Veterinary Record 171:87-94 doi:10.1136/vr.e4966
  • Feature
  • Comparative Physiology

Animal athletes: a performance review

  1. N. C. Craig Sharp
  1. Centre for Sports BVMS, PhD, DSc, FIBiol, FBASES, FPEAUK, MRCVS, Medicine and Human Performance, Brunel University, London UB8 3PH, UK
  1. e-mail: c.sharp90{at}

Major sporting events such as the Olympic Games highlight extraordinary human athletic achievements, performed by the extreme physical outliers of our species. However, there is a range of animal performance that goes far beyond these, both in the wild and in selectively bred ‘sports animals’ such as the racehorse, racing camel, greyhound, sled-dog and racing pigeon. In this selective review, Craig Sharp assesses how human beings measure up in the animal athletic stakes.

WHEN looking at the subject of athleticism, it is worth considering some remarkable feats of animal runners (or ‘innately athletic locomotor extremists’ as Alan Wilson of the Royal Veterinary College calls them), as well as of those animals that fly or dive. It is also interesting to consider some specific aspects of comparative exercise physiology.


Sharp (1997) timed the cheetah (Acinonyx jubatus) to reach 64 mph (104 kph, 29 m/sec). The pronghorn antelope (Antilocapra americana) reaches 55 mph (89 kph, 24.6 m/sec) and the North African ostrich (Struthio camelus camelus) is the fastest running bird at nearly 40 mph (64 kph, 18 m/sec), while sailfish (Istiophorus species) reach a swimming speed of 67 mph (108 kph, 30 m/sec). Garland (1983) quoted a maximum speed of 20 m/sec or more (72 kph, 45 mph) for 17 species of mammal.

Other approximate maximum speeds are:

  • Human beings 23.4 mph (37.6 kph, 10.4 m/sec);

  • Thoroughbred racehorse 43.4 mph (70 kph, 19.3 m/sec);

  • Greyhound 43 mph (69 kph, 19.2 m/sec); and

  • Camel (Camelus dromedarius) 22 mph (35.3 kph, 9.8 m/sec).

Cheetah ‘Sarah’ at Cincinnati Zoo. Note the ‘running-spike’ claws, and that the forefeet and hindfeet will each hit the ground virtually simultaneously, ie, the cheetah is effectively bipedal in its locomotion

Data from the sporting pages of newspapers show that most greyhound races are won at 15 to 16 m/sec, and most horse races of less than one mile (1.6 km) at 16 to 17 m/sec. The fastest horse speed recorded is 55 mph (88 kph, 24.4 m/sec) by the racing quarter horse (Young 2003). Young also noted that the merriam kangaroo rat (Dipodomys merriami), running at 110 body lengths/sec, would relatively outstrip the cheetah's 32 lengths/sec.

Cheetahs have a long streamlined head and body, 112 to 135 cm long. At top speed, a cheetah's extremely flexible spine assists a stride length of 7 to 8 m, which is so long that all four feet are off the ground for more than half the distance (Rich and Rouse 2004). Other speed design features include: tough ridged pads for grip; non-retractable long blunt claws like running shoe spikes, which maximise traction; long legs, with most of the muscle proximally situated near the body (like antelopes, gazelles, greyhounds, horses and camels), giving very long tendons, which can store and release over half the energy at every step. The forepaws strike the ground nearly together, as do the hindpaws – almost as if each pair was a ‘superleg’ effectively giving a longer leg. The hindpaws reach just anterior to the forepaws, crossing mid-air (not on the ground), that is, the forepaws lift off just before the hindpaws come past them in the air – but the hindpaws do not strike the ground in front of the forepaws, as these have already just lifted off. This can only occur due to the cheetah's very flexible spine, which also assists the hindlegs in supporting most (70 per cent) of its bodyweight like a rear-wheel drive car, especially at speed (Pfau and others 2009, Hudson and others 2012). The contribution to a cheetah's speed from its spinal column and associated musculature is approximately 10 per cent, that is, a legless cheetah could ‘run’ at over 6 mph.

The horse shows little change in fore:hind (57 per cent:43 per cent) weight distribution with increasing speed or gaits. Regarding muscle fibre profiles, human sprinters, sports horses, greyhounds and camels all range through approximately 70 per cent to 80 per cent of ‘fast’ (type II) fibres (with, very roughly, 50 per cent each of IIA and IIB). The quarter horse has 93 per cent ‘fast’ – the heavy hunter 68 per cent – but in quarter horses, 45 per cent of these are type IIB glycolytic. The cheetah has approximately 84 per cent type II fibres but, remarkably, as many as 85 per cent of these are the largely anaerobic glycolytic IIB, a value matched only by their prey species.

