Quadrupedalism

Quadrupedalism—locomotion using four limbs for weight bearing and propulsion—is the primitive condition for terrestrial tetrapods. The overwhelming majority of amphibians, reptiles, and mammals are quadrupedal, and the stance is deeply conserved across vertebrate phylogeny. What makes quadrupedalism in dinosaurs remarkable is that it arose secondarily: all dinosaurs trace their ancestry to obligately bipedal forms. Barrett & Maidment (2017) emphasized that the reversion from bipedality to quadrupedality is known exclusively within Dinosauriformes among all tetrapod clades. This stands in sharp contrast to the several independent origins of bipedality from quadrupedal ancestors in marsupials, rodents, hominids, and certain archosaurs. The unidirectional nature of this transition outside dinosaurs underscores the uniqueness of the dinosaurian locomotor evolutionary trajectory.

The common ancestor of Dinosauria and its closest outgroups (lagosuchids, pterosaurs) were bipedal, establishing bipedality as the ancestral state for the clade (Sereno 1997). Quadrupedality subsequently evolved at least four independent times.

Sauropodomorpha underwent the earliest transition, during the Late Triassic to Early Jurassic (~215–200 Ma). Transitional taxa such as Aardonyx celestae (Yates et al. 2010) and Melanorosaurus readi document a spectrum of intermediate conditions between obligate bipedality and obligate quadrupedality. Aardonyx was recovered as the sister taxon to a clade of obligatorily quadrupedal sauropodomorphs (Melanorosaurus + Sauropoda), placing it at the crux of the bipedal-to-quadrupedal transition. Ledumahadi mafube (McPhee et al. 2018), from the earliest Jurassic of South Africa, was a giant (~12 tonnes) that walked on four legs but retained a flexed, semi-sprawled forelimb posture rather than the fully columnar stance of derived sauropods—illustrating that the transition to columnar quadrupedality was not instantaneous. Importantly, phylogenetic analyses of sauropodomorphs reveal multiple switchbacks between bipedality and quadrupedality within the lineage, demonstrating that evolution does not follow a single directional trajectory (UC Berkeley, Understanding Evolution 2018).

Thyreophora (armored dinosaurs) transitioned to quadrupedality early in dinosaur history, during the basal Jurassic. Scelidosaurus harrisonii (Owen 1861) is the earliest known quadrupedal thyreophoran. All subsequent stegosaurs and ankylosaurs were obligate quadrupeds. Stegosaurs exhibited extreme limb length disparity, with hind limbs more than twice the length of the forelimbs, and were undoubtedly graviportal (Britannica).

Ceratopsia transitioned to quadrupedality from the late Early to early Late Cretaceous. Basal ceratopsians such as Psittacosaurus were predominantly bipedal; bone histology studies by Zhao et al. (2013) suggest an ontogenetic postural shift in Psittacosaurus lujiatunensis, potentially from quadrupedal hatchlings to bipedal adults. The transition to obligate quadrupedality occurred within Neoceratopsia, reaching its extreme expression in ceratopsids such as Triceratops.

Hadrosauriformes evolved quadrupedality during the Cretaceous. Iguanodontian ornithopods such as Iguanodon bernissartensis have been variably interpreted as obligate bipeds, obligate quadrupeds, or facultative bipeds/quadrupeds throughout the history of their study. Maidment & Barrett (2014) found that hadrosaurs possess all osteological correlates associated with quadrupedality and were likely predominantly quadrupedal, although capable of bipedal locomotion. A 2026 reassessment by Pintore et al. further examined the locomotor capabilities of Iguanodon and related taxa using osteological correlates.

A possible fifth independent reversion may have occurred in rhabdodontid ornithopods, though the timing remains unclear due to extensive ghost lineages separating this clade from other ornithopods (Weishampel et al. 2003). Silesauridae, the nearest outgroup to Dinosauria, may also have reverted to quadrupedality from a bipedal ancestor during the Middle Triassic, but lack of articulated material makes this difficult to confirm (Langer et al. 2013).

