acting to place the caudal cortex in compression and the cranial cortex in tension. At each gait the magnitude of peak stress in each cortex based on the film/force analysis was 1.5-2 times higher than that determined from the bone strain recordings. In the metacarpus, the magnitude of stress in each cortex calculated from the film/force method was 2-3 times greater at each gait than that shown by the bone strain recordings. However, whereas the film/force analysis indicated that the metacarpus was loaded in sagittal bending (acting to place the palmar cortex in compression and the dorsal cortex in tension), the bone strain recordings showed the metacarpus to be loaded primarily in axial compression at each gait. Because the film/force method depends on an accurate measure of limb segment orientation relative to the direction of ground reaction force, comparatively small errors in calculations of bending moments may lead to a significant difference in the level and distribution of stress determined to act in the bone's cortices. The discrepancy in metacarpal loading obtained by the two methods may be explained in part by the simplicity of the biomechanical model which, for instance, neglected the force exerted by the sesamoids on the distal end of the metacarpus. The records of stress determined from the in vivo bone strain recordings showed that each bone was subjected to a consistent loading regime despite changes of gait. Such a consistent strain distribution should allow these bones to maximize economy in the use of tissue required to support the dynamic loads applied. Peak stresses measured from the bone strain recordings in the radius during locomotion at constant speed (-40.8 +/- 4.1 MN m-2) were significantly larger than those in the metacarpus (-25.1 +/- 2.8 MN m-2), regardless of speed and gait. During acceleration and deceleration, however, peak stress rose dramatically in the metacarpus (-40.6 +/- 3.4 MN m-2) but remained constant in the radius (-37.8 +/- 5.8 MN m-2).(ABSTRACT TRUNCATED AT 400 WORDS)

Muscle-tendon stresses and elastic energy storage during locomotion in the horse.

The stresses acting in muscle-tendon units and ligaments of the forelimb and hindlimb of horses were determined over a range of speed and gait based on recordings of ground reaction forces and limb kinematics. Maximum stresses of 40-50 MPa were calculated to act in several of the principal forelimb (superficial digital flexor (SDF), deep digital flexor (DDF), ulnaris lateralis (UL) and flexor carpi ulnaris/radialis (FCU/R)) and hindlimb tendons (plantaris, DDF) at the fastest galloping speeds recorded (up to 7.4 m s-1). Smaller stresses were found for the gastrocnemius (GAST) tendon (30 MPa) and suspensory ligaments (S-Ligs) (18-25 MPa). Average peak muscle stresses reached 200-240 kPa during galloping. Tendon and muscle stresses increased more steeply with changes of gait and during galloping, than during trotting. Calculations of elastic strain energy storage based on tendon stress showed similar patterns of increase with change of speed and gait, with the greatest contribution to elastic savings by the DDF tendons of the forelimb and hindlimb. In general, the hindlimb contributed two-thirds and the forelimb one-third to overall energy storage. Comparison of tendon elastic energy savings with mechanical work showed a maximum 40% recovery of mechanical work by elastic savings when the horses changed gait from a walk to a slow trot. Percentage of recovery then decreased with increased trotting speed, but increased again with a change of gait to a gallop, reaching 36% recovery at the fastest measured galloping speed (7.4 m s-1). The long length of horse tendons in relation to extremely short pennate muscle fibers suggests a highly specialized design for economical muscle force generation and enhanced elastic energy savings. However, elastic energy savings in terms of percentage of recovery of mechanical work and metabolic energy is less than that observed in wallabies and kangaroos during hopping, but similar to that in humans during running, and greater than that for dogs during trotting and galloping.

A constitutive description of the anisotropic response of the fascia lata.

In this paper we propose a constitutive model to analyze in-plane extension of goat fascia lata. We first perform a histological analysis of the fascia that shows a well-organized bi-layered arrangement of undulated collagen fascicles oriented along two well defined directions. To develop a model consistent with the tissue structure we identify the absolute and relative thickness of each layer and the orientation of the preferred directions. New data are presented showing the mechanical response in uniaxial and planar biaxial extension. The paper proposes a constitutive relation to describe the mechanical response. We provide a summary of the main ingredients of the nonlinear theory of elasticity and introduce a suitable strain-energy function to describe the anisotropic response of the fascia. We validate the model by showing good fit of the numerical results and the experimental data. Comments are included about differences and analogies between goat fascia lata and the human iliotibial band.

Low speed maneuvering flight of the rose-breasted cockatoo (Eolophus roseicapillus). I. Kinematic and neuromuscular control of turning.

Maneuvering flight has long been recognized as an important component of the natural behavior of many bird species, but has been the subject of little experimental work. Here we examine the kinematics and neuromuscular control of turning flight in the rose-breasted cockatoo Eolophus roseicapillus (N=6), testing predictions of maneuvering flight and control based on aerodynamic theory and prior kinematic and neuromuscular studies. Six cockatoos were trained to navigate between two perches placed in an L-shaped flight corridor, making a 90 degrees turn midway through each flight. Flights were recorded with three synchronized high-speed video cameras placed outside the corridor, allowing a three-dimensional reconstruction of wing and body kinematics through the turn. We simultaneously collected electromyography recordings from bilateral implants in the pectoralis, supracoracoideus, biceps brachii and extensor metacarpi radialis muscles. The cockatoos maneuvered using flapping, banked turns with an average turn radius of 0.92 m. The mean rate of change in heading during a complete wingbeat varied through the turn and was significantly correlated to roll angle at mid-downstroke. Changes in roll angle were found to include both within-wingbeat and among-wingbeat components that bear no direct relationship to one another. Within-wingbeat changes in roll were dominated by the inertial effects while among-wingbeat changes in roll were likely the result of both inertial and aerodynamic effects.