Middle-distance running

The pronghorn antelope is reported as running 10 km in 11 minutes (Nowak 1992). Their top speed is alleged to be of the order of 58 mph (93 kph), and herds of pronghorn galloping across the high prairies of Wyoming have been recorded as maintaining 40 mph (64 kph) for 30 minutes or more. It would seem that they still bear a genetic memory of the extinct giant American cheetah, or former plains wolf packs. Lindstedt, comparative physiologist at Northern Arizona University, ran two pronghorns progressively on a treadmill with a gradient up to 11 per cent at 22 mph, noting that: ‘Each consumed between six and 10 litres of oxygen per minute, which is five times as much as a typical mammal of similar size’ (Lindstedt and others 1991).

The horse can run half a mile in 46 sec, a mile in 1 min:35 sec, and maintain 30 mph (48 kph) for 10 minutes – and with 59 kg on its back, although the jockey can aid the locomotion. Pfau and others (2009) showed that jockeys, in the crouched position off the saddle, move to isolate themselves from the movement of the horse, whereby the horse supports the jockey's bodyweight but does not have to move them through each cyclical stride path, gaining some 5 per cent in speed, compared to a seated rider. The crouched posture requires high energy expenditure by jockeys, who have near-maximum heart rates and substantial rises in blood lactate during racing. As a rule of thumb, each 0.9 to 1.3 kg extra slows a horse by about a length over eight to 10 furlongs (Hillenbrand 2001).

Distance running/walking

Camels can maintain 10 mph (16 kph) for over 18 hours, while sled-dogs, such as the Siberian husky, race up to 1200 miles, most famously in the 1000-mile (1600 km) Alaskan ‘Iditarod’, in teams of 12 to 16, over mountain, tundra, forest and frozen rivers. The record, set in 2011, is 8 days:19 hr:47 min (114 miles/day). Dog sled-racing was a demonstration Olympic sport at the VI Winter Games of Oslo (1952) and the XIII Games at Lake Placid (1980). Caribou (reindeer) (Rangifer tarandus) have the longest land animal migration, of up to 3000 miles (4830 km) annually, while wildebeest (Connochaetes taurinus) travel a circle of 1800 miles (3000 km) annually around the Serengeti (collectively giving birth to 400,000 calves). Even cheetah may cover 30 miles (50 km) in one night. African wild dogs (Lycaon pictus) show high levels of endurance, while three examples of human endurance are: Greek runner Yiannis Kouros who ran 635.5 miles (1017 km) in the New York six-day run at 105.9 miles (169.5 km) per day; British runner Bruce Tulloh who ran 2876 miles (4600 km) from Los Angeles to New York in 64 days with an estimated four million footsteps at 44.9 miles (71.9 km) per day; and Istvan Sepos of Hungary, who ran 12,554 miles (20,100 km) in 263.2 days at 47.7 miles (76.3 km) per day. Noakes (2007) has calculated that Scott's Antarctic team man-hauled 2500 km to the South Pole and back with a calculated energy expenditure over 159 days of about 1 million kcal each, probably the greatest recorded sustained endurance human ‘athletic’ performance. Roald Amundsen's sled-dogs expended about 500,000 kcal each over 97 days, probably the greatest recorded terrestrial animal endurance feat.

World record track and field comparisons

A greyhound has run 100 m in 5.8 sec, compared to Usain Bolt's record of 9.58 sec.

In the 200 m the cheetah (Acinonyx jubatus) has been timed at 6.9 sec, the horse (Black Caviar) at 9.98 sec and the greyhound at 11.2 sec, compared to Bolt's record of 19.19 sec.

The racing quarter horse can run 400 m in 19.2 sec and the greyhound in 21.4 sec compared to Michael Johnson's 43.18 sec.

Over 800 m, the pronghorn antelope (Antilocapra americana) with a time of 33 sec and the greyhound with 49.2 sec, beat David Rudisha's 1 min:41.01 sec.

At one mile the pronghorn's 1 min:30 sec compares to Hicham El Guerrouj at 3 min:43.13 sec, while at 5000 m, the pronghorn's 5 min:20 sec compares to Kenenisa Bekele's 12 min:37.35 sec, and over 10,000 m its 11 min compares to his 26 min:17.53 sec.

In the marathon, an endurance horse has run 1 hr:18 min:29 sec, compared to the 2 hr:3 min:38 sec record of Patrick Makau Musyoki.

In the long jump, the red kangaroo (Macropus rufus) has leapt 12.8 m, compared to Mike Powell's 8.95 m, and its high jump of 3.1 m is well ahead of Javier Sotomayor's 2.45 m, who is also overtaken by the springbok (Antidorcas marsupialis), reputed to jump over 3 m, and by the snakehead fish (Channidae species), said to leap 4 m out of the water. Equine high and long jump records are 2.47 m by Huaso, and 8.40 m by SomethingGraphic.

