Explaining the energetics of walking and running has been difficult because the distribution of energy use among individual muscles has not been known. We estimated energy use by measuring blood flow to the hindlimb muscles in guinea fowl. Blood flow to skeletal muscles is controlled locally and varies directly with metabolic rate. We estimate that the swing-phase muscles consume 26% of the energy used by the limbs and the stance-phase muscles consume the remaining 74%, independent of speed. Thus, contrary to some previous suggestions, swinging the limbs requires an appreciable fraction of the energy used during terrestrial legged locomotion. Models integrating the energetics and mechanics of running will benefit from more detailed information on the distribution of energy use by the muscles.

, a value approximately two thirds of the maximal expected value for skeletal muscle. The muscle-mass-specific peak power output during jumping was nearly 800·W·kg -1 , which is more than twice the peak isotonic power estimated for guinea fowl leg muscles. To account for high power outputs, we concluded that energy has to be stored early in the jumps and released later during peak power production, presumably using mechanisms similar to those found in more specialized jumpers.

Running and walking are mechanically complex activities. Leg muscles must exert forces to support weight and provide stability, do work to accelerate the limbs and body centre of mass, and absorb work to act as brakes. Current understanding of energy use during legged locomotion has been limited by the lack of measurements of energy use by individual muscles. Our study is based on the correlation between blood flow and aerobic energy expenditure in active skeletal muscle during locomotion. This correlation is strongly supported by the available evidence concerning control of blood flow to active muscle, and the relationship between blood flow and the rate of muscle oxygen consumption. We used injectable microspheres to measure the blood flow to the hind-limb muscles, and other body tissues, in guinea fowl (Numida meleagris) at rest, and across a range of walking and running speeds. Combined with data concerning the various mechanical functions of the leg muscles, this approach has enabled the first direct estimates of the energetic costs of some of these functions. Cardiac output increased from 350 ml min −1 at rest, to 1700 ml min −1 at a running speed (∼2.6 m s −1 ) eliciting aV O 2 of 90% ofV O 2 ,max . The increase in cardiac output was achieved via approximately equal factorial increases in heart rate and stroke volume. Approximately 90% of the increased cardiac output was directed to the active muscles of the hind limbs, without redistribution of blood flow from the viscera. Values of mass-specific blood flow to the ventricles, ∼15 ml min −1 g −1 , and one of the hind-limb muscles, ∼9 ml min −1 g −1 , were the highest yet recorded for blood flow to active muscle. The patterns of increasing blood flow with increasing speed varied greatly among different muscles. The increases in flow correlated with the likely fibre type distribution of the muscles. Muscles expected to have many high-oxidative fibres preferentially increased flow at low exercise intensities. We estimated substantial energetic costs associated with swinging the limbs, co-contraction to stabilize the knee and work production by the hind-limb muscles. Our data provide a basis for evaluating hypotheses relating the mechanics and energetics of legged locomotion.

SUMMARY We examined the energetic cost of loading the trunk or distal portion of the leg in walking and running guinea fowl (Numida meleagris). These different loading regimes were designed to separately influence the energy use by muscles used during the stance and swing phases of the stride. Metabolic rate, estimated from oxygen consumption, was measured while birds locomoted on a motorized treadmill at speeds from 0.5 to 2.0 m s-1, either unloaded, or with a mass equivalent to 23% of their body mass carried on their backs, or with masses equal to approximately 2.5% of their body mass attached to each tarsometatarsal segment. In separate experiments, we also measured the duration of stance and swing in unloaded, trunk-loaded, or limb-loaded birds. In the unloaded and limb-loaded birds, we also calculated the mechanical energy of the tarsometatarsal segment throughout the stride. Trunk and limb loads caused similar increases in metabolic rate. During trunk loading, the net metabolic rate (gross metabolic rate - resting metabolic rate) increased by 17% above the unloaded value across all speeds. This percentage increase is less than has been found in most studies of humans and other mammals. The economical load carriage of guinea fowl is consistent with predictions based on the relative cost of the stance and swing phases of the stride in this species. However, the available comparative data and considerations of the factors that determine the cost of carrying extra mass lead us to the conclusion that the cost of load carrying is unlikely to be a reliable indicator of the distribution of energy use in stance and swing. Both loading regimes caused small changes in the swing and/or stance durations, but these changes were less than 10%. Loading the tarsometatarsal segment increased its segmental energy by 4.1 times and the segmental mechanical power averaged over the stride by 3.8 times. The increases in metabolism associated with limb loading appear to be linked to the increases in mechanical power. The delta efficiency (change in mechanical power divided by the change in metabolic power) of producing this power increased from 11% in walking to approximately 25% in running. Although tarsometatarsal loading was designed to increase the mechanical energy during swing phase, 40% of the increase in segmental energy occurred during late stance. Thus, the increased energy demand of distal limb loading in guinea fowl is predicted to cause increases in energy use by both stance- and swing-phase muscles.

SUMMARY Uphill running requires more energy than level running at the same speed,largely due to the additional mechanical work of elevating the body weight. We explored the distribution of energy use among the leg muscles of guinea fowl running on the level and uphill using both organismal energy expenditure(oxygen consumption) and muscle blood flow measurements. We tested each bird under four conditions: (1) rest, (2) a moderate-speed level run at 1.5 m s–1, (3) an incline run at 1.5 m s–1 with a 15% gradient and (4) a fast level run at a speed eliciting the same metabolic rate as did running at a 15% gradient at 1.5 m s–1(2.28–2.39 m s–1). The organismal energy expenditure increased by 30% between the moderate-speed level run and both the fast level run and the incline run, and was matched by a proportional increase in total blood flow to the leg muscles. We found that blood flow increased significantly to nearly all the leg muscles between the moderate-speed level run and the incline run. However, the increase in flow was distributed unevenly across the leg muscles, with just three muscles being responsible for over 50% of the total increase in blood flow during uphill running. Three muscles showed significant increases in blood flow with increased incline but not with an increase in speed. Increasing the volume of active muscle may explain why in a previous study a higher maximal rate of oxygen consumption was measured during uphill running. The majority of the increase in energy expenditure between level and incline running was used in stance-phase muscles. Proximal stance-phase extensor muscles with parallel fibers and short tendons, which have been considered particularly well suited for doing positive work on the center of mass, increased their mass-specific energy use during uphill running significantly more than pinnate stance-phase muscles. This finding provides some evidence for a division of labor among muscles used for mechanical work production based on their muscle–tendon architecture. Nevertheless, 33% of the total increase in energy use (40% of the increase in stance-phase energy use) during uphill running was provided by pinnate stance-phase muscles. Swing-phase muscles also increase their energy expenditure during uphill running, although to a lesser extent than that required by running faster on the level. These results suggest that neither muscle–tendon nor musculoskeletal architecture appear to greatly restrict the ability of muscles to do work during locomotor tasks such as uphill running, and that the added energy cost of running uphill is not solely due to lifting the body center of mass.