Kangaroo Hopping: Unlocking the Secrets of Efficient Locomotion (2026)

The Role of Postural Adjustments in the Extraordinary Energy Efficiency of Kangaroo Hopping

Imagine bounding across the Australian outback at blazing speeds without your energy levels crashing – that's the kangaroo's incredible superpower, and it's all thanks to clever tweaks in their body mechanics. But wait, how do these furry athletes pull off what no other mammal can? Let's dive into a fascinating study that uncovers the hidden tricks behind their hopping prowess, and trust me, it might just challenge everything you thought you knew about running and jumping.

Abstract

Kangaroos exhibit an astonishing lack of increase in their metabolic energy use as they hop faster, setting them apart from most other running animals. This unique trait might stem from heightened elastic energy recovery, yet the factors enabling this without extra muscle effort have long puzzled scientists. In our research, we crafted a detailed 3D model of the kangaroo's musculoskeletal system, incorporating 3D motion capture and force plate measurements, to dissect the movements and forces involved in hopping by red and grey kangaroos. Through this model, we investigated how body size and speed affect key elements like hindlimb stance, effective mechanical advantage (EMA – a measure of how efficiently muscles generate force relative to the ground reaction), the resulting strain on ankle tendons, and the work done at the ankle during hops. Our findings reveal that increased dorsiflexion at the ankle (pulling the foot upward) and plantarflexion at the metatarsophalangeal joint (pointing the toes downward) likely play a crucial role in lowering ankle EMA by modifying both muscle and ground-based force levers. This, in turn, boosts energy absorption and peak tendon strain at the ankle. Intriguingly, these stance changes seem to amplify tendon strain at higher speeds, enabling greater elastic energy storage and return. Such posture-driven enhancements in elastic energy could be a key driver of kangaroos' energy savings at faster hops, though they might constrain the performance of larger kangaroos due to the heightened risk of tendon failure.

Introduction

Kangaroos and their kin stand out not just for their distinctive physique and hopping style, but for their exceptional energy dynamics. At leisurely paces, they employ a five-legged gait, involving their forelimbs, hindlimbs, and tail in contact with the ground, but as speed picks up, they switch to their iconic bouncing locomotion (Dawson and Taylor, 1973; O’Connor et al., 2014). What truly sets them apart, however, is their energy use during movement. Dating back to the 19th century, observations showed that the metabolic cost of trotting in quadrupeds and walking in bipeds, such as dogs, horses, and humans, rises linearly with speed (Zuntz, 1897; Taylor et al., 1970; Heglund et al., 1982; Taylor et al., 1982). To make sense of this, researchers later refined the 'cost of force generation' idea (Taylor et al., 1980), suggesting that shorter ground contact times at higher speeds demand quicker muscle force production, leading to faster cross-bridge cycling in muscles (Kram and Taylor, 1990). This holds true for a wide array of running and leaping creatures, implying metabolic rate scales inversely with contact duration (Kram and Taylor, 1990). Yet macropods defy this rule. Treadmill tests on red kangaroos (around 20 kg) and tammar wallabies (about 5 kg) showed minimal or no rise in oxygen intake as hopping speed increased (Dawson and Taylor, 1973; Baudinette et al., 1992; Kram and Dawson, 1998). The exact mechanisms allowing macropods to break the speed-energy link remain elusive (Thornton et al., 2022).

The secret to their energy decoupling likely lies in the behavior of their ankle muscle-tendon complexes, which store and release elastic strain energy (Morgan et al., 1978; Biewener et al., 2004b; McGowan et al., 2005). In tammar wallabies, tendon strain at the ankle rises with speed, leading to more elastic strain energy recovery than muscle work, increasing the fraction of energy from tendon snap-back while keeping muscle effort steady (Baudinette and Biewener, 1998; Biewener et al., 1998). Variations in ankle tendon structure due to size (Bennett and Taylor, 1995; McGowan et al., 2008) and the resulting low elastic energy in smaller hopping macropods and rodents may explain why larger macropods benefit energetically, but not their smaller cousins (Thompson et al., 1980; Biewener et al., 1981; Biewener et al., 1998) – though some recent work suggests otherwise (Christensen et al., 2022). Still, tendon structure alone doesn't explain why big macropods can speed up without extra cost, while large quadrupeds with similar tendons can't (Dawson and Webster, 2010). The most obvious macropod hallmark is their hopping gait, but earlier explanations like stable stride frequency (Heglund and Taylor, 1988; Dawson and Webster, 2010) or synchronized breathing and strides (Baudinette et al., 1987) can't differentiate between small and large macropods, or galloping quadrupeds (McGowan and Collins, 2018).