As indicated in the box above, the horse marathon record is some 45 minutes faster than the human equivalent, but is very dependent on ideal conditions, including a flat course and an equable temperature. Early stagings of the Man v Horse 22-mile race in Llanwrtyd Wells saw easy victories for the horse until the original flat course was re-routed through hilly farm tracks, forestry roads and rough moorland on the edge of the Brecon Beacons, that is, the playing field was unlevelled to make it a more even competition, as such terrain affects the horse more than the man. This caused the margins by which the horse won to drop from 30 minutes to under a minute, and in 2004 Huw Lobb, an international marathon runner, was the first man to win the race, in 2 hr:5 min:19 sec to beat his closest equine rival by 2 min:17 sec. He commented: ‘As an experienced fell-runner I was clambering up steep banks, jumping off ledges and throwing myself down steep hills in a way no horse could do’. Another factor tipping the balance toward people in this 22-mile race is that the horses are required to have a mandatory 10- to 15-minute ‘veterinary break’.

Over 100 years ago, Kennelly (1906) noted that records were already being compiled for equine trotting (harness racing) races of 100 miles, for example, 8 hr:55 min in 1853. In Australia there is a current annual 100-mile race involving one horse and two runners who ride alternately, for which the record stood at 7 hr:3 min:22 sec in 1995. In the popular single-rider 100-mile race (with five veterinary stops), the record is 6 hr:41 min:33 sec, that is, just over four-minute mile pace. In another such race the last 12 miles were sprinted in 35 minutes, that is, under three-minute mile pace. The horses race at a mean heart rate of approximately 140 bpm, at about 50 per cent VO2max. In a 50 mile (80 km) race across the desert in the United Arab Emirates, Tom Johnson, the USA record holder at 100 km, beat the purebred Arab horse Al Barraq. However, Al Barraq, ridden by Jennifer Nice, had two mandatory 40-minute breaks for food and water. Johnson won by about 50 yards in 5 hr:45 min. It was hoped that endurance horse-racing might have been a demonstration sport at the 2012 London Olympic Games.

The human ability for prolonged distance running is more than an athletic curiosity. Daniel Lieberman, Chair and Professor of the Department of Human Evolutionary Biology at Harvard, has focused on how the human body has evolved, especially in terms of an early Darwinian tendency to improve at distance running (Bramble and Lieberman 2004). He believes that the purposeful adaptations of bipedalism and endurance running have played a major part in modifying evolving human anatomy, as summarised below:

  • The marked ligamentum nuchae, which shock absorbs for the head during running;

  • Our excellent sense of balance, which keeps the head stable as we run;

  • The mass of sweat glands for thermoregulation (only equalled by the Equidae);

  • The lack of body fur for the same reason;

  • Shoulders that move independently from the neck (unlike the apes) so that the arms can swing for balance while the head remains stable;

  • Proportionately long legs and narrow waist;

  • Proportionately large joint surface areas in hips, knees and ankles for load-spreading to share and minimise impact shock;

  • An arched foot to help store and return energy (as does the Achilles tendon);

  • Short toes. Increasing toe length by just 20 per cent doubles the mechanical work of the foot. Also the human big toe is parallel to the others, not being divergent as in apes, and it provides the final push-off in running (all this suggests that the foot itself was evolved for running);

  • The very large human gluteus maximus (compared to the great apes) is primarily a running muscle to keep the trunk upright when running (it is barely used when walking);

  • Human capacity to store about 20 miles worth of glycogen in the muscles of locomotion.

These features enabled primitive human beings to outrun animal prey over many hours, allowing them to incorporate meat and fat into their diet, providing required higher calorie intakes for the evolving human brain, more than could be supplied by the gathering of plant food.

This distance running capability still holds good, as human endurance running skills can demonstrate.

Remarkable feats of athleticism. (above) Haile Gabresellasie and Paul Tergat in a sprint finish to the Olympic 10,000 metres in Sydney. (left) Patrick Schoberg setting a world high jump record

Picture: Paul Webster


Many, indeed most, measures of animal physical prowess are loose approximations at best, but none more so than the difficult measure of strength, hence the brevity of discussion here. For example, how would one measure the ‘strength’ of the blue whale (Balaenoptera musculus), the largest ever animal, whose heaviest known example weighed over 193,455 kg, while the longest example measured over 32.9 m? The African elephant (Loxodonta africana) is said to be able to lift 300 kg with its trunk, and to easily carry 820 kg of logs; the grizzly bear (Ursus arctos horribilis) is said to lift 455 kg and the gorilla (Gorilla species) to lift 900 kg. The (accurate!) human world record for the ‘clean and jerk’ is 283 kg, and the world power-lifting record for the squat is 457.5 kg.

Specific animal attributes


Birds display remarkable athletic abilities.

Speed fliers

Peregrine falcons (Falco peregrinus) are variously said to dive at 115 mph (185 kph, 51 m/sec) by radar tracking (Kestenholtz 1998), 161 mph (259 kph, 72 m/sec) by optical tracking (Tucker 1998) or 204 mph (325 kph, 90.3 m/sec) via accelerometers attached dorsally. Either way, the peregrine is the fastest animal on Earth (E. W. Wheeldon, personal communication). Several duck and geese species attain cheetah speeds of 64 mph (103 kph, 29 m/sec) in level flight.

