# Analysis of Asymmetrical Gaits > How does a cheetah’s gallop differ from a horse’s, or a mouse’s bound from a deer’s pronk? Using slow-motion film of 79 mammal genera, Milton Hildebrand showed that just five timing measurements describe every gallop and bound, and built the terminology and gait graph that researchers still use. He also argued that these gaits first evolved to help small animals escape. Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/1977-hildebrand-analysis-asymmetrical-gaits/ ## Citation and access - **Authors:** Milton Hildebrand - **Year published:** 1977 - **Journal or source:** Journal of Mammalogy - **Full citation:** Hildebrand, M. 1977. Analysis of asymmetrical gaits. Journal of Mammalogy 58(2): 131-155. - **DOI:** 10.2307/1379571 - **Access status:** Publisher link only - **Where to find it:** https://doi.org/10.2307/1379571 ## Tags - **Species:** Mammals > African buffalo; Mammals > Baboon; Mammals > Bandicoot; Mammals > Bear; Mammals > Bison; Mammals > Blackbuck; Mammals > Bush hyrax; Mammals > Camel; Mammals > Cattle; Mammals > Cheetah; Mammals > Chimpanzee; Mammals > Chipmunks; Mammals > Coati; Mammals > Dall sheep; Mammals > Deer; Mammals > Domestic cat; Mammals > Domestic dog; Mammals > Duiker; Mammals > Fur seal; Mammals > Galago; Mammals > Gaur; Mammals > Gazelle; Mammals > Gibbon; Mammals > Giraffe; Mammals > Goat; Mammals > Gorilla; Mammals > Ground squirrel; Mammals > Guenon; Mammals > Hare; Mammals > Hartebeest; Mammals > Horse; Mammals > Hyena; Mammals > Impala; Mammals > Jumping mouse; Mammals > Kangaroo; Mammals > Kangaroo rat; Mammals > Kob or waterbuck; Mammals > Langur; Mammals > Lemur; Mammals > Llama; Mammals > Macaque; Mammals > Malagasy giant rat; Mammals > Mangabey; Mammals > Mara; Mammals > Mongoose; Mammals > Mouse; Mammals > Muntjac; Mammals > Nilgai; Mammals > Noctule; Mammals > Okapi; Mammals > Olingo; Mammals > Orangutan; Mammals > Otter; Mammals > Peccary; Mammals > Prairie dog; Mammals > Pronghorn; Mammals > Rhinoceros; Mammals > Roan or sable antelope; Mammals > Sea lion; Mammals > Sheep; Mammals > Skunk; Mammals > Small-toothed palm civet; Mammals > Spider monkey; Mammals > Spiral-horned antelope; Mammals > Springhare; Mammals > Tahr; Mammals > Tapir; Mammals > Tasmanian devil; Mammals > Tree kangaroo; Mammals > Tree squirrel; Mammals > Treeshrew; Mammals > Vole; Mammals > Water deer; Mammals > Weasel; Mammals > Wildebeest; Mammals > Wolverine - **Tracking:** Gaits - **Behaviour and ecology:** Behaviour; Movement - **Methods and technology:** Body-size comparison - **Data type:** Existing data; Modelling or theoretical; New field data - **Observer qualification:** Not needed - **Discipline:** Behaviour and ecology - **Year published:** 1977 ## Detailed keywords - **Species:** Studied: mammals of 79 genera in 11 orders, recorded at genus level only (the author leaves out species names because the data do not justify that level). They include marsupials (Tasmanian devil Sarcophilus, bandicoots Isoodon and Perameles, tree kangaroo Dendrolagus, kangaroo Macropus), a bat (Nyctalus), tree shrews (Tupaia, Urogale), primates (lemur, galago, spider monkey, mangabey, guenon, macaque, baboon, langur, gibbon, chimpanzee, gorilla, orangutan and a human child), hares (Lepus), rodents (prairie dog Cynomys, chipmunk Eutamias, squirrel Sciurus, ground squirrel Spermophilus, kangaroo rat Dipodomys, springhare Pedetes, Malagasy giant rat Hypogeomys, vole Microtus, mouse Mus, jumping mouse Zapus, mara Dolichotis), carnivores (dog Canis, bear Ursus, olingo Bassaricyon, coati Nasua, wolverine Gulo, weasel Mustela, skunks Mephitis and Spilogale, otter Lutra, small-toothed palm civet Arctogalidia, mongooses Herpestes and Mungos, hyena Hyaena, cheetah Acinonyx, cat Felis), sea lion Zalophus and fur seal Callorhinus, hyrax Heterohyrax, horse Equus, tapir Tapirus, rhinoceros Rhinoceros, and artiodactyls (peccary Tayassu, camel Camelus, llama Lama, muntjac Muntiacus, water deer Hydropotes, deer Odocoileus, giraffe Giraffa, okapi Okapia, pronghorn Antilocapra, bison, cattle Bos including gaur Bos gaurus, nilgai Boselaphus, buffalo Syncerus, Tragelaphus, duiker Cephalophus, hartebeest Alcelaphus, wildebeest Connochaetes, Hippotragus, Kobus, impala Aepyceros, blackbuck Antilope, gazelle Gazella, goat Capra, tahr Hemitragus, sheep Ovis including Ovis dalli). Common names here follow the genus names in the paper’s Table 1; the paper itself gives mostly genus names. Discussed but not filmed: frogs and toads (anurans), small crocodiles (Crocodylus), lizards and extinct reptiles. - **Location and habitat:** Not reported as field sites. Animals were filmed in the open or in enclosures (none on treadmills; Heterohyrax in an exercise wheel). Film also came from the Institut für den Wissenschaftlichen Film (Göttingen), the