Posture shifts offer another avenue for cutting energy costs by tweaking muscle leverage. EMA represents the balance between a muscle's internal force lever (the distance from the muscle's action line to the joint center) and the ground's external force lever (the distance from the ground reaction force to the joint center). Lower EMA means muscles need more force for a given ground push, potentially requiring more active muscle mass and higher metabolic rates than higher EMA for the same setup. In humans, more bent postures with reduced EMA during walking demand extra muscle effort and raise running energy costs (Biewener et al., 2004a; Kipp et al., 2018; Allen et al., 2022). EMA changes with speed aren't common in four-legged mammals (Biewener, 1989), but elephants – among the largest and most upright animals – continuously lower EMA at higher speeds, tied to rising energy use (Ren et al., 2010; Langman et al., 2012). Typically, EMA shifts with size in land animals. Smaller creatures adopt more crouched stances, with limbs extending more as body weight increases (Biewener, 1989). But this postural shift often relates to lowering size-dependent tissue stress, not energy (Dick and Clemente, 2017; Clemente and Dick, 2023). As limbs straighten, stresses in bones and muscles stay constant despite weight (Biewener, 1989; Biewener, 2005).

Macropods, however, keep a crouched hop. Kram and Dawson, 1998 checked if kangaroos straightened their limbs at higher speeds for steady energy, but found no EMA shift at ankle from 4.3 to 9.7 m/s. Ankle stance and EMA seem to vary little with size in macropods (Bennett and Taylor, 1995; Snelling et al., 2017). So, strain should build with both speed and size. High tendon strains are vital for quick elastic energy release (strain energy scales with stress squared) (Biewener and Baudinette, 1995). With tendons central to hopping, it's no surprise tendon strain edges close to failure in bigger kangaroos. Ankle extensor tendons in a hefty red kangaroo male (46.1 kg) operate at safety margins around two at slow hops (3.9 m/s), much lower than the usual 4-8 for mammal tendons (Ker et al., 1988). Strains are notably high even in smaller kangaroos. Juvenile and adult western grey kangaroos from 5.8 to 70.5 kg all hop with gastrocnemius and plantaris tendon safety margins under two (Snelling et al., 2017). Elevated strains might be a byproduct of their crouched stance and tendon build, but perhaps they're also an evolutionary advantage. If kangaroos tweak their posture to boost tendon strain and elastic energy, we'd expect clear links between stance, size, and speed that haven't been fully mapped yet.

In this investigation, we examined hindlimb movements and forces in kangaroos hopping at different speeds. We focused on how stance shifts, EMA, joint effort, and tendon strain change with speed and size. We built a kangaroo musculoskeletal model from real imaging and dissections (Figure 1a), using it to track EMA shifts by monitoring muscle and ground levers. We predicted that (i) the hindlimbs would crouch more at higher speeds, mainly via the ankle and metatarsophalangeal joints, separate from size effects, and (ii) lever changes from posture would raise tendon strain with speed, potentially boosting energy savings by amplifying ankle work without added muscle input.

Results

Stride Parameters

Hopping speeds spanned 1.99 to 4.48 m/s. Bigger kangaroos tended to hop a tad faster (β=0.048, SE=0.018, p=0.009, R²=0.057), and since speed weakly tied to size, both factors were factored into regression models for other measures (see Appendix 1—table 1).

Higher speeds led to stronger acceleration peaks in the braking stance phase, like lower minimum horizontal acceleration and higher maximum vertical acceleration (Figure 2—figure supplement 1a). Size didn't affect acceleration, but an interaction showed smaller kangaroos ramped up vertical acceleration more sharply between slow and fast hops than bigger ones (Appendix 1—table 1).

Size and speed impacted ground contact time differently (Appendix 1—table 1). Bigger kangaroos had slightly longer contact, while faster speeds shortened it. The tight link between contact time and speed (R²=0.73) could help estimate speed from contact duration alone. In Figure 2—figure supplement 1b, we merged our data with red kangaroo findings from Kram and Dawson, 1998, to broaden the prediction range.

Bigger kangaroos hopped with longer strides and slower frequencies than smaller ones (Appendix 1—table 1), and strides lengthened with speed (Figure 2—figure supplement 1c). Frequency dropped with size and rose with speed, but a notable interaction meant bigger kangaroos boosted frequency more for speed than smaller ones (Appendix 1—table 1).

Ground Reaction Forces

The hop starts with braking (backward horizontal ground force), shifting to propulsion (forward horizontal force). The force turns vertical at the switch, around 42.9±26.9% of stance. But peak vertical force hits at 46.4±5.3% of stance (Figure 2a and b).