Some avian species are outstanding. In terms of muscle power, Askew has noted that: ‘Pheasant and grouse families generate 400W power per kg, which is five times as powerful as trained humans. Hummingbirds are also powerful, at 200W/kg’. Askew and others (2001) filmed and plotted the trajectory and velocity of take-off of birds of the pheasant family, bred in captivity, namely, blue-breasted quail (Coturnix chinensis) (weight 40 g), grey- and red-necked pheasant (Phasianus species) and peafowl (Pavo species) (4.5 kg), which all generated similar bursts of power. ‘We were very surprised as nothing has been measured generating power like this. While well-designed for short bursts of intense activity and swift take-off, the muscle of game birds is poorly adapted for longer flights.’

Nevertheless, the equine biceps has been estimated to release 243 J in 0.11 sec in a gallop (Wilson and others 2003), which equates to a peak power of over 4000W. Yet its conventional power output would be under 200W. These massive peaks, necessary for such fast limb protraction in the gallop, are due to a catapult mechanism, and will presumably be found in other species, possibly including these birds, and indeed cheetah.

Distance fliers

The Arctic tern (Sterna paradisaea) breeds in the Arctic, and with a wingspan of 76 to 85 cm, flies to the Antarctic and back, an annual 50,000 miles (80,000 km), the longest migration of all, during which they see more daylight than any other creature. Living for up to 34 years, they could fly 1,500,000 miles – equivalent to travelling to the moon and back three times.

Endurance fliers

The sooty tern (Onychoprion fuscatus), with the physiological advantage of a minimalist sleep regimen of very large numbers of naps of a few seconds, may be continuously airborne for five years, especially before their first breeding. Electronically tagged bar-tailed godwits (Limosa lapponica) have been tracked flying non-stop for six to nine days between 4355 and 7258 miles (7000 and 11,600 km). The flight path showed that the birds could not have fed en route and would be unlikely to sleep. Researchers from the University of Groningen believe that the birds flap their wings non-stop for the entire journey, and that their utilising energy at a measured eight times resting basal metabolic rate (BMR) is, relatively, the greatest endurance feat in the vertebrate kingdom. The nearest equivalent human endurance effort is performed by professional road cyclists, who manage some five times BMR, for about five hours each day. ‘Lance Armstrong would be no competition for these birds’, they note. It is noteworthy that in the longer pigeon races, the owners ‘carbohydrate-load’ their birds, like human marathoners.

Altitude fliers

Ruppell's vulture (Gyps rueppellii) reaches an altitude of 11,277 m, some 2438 m higher than Mount Everest, while the bar-headed goose (Anser indicus) flies directly over the Himalayan mountain range during its semi-annual migration from wintering grounds in India to breeding grounds in Tibet. Flocks have been sighted above the summits of Mount Everest (8840 m) and Annapurna I (8077 m). At these altitudes their journey may be completed in a single, non-stop flight, with no acclimatisation time, flying from near sea-level in India to the altitude of Everest in less 24 hours. Oxygen levels at this altitude are only 20 per cent of sea level, yet the bar-headed goose increases its oxygen consumption 10- to 20-fold during its flight. Rheinhold Messner and Peter Habeler climbed Everest without supplementary oxygen in 1978, and Messner repeated this, solo, in 1980. Eland [Taurotragus species] may be found at 5000 m on Kilimanjaro, where a leopard [Panthera pardus] once died at 5600 m, commemorated at ‘Leopard Point’.)

Deep diving

The northern elephant seal (Mirounga angustirostris) can dive for two hours, and to 1275 m. Handrich and others (1997) noted that the ability to dive for long periods increases with body size. King and emperor penguins (Aptenodytes patagonicus and Aptenodytes forsteri) of 12 kg and 30 kg, respectively, dove to depths of 304 m and 534 m for 7.5 and 15.8 minutes. Blood oxygen levels were also measured by backpack recorders during the dive, using an oxygen electrode for continuous measurement. Emperor penguins and elephant seals tolerate exceptionally low levels of oxygen saturation, far below the limits of human beings and other animals. Sperm whales (Physeter macrocephalus) have been recorded as diving to 3000 m, while brotulid fish (Petrotyx species) descend to 8300 m.

In people, ‘free diving’ is the sport of diving without external breathing apparatus, using artificial ballasts, cables and sleds, to propel them deep on a single breath. In 2007, Austrian Herbert Nitsch made a record dive to 214 m in Spetses, Greece. The human water submersion (non-diving) ‘static apnoea’ breath-holding record is 11 min:35 sec, by Stefan Mifsud in 2009. This should not to be confused with the Guinness world record underwater breathing record, which permits the breathing of 100 per cent oxygen for up to 30 minutes beforehand, resulting in a breath-holding record in 2010 of 20 min:21 sec by Ricardo Bahia.

