Phil Dhingra
Prof. Jablonski
Anth. Sci. 131/231 - Primate Evolution, Stanford University
Published on Philosophistry
May 25, 2004
COMPARATIVE BIPEDALISM - how the rest of the animal kingdom walks on two legs
Sigmund Freud often remarked that great revolutions in the history of science have but one common, and ironic, feature: they knock human arrogance off one pedestal after another of our previous conviction about our own self-importance. --Stephen Jay Gould

This anthropocentric bias extends to discussions about bipedalism in humans. Bipedality is often casually cited as the prime evolutionary adaptation that distinguishes humankind by enabling the use of tools. While it is indeed true that the evolution of bipedalism in humans is coincident with the large increases in brain size that separate humans from other primates, bipedality in general has a long and varied history among many animals. What follows is a tour of examples from living species of animals, including lizards such as the basilisk, birds such as the ostrich, mammals such as kangaroos, and finally primates and humans. Two unifying and often-times overlapping theories about the evolution of bipedalism are explored: (1) bipedalism evolved to increase speed to escape from predators or to better catch prey, such as in lizards, and (2) bipedalism evolved as a result of different selection pressures acting on the forelimbs and hindlimbs, such as in birds with their dependence on wings for flight and humans with their dependence on hands for feeding.


Bipedalism is not a trait that uniquely evolved in humans but has occurred in five different instances among vertebrates. Twice it evolved in reptiles, first in thecodonts-the reptilian ancestors to crocodiles, birds, and dinosaurs-and second in lizards, such as the Callisaurus draconoides (a.k.a. zebra-tailed lizard). Birds could represent another instance of bipedalism but it is commonly believed that they evolved from dinosaurs. Finally, there are three instances of bipedal evolution in mammals: the marsupial ricochet form (kangaroos), the placental ricochet form (kangaroo mice), and primates and humans with their alternating gait (Snyder 1962). By definition, these five independent evolutions of bipedalism indicate that using solely two legs for locomotion is a convergent trait. This is important because it knocks down the notion that bipedalism is something that makes humans special; there has been an anthropocentric bias toward focusing on bipedalism as a defining trait of man (Du Brul 1962). Perhaps it defines humans as unique among primates, but it is by no means the defining trait for humankind among the whole Animal Kingdom. Plus, during the Mesozoic Era 251 to 65 million years ago, dinosaurs "ruled the earth," many of which, such as the Tyrannosaurus Rex, the Pterodactyl, and the Velociraptor, were bipedal.

There are, however, within the individual branches of bipedalism, interesting differences and ways of organizing the different types. A convenient categorization is into three areas: bipedalism to gain speed in order to escape predators or in order to improve predation, such as is seen in lizards; bipedalism that evolves to free the forelimbs for feeding such as in primates and humans; and thirdly, bipedalism to allow for the flight of birds (Howell 1944; Du Brul 1962).


A few living species of reptiles have bipedal postures and it is widely assumed that many dinosaurs were bipedal as well. Some bipedal lizards include the basilisk, the zebra-tailed lizard (Callisaurus draconoides), and the spiny-footed lizard (Acanthodactylus erythrurus). Bipedalism in lizards is characterized by an intermittent running style with high maneuverability. Since lizards are horizontally balanced-as opposed to vertically balanced in humans-the variable location of the center of mass is important for the extent to which bipedality is used. Their center of mass is closer to the rear, which allows for heading to be easily changed during movement. This also permits lift to occur when the lizard accelerates against air resistance (Aerts 2003). Vertical inclinations and the position of the lizard's tail move the center of mass closer to the pelvis and create a semi-erect posture. It has been observed that the tail is of principal importance to bipedalism in lizards. When Snyder (1949) cut off half to two-thirds of the tail of bipedal lizards, they could not maintain their bipedal posture. Another biological feature of bipedal lizards is shorter forelimbs which pushes the center of mass further back (Snyder 1962). The benefit of bipedalism for lizards is that it increases maximal running speed and is energetically advantageous (Aerts 2003). Snyder (1952) and Howell (1944) also noted that speed increases with bipedalism. The benefit of top-speed increases would be for occasional, but critical, escape or predation. In the case of the basilisk lizard, bipedalism also allows for running over water. An important distinction between bipedal lizards and other bipedal animals is that there are not many differences in the musculoskeletal structure of quadrupedal and bipedal lizards (Snyder 1962; Du Brul 1962). Bipedal lizards only have slight modifications that confer them the ability to balance on two legs. This distinction has important implications as to how bipedalism evolved in lizards.