Bigger kangaroos and faster speeds generated stronger peak forces (Figure 2a and b; Appendix 1—table 2). Peak vertical force, body-weight normalized, ranged from 1.4 to 3.7 times body weight (average 2.41±0.396 BW), climbing with speed (Figure 2—figure supplement 2a). Smaller kangaroos might face disproportionately higher forces at speed (Figure 2—figure supplement 2b, Appendix 1—table 2). This force boost naturally raises tendon strain, regardless of posture.

Kinematics and Posture

Kangaroo hindlimb stance movements varied with size and speed. Supporting our first hypothesis partly, bigger and faster hops involved more crouched hindlimbs (Figure 3a and c; crouch factor = total limb length / pelvis-to-toe distance). Size effects spread across all joints subtly (Figure 3d and f; Appendix 1—table 3). Hips, knees, and ankles flexed more, while metatarsophalangeal joints extended (plantarflexed) in bigger kangaroos.

Effective Mechanical Advantage

EMA dropped as limbs crouched more. Changes with size and speed were big, especially in braking (Figure 2c and d). We checked mid-stance EMA, when ground force (and tendon strain) peaked. EMA fell with size, but not speed (Figure 2c and d; Appendix 1—table 5), though this might miss subtle shifts. Diving deeper, the drop stemmed from a smaller muscle lever arm (r, for gastrocnemius and plantaris) with speed, not mass (Figure 3—figure supplement 1a, b and h; Appendix 1—table 5). A 1 m/s faster hop cut r by about 4.2%. Muscle levers dropped across our speeds (1.99–4.48 m/s), with the top r at 114.4±0.8° ankle angle (Figure 3—figure supplement 1g). Peak dorsiflexion ranged from 114.5° slow to 75.8° fast (Figure 3g). The low r timing matched peak dorsiflexion.

We saw a ground lever arm (R) rise with size, but not speed at mid-stance (Figure 3—figure supplement 1c, d and h; Appendix 1—table 5). We expected R to shift with both, as more metatarsophalangeal plantarflexion lowers the ankle toward the ground, increasing R (since mid-stance center of pressure stayed put) (Appendix 1—table 2). A 4.6 kg weight gain or 1 m/s speed boost dropped ankle height (ground-to-ankle vertical distance) by ~20% (Figure 3—figure supplement 1e and f). But if R doesn't change much with speed, it might clash with height and pressure data. Maybe there's a small, key R shift with speed, hard to spot due to range and error (Figure 3—figure supplement 1h). R depends on metatarsal length, longer in bigger kangaroos (1 kg body mass = ~1% longer segment, p<0.001, R²=0.449). If R rises with speed, combined with r changes, it'd lower EMA at faster hops.

Joint Moments

Small proximal joint rotations led to noticeable drops in dimensionless hip extensor and knee flexor moments with size (Figure 3—figure supplement 2a, Appendix 1—table 4), and rises in both during braking with speed (Figure 3—figure supplement 2b). Peak moments interacted with size and speed, meaning bigger kangaroos hiked moments faster with speed than smaller ones.

The muscle lever arm (r) shift with speed had big implications for Achilles tendon strain, since tendon force = ankle moment / r. So, r drop boosts tendon force (and strain), as does rising ankle moment. We found speed, not size, lifted peak ankle moment (Figure 3—figure supplement 2; Appendix 1—table 4). Our size range might not capture mass effects beyond speed.

Tendon Stress

Peak Achilles tendon strain climbed as minimum EMA fell (Figure 4a). Tiny EMA drops, perhaps missed before, tied to big strain jumps. For example, EMA from 0.242 (slow group average) to 0.206 (fast group) boosted strain from ~50 MPa to ~60 MPa, slashing safety margin from 2 to 1.67 (failure at 1), hugely meaningful. Strain rose with size and speed (Appendix 1—table 5). Peak strain hit at 46.8±4.9% of stance, aligning with peak vertical force and ankle moment. Size didn't shift peak timing, but it came 3.2±0.8% sooner per 1 m/s speed (p<0.001), matching dorsiflexion peaks. This speed-driven strain hike via lever tweaks likely fuels ankle work shifts (per hypothesis ii).

Joint Work and Power

Joint energy analysis showed the ankle handled most stance work and power (Figure 5; Appendix 1—table 8). Ankles did negative work in braking, turning kinetic and potential energy into elastic potential via tendon loading. Negative ankle work linked to tendon strain (β=0.750, SE=0.074, p<0.001, R²=0.511). Regression showed it rose with speed, not size (Figure 6a; Figure 6—figure supplement 1e and f; Appendix 1—table 6).