A loggerhead turtle (Caretta caretta) has been tagged and recorded as being submerged for 10 hr:14 min, the longest-duration dive for an air-breathing marine vertebrate.

Deep-diving whales and seals, relying on large oxygen stores in blood and muscle, have mass-specific blood volumes three to four times those found in terrestrial mammals, for example, 200 to 250 ml blood/kg, compared to the human value of 70 ml blood/kg. Their haemoglobin concentration is twice that of people, and their myoglobin is some 10 times that of human muscle. The bottlenose whale (Hyperoodon ampullatus) and sperm whale are exceptional divers, with reports of dives lasting as long as two hours after they were harpooned, and modern sonar tracking and attached time-depth recorders have recorded dives of 1.5 km, usually lasting between 20 and 60 minutes. These depths result in mechanical distortion and tissue compression, especially the lungs and other gas-filled body spaces, such as the middle ear cavity and sinuses in the skull, leading to what human divers call ‘the squeeze’. In some cetaceans, the middle ear cavity is lined with an extensive venous plexus, thought to engorge at depth, reducing the air space and preventing the squeeze. Kooyman and Ponganis (1998) noted that marine mammals lack the frontal cranial sinuses of terrestrial mammals, and also that ‘in the emperor penguin the swimming muscles alone can contain 19 per cent more oxygen than in their entire blood volume’.

Built for running. A number of animals show a range of physiological adaptations that make them efficient at speed or distance.

(clockwise from top left) The racehorse; greyhound; red kangaroo; camel; sled-dogs

Exercise physiology features

Oxygen uptake

The maximal oxygen uptake is a major parameter of aerobic performance. Elite human athletes range up to 96 ml O2/kg/min, horses up to 230 ml, greyhounds approximately 150 ml, with camels a surprisingly low 50 ml O2/kg/min. However, camels can sustain almost 100 per cent VO2max for 10 km in about 18 minutes. The VO2max of the pronghorn is reported as 9.5 l/min in a 32 kg animal, that is, 298 ml O2/kg/min (Lindstedt and others 1991), while the bat (Chiroptera species) and the pigeon range up to 300 ml, but with the humming bird (Trochilidae) and the Etruscan shrew (Suncus etruscus) topping them at 400 ml (Jones and Lindstedt 1993). However, non-allometric scaling, as above, markedly distorts such values according to size, so these are very approximate comparisons.

Gaits and running economy

The horse has a variety of gaits (eg, walk, trot, pace, canter and gallop), each optimising economy (oxygen usage per metre) at the progressively increasing speeds (Hoyt and Taylor 1981) – a typical pony would be about 1 m/sec for the walk, 3.2 m/sec for the trot, and 6.5 m/sec for the gallop.

Horses can be bred to ‘pace’, as a particular sport. Similarly, the camel tends to walk, pace or gallop, while children walk, skip or run, and adult humans walk or run, breaking naturally into a run just when walking becomes more energy costly at around 4.5 mph (7.2 kph), hence brisk walking is better exercise than slow jogging! Horses utilise the more stable and economic ‘transverse gallop’ (eg, lh, rh, lf, rf), but greyhounds (and cheetah and gazelle), with far more flexible backs and a need for agility as well as acceleration, use a rotary gallop (eg, lh, rh, rf, lf) with two suspension phases, contracted and extended. Camels (and giraffe [Giraffa camelopardalis]), being particularly long-legged, make considerable use of ‘pacing’ (lh and lf together, then rh and rf), this being their natural economic fast gait. In part, camels' very good running economy is due to their particularly long legs, although the counter-intuitive terrestrial champion of the ‘lowest normalised metabolic cost of transport’ appears to be the elephant.

Cardiac aspects

Approximate resting and maximal heart rates range from 30 to 190 bpm in human athletes, from 25 to 250 bpm in horses, from 100 to 300 bpm in greyhounds, and from 30 to 150 bpm in camels. The heart rate of the Nile crocodile (Crocodylus niloticus) in ‘drought hibernation’ can decrease to 2 bpm, and during an emperor penguin's 16 minute dive, its heart rate may decrease to 6 bpm over five minutes during the dive, with a minimum of 3 bpm. Equine cardiac outputs may rise to 250 to 450 l/min (L. E. Young, personal communication), compared to elite human rowers with up to 45 l/min.

Leinward (2011) has reported a remarkable cardiac anomaly occurring in the Burmese python (Python molurus), which shows an extraordinary and rapid increase of 40 per cent in ventricular mass in only 48 to 72 hours after their bi-annual feeding. Their metabolic rate may rise 40-fold; a remarkable example of extreme physiological up-regulation. Leinward (2011) at the University of Colorado identified a specific set of fatty acids that appear to promote myocardial hypertrophy, not hyperplasia, and a greater expression and activity of superoxide mutase – a cardioprotective free-radical-scavenger. This combination of fatty acids from postprandial python blood promoted healthy cardiac growth when injected into both pythons and mice. A possible future human therapeutic approach?