According to one theory, lizards developed bipedalism as a result of having stronger hindlimbs for speed combined with shorter forelimbs and a natural lift that occurs when a lizard runs. Du Brul (1962) offered the following suggestion as to how bipedalism evolved in lizards. Lizards started with a trot. Then, as increases in speed were selected for, larger hindlimbs evolved; this is perhaps due to inherent limitations of position as well as the structure of moving a cylindrical body (for example, there may be mechanical advantages to having real-wheel drive cars as opposed to front-wheel drives). This makes the length of the rear-limb stride greater than the front, making the front feet interfere with the movement of the hindlimbs. Smaller front feet are therefore selected for, and as the lizard gains speed, its body achieves lift and the specimen starts running. Having a tail also helps move the lizard's center of mass to the rear to maintain balance (Snyder 1949). Even if larger hind limbs do not necessarily offer a natural mechanical advantage, such as in the real-wheel drive example, evolution could select for either forelimbs or hindlimbs of larger size by chance. In the case where a lineage evolves with larger hindlimbs, then smaller forelimbs would be selected for by reasons mentioned above, and then lift would be achieved while running.

But could bipedalism in lizards have started as a spandrel? A spandrel is a term in evolutionary science that frames an evolved feature in a species as an accidental outcome of other evolved processes. Aerts (2003) theorized that bipedalism in lizards is a spandrel-like accident. He first noted that lizards have a caudal, or rear-located, center of mass-a feature selected for to aid in higher maneuverability and an increased ease of changing heading while moving. Aerts then modeled the lizard Acanthodactylus erythurus and simulated it going through motion. He discovered that acceleration of the body caused important lift which would allow for passive bipedality. Also, the existence of the lizard's tail allowed for instantaneous manipulation of the position of the center of mass, further aiding in bipedal locomotion. Because of the ease with which lift occurn, then, the theory is that bipedalism evolved as an exploitation of the existing body structure of lizards (Aerts 2003). The spandrel theory is probably partially true as the ease at which lift occurs cannot be discounted. However, bipedality, even if it is passive, requires some locomotor skills. In lizards, this would be the skills needed to manipulate the tail when lift is desired or manipulate the hindlimbs in a non-quadrupedal pattern. Whether or not bipedal lizards already had these instincts is hard to tell. It is understandable why we are tempted to hypothesize that lizards would just figure out how to run when they encounter lift; since humans exhibit flexibility with the manipulation of our limbs for different locomotor patterns (crawling, swimming, jumping, climbing, running, walking, sitting), we naively assume other animals have similar skills. The evolution of bipedalism in lizards, then, was a hybrid spandrel-adaptive process, formed by a combination of existing larger hindlimbs, smaller forelimbs, selection pressures for speed, and a natural lift that occurs upon acceleration.


All birds are bipedal. In most species, bipedalism is used for landing and obtaining proximity to food. Two major taxa have lost the capacity for flying: the Palaeognathiformes (flightless birds like emus and ostriches) and Sphenisciformes (penguins) (Cubo & Casinos 1997). What is interesting about these two is that walking represents a global adaptive strategy. Furthermore of interest is that a comparison of the wing bone structures of Palaoegnathiformes and Sphenisciformes reveals that parallel processes led to flightlessness. In other words, flightlessness is a convergent trait in these birds (Casinos & Cubo 2001). This bolsters the notion that full dependence on bipedalism can evolve easily in evolution. According to Cub and Casinos (1997), it only takes a few thousand years for the flightless condition to emerge.

Bipedalism evolved in birds through selective pressures acting separately on the forelimbs and the hindlimbs. For birds, the forelimbs ultimately evolved for flight and so were free of axial loads. The leg bones, on the other hand, were strained by axial, bending, and twisting loads. Whatever benefits were afforded by flight required that the hindlimbs bear the burden for terrestrial locomotion (Casinos & Cubo 2001). A competing theory based on holism might counteract this "differential selective pressures" theory (Dullemeijer 1974). According to holism, the whole organism, all features included, is selected for and not individual parts. This is relevant because the legs of birds are also specially adapted to take on the hard impact of landing. However, generally speaking, different selective pressures can be traced to different limb systems of the same animal (Casinos & Cubo 2001). The notion of "different selective pressures" connects the evolution of bipedalism in birds to, as we will see further, the evolution of bipedalism in humans: while birds depend on their forelimbs for flight, primates depend on their hands for feeding (Du Brul 1962). Cursorial locomotor adaptations would therefore naturally lead to bipedalism in birds and humans.