Granada TV Network (London), Walt Disney Productions, university film libraries and about 20 other sources. Author based at the Department of Zoology, University of California, Davis, United States. - **Topics:** asymmetrical gaits (gallops and bounds); gait analysis; footfall timing; contact intervals; fore and hind leads; leading and trailing feet; midtime lag; gathered and extended suspensions; lead sequence (transverse gallop, rotary gallop, half bound, bound, pronk or stott, crutch walk); canter and lope; gait diagrams, gait graphs and footfall formulas (104 formulas); gait terminology; gait and body size, speed, agility and maneuverability; turning (the inside forefoot leads); evolution of asymmetrical gaits for escape; interface between symmetrical and asymmetrical gaits; companion to Hildebrand (1976) on symmetrical gaits. - **Statistics or models:** Smoothed records (“plots”) averaging three or more successive strides; 100 exploratory two-variable (bivariate) graphs by order, five weight classes (logarithmic) and six body-proportion classes; visual centres of weight as approximate averages; graphical gait graph (hind support against midtime lag) with overlays. No statistical tests. The author judges multivariate analysis questionably valid here. Notes: the abstract gives contact intervals of 16 to 70 percent of the cycle, while the text says fore contacts are “rarely less than 13 percent” for consecutive strides and hind contacts fall below 18 percent only in single strides of exceptional runners. Fig. 20’s caption keys formulas to Figs. 11 to 17 and Fig. 21’s to Figs. 10 to 17. - **Data type:** New field data: original analysis of slow-motion film, mostly at 64 frames per second (some at 18, 72, 100 or 200), 331 passes by animals of 79 genera; dates of filming not given. Existing data: all or most of the data for Dendrolagus, Mungos, Felis, Nyctalus and Connochaetes came from other authors. Modelling or theoretical: a system of analysis, a terminology and an evolutionary interpretation. Submitted 23 April 1976, accepted 23 September 1976. ## Synopsis ### 1. What they did Gaits come in two classes. In symmetrical gaits (walks, the pace and the trot) the two feet of a pair land evenly spaced in time. In asymmetrical gaits (gallops and bounds) they do not: the two feet of a pair land close together, then there is a pause. The author had already analysed symmetrical gaits, and this companion paper is the first general analysis of asymmetrical ones. He studied slow-motion film of 331 passes in front of the camera by mammals of 79 genera in 11 orders, from bandicoots and mice to horses, giraffes and cheetahs. Most of the film was his own; some came from film libraries and television and film companies. For each record he averaged three or more successive strides to smooth out the quirks of single strides, then built gait diagrams showing when each foot was on the ground. He set out to: - Find the smallest set of measurements that fully describes any asymmetrical gait. - Propose a terminology for these gaits. - Describe the asymmetrical gaits of many mammal genera. - Interpret gait choice in relation to animal group, body size, speed, way of life and habitat. ### 2. What they found Five measurements describe every gallop and bound All the timing of events at the ground can be captured by five variables: how long the hind feet touch the ground (hind contact), how long the forefeet touch (fore contact), the hind lead and the fore lead (the delay between the first, or trailing, foot of a pair and the second, or leading, foot), and the “midtime lag” (how far the middle of forefoot support follows the middle of hind foot support in the stride). Contact times - Contact times shorten as animals go faster. In fast runners from the horse, dog and cheetah to the jumping mouse and rabbit, the hind foot is down for about 20% of the stride. In steady movement neither fore nor hind contact is likely to be over 70% of the stride. - Fore and hind contacts are about equal for horses and their relatives, most even-toed hoofed animals and most carnivores. Fore contacts are about 80% of hind contacts for most rodents, several carnivores (including the cheetah at speed) and some hoofed animals, and 70% or less for hares at speed, jumping mice, kangaroo rats and springhares. - Smaller, lighter animals tend to have relatively shorter fore contacts and more variation from stride to stride, which seems to go with agility, maneuverability and rough ground relative to body size. Leads - Leads vary widely within a genus. The fore lead is shorter than the hind only for most apes and a few others. It is about equal to, or up to 1¼ times, the hind lead for most carnivores, hoofed animals and horses, and more than 3 times the hind lead for the bounding and half-bounding