Pulmonary ventilation on exercise is over 2000 l/min and over 700 l/min respectively for the horse and camel, and up to 300 l/min for people (ie, for large elite rowers doing four-limbed exercise). Especially in the galloping horse, ventilation is strongly entrained to stride rate. The horse only breathes through its nostrils, and from 1800 BC in Egypt to 17th century Europe this was thought to be a limitation, hence many horses during that time had their nostrils slit to achieve the (equally futile) widening effect of the nasal strip as worn by some modern endurance athletes. However, the horse can ventilate 2000 litres air/min at a racing gallop, extracting about 60 litres O2/min, compared to a human maximum of 8 l/min from 300 litres of air.

The horse has no clavicle, so its forelimb motion is tied more to the ribs and spine giving greater compressive loading of the chest via the scapulae, especially when both forelegs strike the ground nearly simultaneously. In addition, the head and neck then act as a pendulum which, on lowering, presses the rib cage posteriorly adding to the scapular lateral compression of the thorax. Also simultaneously involved is the very important ‘visceral piston’ effect. A 500 kg horse has some 170 kg of abdominal contents anchored to the diaphragm; these move with the phase of the stride, shifting anteriorly due to body retardation on foot strike into stance phase, or moving posteriorly due to residual inertia on toe-off into suspension phase, both also being influenced by the change in slope of the body axis.

In the gallop these effects give, of necessity, a stride:breath ratio of 1:1 and the diaphragm itself may generate its greatest power at a natural resonant frequency that matches the preferred galloping stride frequency. At the canter, there is still the same 1:1 ratio, but cycles are shifted slightly out of phase – diagonal leg pair (non-lead fore and lead hind) strike together at the canter, transmitting less force to the rib cage.

Human athletes in swimming, rowing and canoeing also necessarily tie ventilation to stroke rate, though not to their advantage, and the wallaby (Macropodidae species) takes one breath per hop with its own visceral piston effect.

A tired horse may alter its gait to gain more air by prolonging the suspension phase; or it may exaggerate the vertical movements of head and neck, to pump the pendulum harder. Both lead to a reduction in speed. There is no comparable coupling of ventilation at trot, pace or walk. Left then right pressures on the rib cage tend to minimise the effect, and the level carriage of head and neck at trot and pace minimises the pendulum. Trotting dogs may show a form of respiratory coupling, where anterior left and right lobes fill alternately, corresponding to leg movement on a particular side, while the posterior lobes are driven together by a visceral piston. (A. M. Wilson, personal communication).

Loggerhead turtle (Caretta caretta). One turtle has been recorded as being submerged for over 10 hours, the longest-duration dive for an air-breathing

Photograph: Barcroft, Dunham

The aetiology of the common equine exercise-induced pulmonary haemorrhage (EIPH) is widely believed to be stress failure of pulmonary capillaries, but its pathophysiology is equivocal (L. E. Young, personal communication). Since the discovery of pulmonary hypertension and the elevated left atrial pressure (LAP) during maximum exercise in the horse, the role of the left heart in EIPH has been considered. Yet it is unclear whether LAP is a cause or effect of the comparative pulmonary hypertension. Dr Young notes that EIPH is relatively unique to the horse, although cyclists in highly competitive time trials may taste blood. Sudden death occurs for various cardiac reasons in human sport, but she describes it splendidly graphically, truly bringing pathophysiology to life, regarding flat-racing horses: ‘They usually drop dead right in front of the grandstand at the end of the race, when all of the products of metabolism are bathing the heart in an arrhythmogenic stew and the autonomic nervous system is in chaos.’ ‘Cyprinoid fish utilise acetaldehyde dehydrogenase and ethanol dehydrogenase to convert pyruvate to ethanol (alcohol), instead of to lactic acid, an interesting adaptation that might have made anaerobic work and training far more enjoyable for people!’

The haematocrit at rest is about 40 to 50 per cent in human beings, rising only slightly on exercise; equine values range from about 32 to 46 per cent at rest, but rise dramatically to 50 to 70 per cent on exercise, with the splenic erythrocyte release kicking in at heart rates of approximately 180 to 200 bpm. Values in greyhounds are from about 50 to 55 per cent at rest, also rising sharply to 60 to 65 per cent on exercise, while camels' values rise only slightly on exercise, from about 33 per cent at rest. In some countries, somewhat surprisingly, the doping agent erythropoietin (EPO), and possibly continuous erythropoietin receptor activator, has been used in horses, although there are reports of EPO triggering an equine aplastic anaemia in horses. Given that horses already physiologically ‘blood dope’ on the move, EPO use seems surprising. As with trained or pregnant people, the equine total erythrocyte mass increases with training (Evans 1994).