Humans have complete dependence on bipedalism. Many architectural changes accompany this feature in humans and in some primates as well. Many of the major changes are in the locomotor apparatus itself, the legs and the feet: the feet are less hand-like and more like pedestals with the tarsals (toes) shrunken; the legs are bent in at the knee and able to lock in to allow for sustained erect postures. The backbone in humans is also bent more like a spring to reduce the shock from walking and to bring the body more vertically oriented above the center of gravity. Thirdly, humans have a flatter figure, as if they are walking tablets: the thorax or torso is thinner from front to back but wider from side to side and the face is flattened (Du Brul 1962). This last bit is interesting because it shows a marked structural body change associated with "new uses of the dimensions of space." Early creatures in the Proterozoic Era were bacteria-like and not very mobile, with hardly any symmetry to their structure as if they only used a single dimensional point of space. Chordates then developed the innovation of having an axis with two end points: they developed a head with a concentrated bundle of neurons and a tail for navigation. Limbs developed first as fins for more navigation, but then became legs and arms on the land which allowed the animal to lift itself up into space. Primates exhibit extended limbs to really take advantage of three dimensions by crawling up trees and extracting food sources. Humans tap beyond three dimensions by becoming a-dimensional: flatter tablet-like figures permit unprecedented communication between humans via body language, facial gestures, and conversation which create a dimensionless network among groups. While this "a-dimensional theory" does not explain how bipedalism evolved in humans, it describes how significant locomotor apparatus is to how an organism will conduct its life. Below are some theories for how bipedalism evolved in humans.

The hypotheses for the evolution of bipedalism are many and varied and are summarized nicely by Videan (2002). Videan experimented with bonobos and chimpanzees--humans' nearest relatives-and focused on four theories that provide a nice tour and test for how human bipedalism developed. First are two theories that tested positive by Videan's experiments. The Carry Hypothesis establishes that bipedalism developed to permit the carrying of infants, tools, and food over large distances. This, as is suggested, evolved to exploit widely dispersed resources (Hewes 1961; Lovejoy 1988). This is relevant given the Earth history during the evolution of hominids which saw frequent declines in the areas of tropical forests, the main niche of primates. Videan's tests revealed that the availability of portable food-items items successfully predicted an increased rate of bipedality in bonobos and chimpanzees. Videan also tested the Forage Hypothesis which asserts that bipedalism's evolutionary advantage was the ability to reach the low branches of woody vegetation and to permit travel between food-patches stretched across long distances (Rose 1984; Hunt 1994). The researcher's experiments showed that the presence of food items in elevated locations increased bipedality in chimpanzees (Videan 2002). This behavior of using a locomotor adaptation to occasionally shift between dispersed food sources is similar to flamingos. Flamingos, a mostly bipedal species, will only fly when there is a predator or when their food supplies run short, in which case they move to another lake.

Videan (2002) also tested two other theories for bipedalism. The Vigilance Hypothesis, as its name suggests, describes bipedalism as increasing the visual horizon of the primate, providing a competitive advantage for foraging in savanna-woodland habitats and to allow for better escaping from predators (Day 1986). Vigilance is also frequently correlated with bipedality in other mammals such as nonhuman primates, bipedal rodents, and kangaroos (Videan 2002). However, when Videan (2002) introduced visual barriers about a meter high, bipedality did not increase in either chimpanzees or bonobos. A fourth theory, the Display Hypothesis, offers that primates stand up in order to threaten others by increasing their apparent size. Upright posture also frees the hands to display objects (Kortlandt 1980; Jablonski & Chaplin 1993). Videan (2002) introduced branches in order to coax the chimpanzees and bonobos to use them for more frequent displays; his results, however, showed that the frequency of displays did not increase. All four theories, though, appear credible. Three of the theories-the Carry, Vigilance, and Forage Hypotheses-benefit from connecting bipedalism as a specific adaptation to possible ecological variables such as the decreasing forest areas during the evolution of hominids. The Display Hypothesis benefits from being a simple way in which females could propel sexual selection by selecting more and more erect males who win in display contests.