marsupials, carnivores, rabbits, rodents and hyrax. - The fore lead tends to grow relative to the hind lead as body size decreases, as agility and speed increase, and as contact times shorten. - All galloping mammals use both right and left leads. When turning, the inside forefoot leads. Suspensions and the gait graph Plotting hind support against midtime lag gives a “basic gait graph” with four zones: gaits with no airborne phase (slow gallops of many mammals, fast ones of the giraffe), a gathered suspension with legs under the body (fast gallops of horses and camelids), an extended suspension with legs stretched out (deer, some rodents), and both suspensions (fast gallops of rabbits and many carnivores and hoofed animals). Lead sequences - In a transverse gallop the forefeet lead on the same side as the hind feet; in a rotary gallop on the opposite side. In a bound both pairs land together; in a half bound the hind feet land together but the forefeet do not; in a pronk (stott) all four land at about the same moment. - The transverse gallop is typical of primates, most carnivores, horses and rhinoceros, and camelids. The rotary gallop is preferred by the tapir and giraffids, and by the fastest, most maneuverable runners (cheetah, blackbuck, gazelle). The half bound and bound are typical of most marsupials studied, tree shrews, rodents, small carnivores, sea lions and fur seals, and hyrax. Several mammals switched to the half bound or bound as speed increased. - Overall, heavier, less agile animals and slower movement go with transverse gallops with no suspension or a gathered one; lighter, more maneuverable animals and faster movement go with an extended suspension or both, and with rotary gallops, half bounds or bounds. Footfall formulas He mapped 104 footfall formulas (the order in which combinations of feet support the body) onto the gait graph and showed those typical of 55 genera. There are many more formulas than had been described before, and the horse alone may use about 21. Terminology He proposes describing a gait in words by combining speed, lead sequence, the length of the fore and hind leads, and the kind and length of suspension. For example, a cheetah’s very fast gallop is “a very fast rotary gallop having a short hind and very long fore lead, and short gathered and extended suspensions.” ### 3. What they concluded - Five variables are both necessary and sufficient to describe asymmetrical gaits. Published descriptions should give variables from which the full gait diagram can be rebuilt. - Despite wide variation within and overlap between genera, gait choice correlates roughly with lead sequence, body weight, speed, agility and maneuverability. - Asymmetrical gaits probably evolved in small animals, at least once in amphibians (a frog’s hop is a kind of bound), several times in reptiles and probably in some mammals, first for rapid escape. Bounds, half bounds and gallops in both lead sequences are probably all very old. Gaits with short leads or none and an extended suspension are considered primitive. - Most small mammals are versatile in their gaits, probably more than has been reported; large or highly specialised animals such as the rhinoceros, kangaroo and giraffe give up versatility. - Symmetrical gaits are a special case of the general relationships described here, and the two classes meet at a broad interface. - The author hoped to prepare films of these gaits for study and teaching. ### 6. Other technologies, skills or innovations - Slow-motion film analysis frame by frame, at 18 to 200 frames per second. - Computer plotting of 100 exploratory bivariate graphs. - Gait diagrams, the basic gait graph with overlays, and a standard footfall-formula notation (always shown with a right hind lead and starting from the left hind footfall). ### 8. Implications for tracking (Original Wisdom's observations) _These are Original Wisdom's observations, not the authors'._ - The order and spacing of footfalls in gallops and bounds is what produces the track groups trackers read in the field. The paper’s terms (transverse and rotary gallops, bounds, half bounds, pronks, leading and trailing feet) give a precise vocabulary for describing those patterns. - The paper notes that, although leads are measured in time, they correlate with the length of the step taken by the leading foot relative to leg length, and that the inside forefoot leads when turning. Both points may help when interpreting changes in a gallop trail. - The wide stride-to-stride variation within a genus is a reminder that a single track group may not show an animal’s typical gait. ## Cite this summary Lawrence, K. 2026. Summary of: Analysis of Asymmetrical Gaits (Hildebrand, 1977). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/1977-hildebrand-analysis-asymmetrical-gaits/