Blood lactate concentrations are a useful measure of anaerobic status, although its function in muscle fatigue has been strongly questioned. From resting levels of about 1 mmol/l at rest, these rise maximally in humans to approximately 24 mmol/l, in horses to approximately 36 mmol/l, in dogs to approximately 34 mmol/l, and in camels to approximately 24 mmol/l. Regarding muscle buffering, Wise and others (2007) reviewed pH homeostasis in muscle during intense exercise, which is maintained by exporting H+ and by physicochemical buffering, via organic phosphate, bicarbonate and the histidine-containing dipeptides (HCD) (carnosine, anserine and balenine). Of these, phosphates and bicarbonate are limited due to their involvement in other metabolic reactions. HCDs on the other hand show a 40-fold variation in muscle across species, being uninvolved in metabolic processes. Carnosine is the only HCD found in human beings. Anserine is found in dogs and camels, and balenine in deer (Cervidae) (along with carnosine) and whales and dolphins (Delphinidae). Quantitative HCD differences between species seem proportional to the H+ ions generated during bouts of exercise. Whale muscle contains the highest level of HCDs of any species, over 350 mmol/kg dry matter, which exceeds even their level of stored carbohydrate.

Wise and others (2007) noted HCD levels in the thoroughbred horse of approximately 110 mmol/kg, and of 82 mmol in the greyhound. Racing camels have 70 mmol and people vary from 13 to 44 mmol/kg, depending on diet and sport. Avian flight muscles in fast take-off birds have particularly high HCD content, such as the turkey (Meleagris species) with 275 mmol/kg and the pheasant with 220 mmol/kg, but levels are much lower in endurance fliers, such as geese, with approximately 60 mmol/kg. Humans have twice the levels of HCD in type II (‘fast’) fibres compared to type I (‘slow’), but racehorses have five times as much HCD in type II as type I. The overall low levels of HCDs in people are consistent with their original lesser need for sprinting for survival, illustrating that levels in different species reflect their relative needs for H+ ion buffering, probably indicating evolutionary adaptation.

What one might call the ‘macho anaerobes’ (apart from bacteria) are intertidal bivalves such as oysters, mussels and clams. Oysters can live indefinitely without oxygen. Cyprinoid fish, such as carp and goldfish, can survive hypoxia for over 100 days. They utilise acetaldehyde dehydrogenase and ethanol dehydrogenase to convert pyruvate to ethanol (alcohol), instead of to lactic acid, an interesting adaptation that might have made anaerobic work and training far more enjoyable for people!


Camels and many antelope (eg, eland [Tragelaphus oryx]), act as a heat sink, which allows body temperature to rise by up to 6°C in a day. The hump in camels acts as a dorsal heat shield, and they have a large sternal ‘pedestal’ together with heavily calloused knees to insulate them when they kneel on the hot sand. Human beings can sweat over 2 l/hr, and approximately 11 l in eight hours has been recorded. A horse can sweat 1.5 l in two minutes, and up to 16 l/hr. Human beings and horses are the ‘sweatiest’ animals, with horses sweating far more in absolute terms. However, using body mass as a criterion, people come out sweatier, 46 ml/kg/hr compared to 36 ml/kg/hr – but using the surface area criterion horses come out sweatier at 3.2 l/m2/hr compared to human rates of 1.8 l/m2/hr. This is because the horse has about 1 m2 of skin per 90 to 100 kg bodyweight, compared to an adult human's 1 m2 skin per 35 to 40 kg bodyweight, that is, the horse has only about 40 per cent surface area relative to body mass, compared to people, so it has to lose about two-and-half times more heat per unit area. However, equine sweat contains a detergent, latherin, which makes the sweat ‘froth’, thus more water stays on the skin to remove heat via latent heat of evaporation. Horses may drink up to 80 litres a day, but camels can drink 80 litres in 10 minutes. The broad-tailed humming bird (Selasphorus platycerus) can drink up to five times its bodyweight of water (in nectar, a 30 per cent glucose solution). ‘Human beings and horses are the “sweatiest’ animals, with horses sweating far more in absolute terms. However, using body mass as a criterion, people come out sweatier'

Camels conserve fluid via a comparatively low glomerular filtration rate, very long loops of Henle, a slow gastrointestinal absorption of water and a very delayed water diuresis, and they cyclically secrete and reabsorb water into their forestomach and produce desiccated faeces (which can be immediately burned on the fire as fuel). Dogs may lose 60 per cent of their body heat via evaporation from their tongue, and birds do not sweat.

Regarding cold, Handrich and others (1997) showed that the abdominal temperatures of king penguins may fall to as low as 11°C during sustained deep diving (this is lower than it was calculated to be due to ingestion of cold food), and suggest that this ‘temperature induced metabolic suppression’ helps maintain the dive. The wood frog (Rana sylvatica) can survive –8°C over winter (as can its parasites), even its eyeballs freeze. Its rate of hepatic glycolysis is accelerated at around 0°C, and the released glucose acts as antifreeze (Carwardine and Cox 2005).