Bipedalism occurs in other mammals as well. The kangaroos, which employ a hopping locomotion, are the most obvious example. Also of interest are the bipedal kangaroo mice and kangaroo rats. An interesting study by Harris (1984) compared bipedal kangaroo rats with quadrupedal pocket mice among open and closed microhabitats and with varying dispersals of seeds. The bipeds gathered more food in the open habitats and tended to select large or clumped seeds there. The quadrupeds, on the other hand, selected small or evenly scattered seeds in shrubby microhabitats. The conclusion is that the difference lays in the energetic advantages of bipedalism that aid in exploiting open habitats despite increased exposure to predators (Harris 1984). Kangaroos are also known for exploiting open habitats in Australia, and so one could assume that similar adaptive benefits to bipedalism came to them as well. Bennet (2000) suggested that bipedal hopping may have evolved in Kangaroos "from the bounding or half-bounding gaits that are commonly used by many of the smaller marsupials, such as bandicoots, dasyurids, potoroos, and some of the smaller macropods" (Bennett 2000, pp. 726). The image that comes to mind considering kangaroos and their bipedal exploitation of open habitats is that of the ancestral hominid in the open plains of the savannah; similar ecological pressures may have faced both hominids and kangaroos when bipedalism was evolving.


A major unifying trait of bipedalism is that it is associated with occasional but critical cursorial movements. Lizards, which are quadrupedal most of the time, switch to bipedal running in order to escape predators (or catch prey). In hominids, during the deforestations after the Miocene Epoch (twenty-four to five million years ago) bipedalism was critically useful to migrate to new locations for food. Flamingos, likewise, will do the same thing, occasionally shifting between lakes.

Another unifying trait in bipedalism is the two general ways in which it has developed. In the case of thecodonts, lizards, and the non-primate mammals, bipedalism evolved for its advantages to speed in either escaping prey or pursuing predators. In the case of primates and birds, bipedalism evolved because the forelimbs were highly specialized for other purposes.

Framing human evolution into the larger picture of the evolution of all members of the Animal Kingdom provides broader insight and accuracy into understanding our origins. This kind of perspective may not have been easily accessible ten years ago. Recent innovations, such as the Internet and fast library databases, have put "information at our fingertips," allowing researchers to quickly cull bits of information from a diverse set of fields. Having access to more organized data helps students synthesize and grasp were we came from. Ten years from now, though, it will be even more surprising what novel insights await us.

Literature Cited

Aerts P, Van Damme R, D'Août K, Von Hooydonck BV. 2003. Bipedalism in lizards: whole-body modeling reveals a possible spandrel. Online: University of Antwerp.

Bennett MB. 2000. Unifying principles in terrestrial locomotion: Where do hopping Australian marsupials fit in? Physiological and Biochemical Zoology 73.6: 726-735

Casinos A, Cubo J. 2001. Avian long bones, flight and bipedalism. Comparative Biochemistry and Physiology Part A Molecular & Integrative Physiology 131: 159-167

Cubo J, Casinos A. 1997. Flightlessness and long bone allometry in Palaeognathiformes and Sphenisciformes. Netherlands Journal of Zoology 47.2: 209-226
Day MH. 1986. Bipedalism: pressures, origins, and modes. In Major Topics in Primate and Human Evolution, ed. Wood B, Martin L, Andrews P, pp. 188-201. New York: Cambridge University Press

Du Brul EL. 1962. The general phenomenon of bipedalism. American Zoologist 2: 205-208

Dullemeijer P. 1974. Concepts and approaches in animal morphology, pp. 6. The Netherlands: Koninklijke Van Gorcum & Comp. B.V.

Harris JH. 1984. An experimental analysis of desert rodent foraging ecology. Ecology (Tempe) 65.5: 1579-1584

Hewes GW. 1961. Food transport and the origin of hominid bipedalism. American Anthropology 63: 687-710

Howell AB. 1944. Speed in animals, pp. 270. Chicago: University Chicago Press

Hunt KD. 1994. The evolution of human bipedality: ecology and functional morphology. Journal of Human Evolution 26: 183-202

Jablonski NG, Chaplin G. 1993. Origin of habitual terrestrial bipedalism in the ancestor of the Hominidae. Journal of Human Evolution 24: 259-280

Kortlandt A. 1980. How might early hominids have defended themselves against large predators and food competitors? Journal Human Evolution 9: 79-112

Lovejoy CO. 1988. Evolution of human walking. Scientific American 259: 118-125

Rose MD. 1984. Food acquisition and the evolution of positional behavior: the case of bipedalism. In Food acquisition and processing in primates, ed. Chivers DJ, Wood BA, Bilsborough A, pp. 509-524. New York: Plenum Press.

Snyder RC. 1949. Bipedal locomotion of lizard Basiliscus basiliscus. Copeia 2: 129-137

Snyder RC. 1962. Adaptations for bipedal locomotion in lizards. American Zoologist 2: 191-203

Videan EN, McGrew WC. 2002. Bipedality in chimpanzee (Pan troglodytes) and bonobo (Pan paniscus): Testing hypotheses on the evolution of bipedalism. American Journal of Physical Anthropology 118.2: 184-190