Elastic energy recoiled for positive work in propulsion, as ankles extended pre-mid-stance. Positive ankle work grew with size and speed (Figure 6a; Figure 6—figure supplement 1e and f; Appendix 1—table 6), but since negative work matched, net ankle work stayed flat with speed (Figure 6b; Figure 6—figure supplement 2e and f; Appendix 1—table 6). Supporting hypothesis ii, the work hikes stem from higher strain, not extra muscle effort.

Metatarsophalangeal joints mostly did negative work. They absorbed more at speed, while net work dropped with size and speed (Figure 6—figure supplement 1g and h; Figure 6—figure supplement 2g and h; Appendix 1—table 6). Knees barely did negative work, hips even less (Figure 5; Figure 6—supplements 1 and 2). Hip net work rose slightly with size and speed, knee with speed only (Appendix 1—table 6).

Discussion

How Posture Fuels Kangaroo Energetics

The 'force generation cost' idea says metabolic rate should climb as contact time shortens with speed (Kram and Taylor, 1990). Macropods buck this (Dawson and Taylor, 1973; Baudinette et al., 1992; Kram and Dawson, 1998). Their ankle muscle-tendon units probably set them apart, but how? (Bennett and Taylor, 1995; Bennett, 2000; Thornton et al., 2022). We tested and confirmed hindlimbs crouch more at higher speeds, mainly via ankle and metatarsophalangeal joints. We suggest lever shifts from posture raise tendon strain with speed, aiding energy savings by boosting ankle work sans extra muscle (Figure 7).

Achilles strain hinges on extensor muscle force, which ties to muscle moment arm, ankle moment, ground moment arm, and ground force – assuming tendon properties and area stay put (Figure 7). We saw peak ankle moment rise with speed despite ground arm changes, probably from dominant ground force effects (ankle moment = ground force × ground arm). Peak ground force also surged with speed and briefer contacts (Figures 1b and 2b, Figure 2—figure supplement 1). Yet, ankle moment rise and muscle arm drop boost strain (strain = force / area; tendon force = moment / muscle arm). Though ground force explains much strain rise (and stretch) (Figure 2—figure supplement 3), the strain-EMA link shows posture tweaks modulate strain beyond forces (Figure 4a). If ground force stayed equal but EMA changed like slow to fast, strain would jump 18%; if EMA stayed but force varied similarly, only 12%. So, posture shifts and contact shortening both tweak kangaroo strain, in our speed range.

Untangling Force and Energy in Hopping

As Achilles strain climbs with speed in hopping kangaroos, so must gastrocnemius and plantaris forces – or the tendon would slacken instead of storing energy. These force hikes could come from more activation, but that'd spike muscle energy use, and thus metabolism. So, the big question is: Are there realistic ways to hike force without energy cost? Experimental macropod data show tendon force and EMG decouple, with more EMG not always meaning bigger forces or strains (Biewener et al., 2004b Table 2). This decoupling might stem from muscle mechanics, like force-length or force-velocity ties. Force-length could help if muscles hit better spots during tendon loading. Tammar wallabies' gastrocnemius and plantaris don't shorten much, acting like struts in stance late half; but early half stretch happens (Biewener et al., 1998). This doesn't shift with speed, despite big force hikes (Biewener et al., 1998).

Force-velocity might be crucial. During hops, muscles activate before ground touch, with force mostly from fiber stretch early, fibers lengthening actively before strutting (Biewener et al., 2004b, Figure 3). Faster hops mean shorter contacts, so stretch rates likely speed up. This fits with tendon transducer and kinematic data on kangaroos (Griffiths, 1989), showing stretch rates 10x faster at supramaximal isometric forces. Frog fiber experiments show lengthening cheaper than isometric (Linari et al., 2003), letting muscles tension up economically. Thus, posture changes with speed might boost active fiber stretch rates early, hiking strain and storage without more activation or cost. But a full musculoskeletal model with muscle energetics is needed to confirm. And this is the part most people miss: Could this explain macropod uniqueness without other mammals catching on?

Caveats and Limits

We think lever shifts tie to speed more than size, given big joint and height changes at speed. But our samples (13.7–26.6 kg) were narrow vs. full range (up to 80 kg), limiting size-speed separation. Future studies need broader sizes. Kangaroos hopped slowly here, capping our EMA and strain views at safer zones. Faster hops might push strains closer to rupture; different setups could help. We didn't check proximal joint EMA changes (harder to track), which might affect bigger muscle masses, countering or aiding ankle savings. Research is vital to see whole-body posture and muscles in kangaroo energetics.