Energy storage in collagen

A running horse, camel, greyhound, antelope, cheetah or a bounding kangaroo – or a running person – all behave biomechanically like a pogostick or a bouncing ball. The ball is deformed and stopped by the impact, losing kinetic energy, but by deformation it stores elastic strain energy. The elastic recoil propels the ball almost to the original height, losing elastic strain energy and regaining kinetic energy. The ‘tendon can store a remarkable energy density owing to the high strains it can withstand and its high elastic modulus. Strained tendon can store about 60 times as much energy per unit mass as a helical steel spring’ (Wilson and Lichtwark 2011). Human and equine runners use only approximately 50 per cent of the oxygen expected, and kangaroos only about 35 per cent, by converting kinetic energy to stored elastic strain energy primarily in the tendons and ligaments. Human, horse and kangaroo tendon are similar. In the kangaroo particularly, oxygen uptake does not increase proportionately with speed, due in part to the increased relative contribution of recycled recoil energy. Around 93 per cent of work done in the stretching tendon is returned as elastic recoil, with an optimal stretch of around 5 to 12 per cent. Above this, it tends to rupture. Rubber can stretch 200 per cent, but a strong material, such as collagen, stretching a little can store as much energy as a weak material, such as elastic, stretching a lot.

Citius, Altius, Fortius. The peregrine falcon is the fastest animal recorded, and has been measured travelling at 161 mph when in the dive

McNeill Alexander (1992) has shown that people running at 4.5 m/s (270 m/min – approximately six-minute mile pace) exert peak ground forces of 2.8 times bodyweight, with forces over three times this on the Achilles tendon (4.7 kN at 70 kg bodyweight). This would stretch the Achilles by approximately 5 per cent, to store about 35 J of elastic strain energy in the tendon. Elastic strain energy is also stored in the ligaments of the human foot, where on foot-strike, it flattens by about 10 mm, and the arch-ligament system returns 78 per cent of the energy in its elastic recoil. A typical human middle-distance runner uses about 100 J at each stride, losing and regaining approximately 35 J elastically in the Achilles tendon and 17 J in the arch of the foot, with the remainder coming from aerobic and anaerobic metabolism in muscle. In the running animals, the muscular work may be similarly or even more reduced by storing and returning elastic strain energy in highly specialised spring-like muscle-tendon units, with each leg acting like a pogostick which in the horse is tuned to stretch and recoil at about 2.5 strides per second (Wilson and others 2001). Wilson and others note that in a 500 kg horse, about 1000 J of elastic energy are stored in the digital flexor tendons and suspensory ligaments of each leg in each stride. A peak tendon strain of 8 to 12 per cent during locomotion equates to a tendon elongation of 70 mm and, assuming similar aponeurosis strains, a muscle elongation of 40 mm. One problem with this degree of energy storage is that the tendons can ‘over-heat’. The 93 per cent of the energy that is returned, implies that 7 per cent is retained as heat, which can raise tendon temperature to 44 to 46°C, to the possible detriment of the tendon's fibroblasts (Wilson and Goodship 1994).

Citius, Altius, Fortius

In conclusion, ‘Citius, Altius, Fortius’ is the Olympic motto, but if we allowed the rest of the animal kingdom into the Games, and it were to select the peregrine falcon (161 mph), Ruppell's vulture (37,000 feet) and the 190-ton blue whale as its representatives for ‘Faster, Higher, Stronger’, we could not offer much competition! Even if restricted to terrestrial animals, we could be up against the cheetah (65 mph), the red kangaroo (3.1 m) and the 12-ton bull African elephant – worth a thought when viewing the adulation given to our species' Olympic outliers in July. At least human beings are physically the most versatile single species, which is what the Games display overall.


Data in this paper have come from many sources, mainly journals, texts, presentations and personal communications; however, some of the performance data are unavoidable ‘best estimates’. Regarding personal communications, I would like to give special thanks to Alan Wilson (RVC) and Lesley Young, for invaluable help and information, and also to thank: Brian Wheeldon, Hilary Clayton (University of Saskatchewan), David Robertshaw (Cornell), R. McNeill Alexander (Leeds University), Dan Lieberman (Harvard) and Tim Noakes (Capetown University); and finally, former colleagues at the International Equine Institute, Limerick University, particularly Frank McGourty, together with Sean Arkins and Justin Keating. My brilliant and now sadly recently deceased Glasgow early mentors were Professor Sir J. W. Black and Professor W. F. H. Jarrett, all gratitude to them.


  • Emeritus Professor Sharp is Co-founder and former Director of the British Olympic Medical Centre, Northwick Park Hospital, Harrow, Middlesex HA1 3UJ


Sign up for job alerts

Vet Record Careers features a comprehensive database of vet jobs.
Browse for your next job, or set personalised job alerts so you hear about vacancies as soon as they go online.
Find out more at Vet Record Careers

Free Sample Issue

This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of the Veterinary Record.
View free sample issue >>