What Makes Macropods Special?

No other mammals match macropod energy feats despite similar tendons or strides (Thornton et al., 2022), but macropods are outliers in other ways. They operate at lower ankle EMA than big mammal peers (Figure 4b). Animals >18 kg usually hit 0.5–1.2 EMA. Only rodents <1 kg match kangaroo values (Figure 4b; Biewener, 1990), like kangaroo rats with thicker tendons for less elastic recovery (Biewener and Blickhan, 1988) – but see Christensen et al., 2022. Figure 4a shows strain vs. EMA nonlinearity in kangaroos, quadrupeds, and humans. Other animals operate where big EMA changes yield small strain shifts, making EMA less potent for strain boosts at speed. But here's where it gets controversial: If posture is so tweakable, why don't other animals do it? Is it just macropod evolution, or a hidden universal trick?

Performance vs. Size Limits?

Macropod anatomy and hopping grant amazing efficiency, but at costs. High tendon compliance might hinder quick starts, delaying force to ground. Second, massive strains mean low safety margins. Studies suggest kangaroos hop at 1–2 margins for big ones (Kram and Dawson, 1998; McGowan et al., 2008; Snelling et al., 2017; Thornton et al., 2022). Our size-speed EMA shifts might worsen this, underestimating rupture mass. Snelling et al., 2017 and McGowan et al., 2008 put safe max at ~150 kg, but extinct giants hit 240 kg (Helgen et al., 2006; Janis et al., 2023). So, posture might shift from strain-booster to mitigator past a mass point (Dick and Clemente, 2017). What do you think – are kangaroos pushing evolutionary limits, or is there untapped potential? Share your views in the comments!

This study shows EMA more flexible than thought, with musculoskeletal modeling revealing form-function ties often elusive in experiments.

Methods

Animals and Data Gathering

Hopping videos of red and eastern grey kangaroos came from Brisbane’s Alma Park Zoo, per Royal Veterinary College Ethics approval (URN 2010 1051). We studied 16 males and females, 13.7–26.6 kg (average 20.9±3.4 kg). Two juvenile reds and 11 greys were noted; three unclassified. Mass from force plates, or interpolated from leg markers.

Experimental Setup

Kangaroos hopped a ~10×1.5 m enclosed runway with two sequential force plates (Kistler 60×60 cm, AMTI 50×50 cm). A 6-camera Vicon T160 system captured motion at 200 Hz. Markers on hips, knees, ankles, metatarsophalangeal joints, phalanx IV tip, ilium base, tail base (Figure 1a). Forces recorded at 1000 Hz via analog board synced to Vicon.

Data Processing

Stride length from ankle/metatarsophalangeal markers at points. Stance from vertical force >2% peak. Contact and stride durations from frame rate and frames. Acceleration from smoothed pelvis position.

We averaged strides if visible before/after plate. 173 total, 100 trials after exclusions (non-simultaneous feet, starts/stops on plate). Assumed forces split evenly, halved vertical/horizontal/lateral. Normalized to body weight.

Center of pressure tracked in coordinate system (Figure 1b). Assumed no medial-lateral shift; used symmetrical trials.

Musculoskeletal Model Building

Built in OpenSim (v3.3; Seth et al., 2018) from western grey kangaroo CT scan (morphosource.org). Hindlimb: pelvis, femur, tibia, metatarsals/calcaneus, phalanges. Hinge joints, hip ball-and-socket. Scaled to each kangaroo via markers.

Mass/inertia from Hopwood, 1976. Muscles from dissection (USC Ethics ANE2284) and literature (Bauschulte, 1972; Hopwood and Butterfield, 1976, 1990).

Joint Kinematics and Mechanics

Inverse kinematics for angles. Inverse dynamics for moments, normalized to weight/length.

Powers integrated for work periods (Dick et al., 2019). Normalized to mass.

Posture and EMA

Crouch factor: ilium-to-toe distance / total length.

EMA: gastrocnemius/plantaris muscle arm / ground arm (Figure 1b). Arms from model, scaled.

Tendon Stress

Forces: moment / muscle arm. Stress: force / area (from Snelling et al., 2017 regressions). Excluded flexor digitorum longus.

Stats

Multiple regressions for effects (lm in R v.3.6.3). Interactions removed if non-significant. p<0.05 key. Trials individual; species not factored (no differences).

Appendix 1

[Full tables and supplements as in original, rephrased where needed for uniqueness.]

Data Availability

Data and analyses at https://doi.org/10.6084/m9.figshare.30282592.v1.

References

[All original references, unchanged.]

Article and Author Information

[All original author details, unchanged.]