The transition from bipedality to quadrupedality in dinosaurs involved a suite of skeletal modifications that have been the subject of extensive study. Barrett & Maidment (2017) and Maidment & Barrett (2014) identified several osteological correlates that are strongly associated with quadrupedal stance across dinosaur phylogeny. These include the presence of an anterolateral process on the ulna that cups the radius and restricts forearm rotation; hoof-like manual ungual phalanges adapted for weight bearing rather than grasping; a femur longer than the tibia; a straight femur in lateral view; reduction of the fourth trochanter on the femur; a transversely expanded iliac blade; and relatively shorter feet in proportion to the combined femoral and tibial length.

Mallison et al. (2014, PLOS ONE) conducted a quantitative geometric morphometric analysis of the radius across 293 specimens (189 mammals, 49 dinosaurs, 35 squamates, 16 birds, and 5 crocodilians) and demonstrated that no dinosaurian clade had the ability to cross the radius over the ulna. This means that all quadrupedal dinosaurs maintained parallel antebrachial elements, unlike parasagittal mammals which cross the radius over the ulna for active pronation. This finding confirms that quadrupedal dinosaurs evolved a fundamentally different forearm configuration from mammalian quadrupeds, constrained by their bipedal ancestry. The anterolateral process of the ulna, convergently present in all quadrupedal dinosaur clades, may have served primarily to stabilize the radius during locomotion rather than to actively restrict rotation.

Despite convergently acquiring quadrupedal gait, different ornithischian clades achieved this through disparate musculoskeletal mechanics. Dempsey et al. (2023), using three-dimensional multi-body dynamics models of 17 ornithischian taxa, demonstrated quantitatively that the three major quadrupedal ornithischian clades evolved distinct forelimb musculature, particularly around the glenohumeral (shoulder) joint. Major differences in glenohumeral abduction–adduction and long axis rotation muscle leverages were the key drivers of mechanical disparity.

Stegosaurs and ankylosaurs evolved high glenohumeral abduction and lateral rotation moment arms consistent with a wide-gauge, splayed "press-up" stance, although the two groups achieved this through different osteological configurations. Hadrosauriform ornithopods evolved high glenohumeral adduction moment arms consistent with a narrow-gauge stance. Ceratopsids exhibited yet another pattern, with high glenohumeral lateral rotation moment arms but also strong adduction capabilities. At the elbow joint, high extension moment arms were convergent among most quadrupeds, but abduction–adduction mechanics differed strongly.

These findings refute earlier qualitative hypotheses of functional convergence in major quadrupedal clades and demonstrate that morphological convergence does not necessarily imply functional similarity—an important principle in evolutionary biomechanics. Phylogenetic constraints from ancestral bipedal forms within each clade appear to have shaped the divergent ecomorphological pathways of their quadrupedal descendants.

The relationship between quadrupedal stance and large body size in dinosaurs is complex. In sauropods, the transition to quadrupedality was clearly a prerequisite for the evolution of extreme gigantism. Sauropods evolved fully columnar, parasagittal limbs—analogous to the pillar-like legs of elephants—that positioned the limb bones directly beneath the body to support masses exceeding 50 tonnes (Hutchinson 2021). McPhee et al. (2018) noted that sauropods were unique among quadrupedal dinosaurs in having this fully erect, graviportal stance, allowing efficient weight support in accordance with biomechanical scaling principles: as body mass increases, limbs must become more columnar to maintain adequate safety factors.

However, Barrett & Maidment (2017) cautioned that body size increase alone was not the primary driver of the quadrupedal transition in ornithischians. They found that neither increased body size nor the acquisition of dermal armour appeared to have played a significant direct role in triggering the ornithischian quadrupedal reversions. Instead, increased head size may have shifted the center of mass anteriorly in ceratopsians, and the evolution of herbivory—with an enlarged gut displacing the center of mass forward—is a plausible driver in both ornithischians and sauropodomorphs, though this hypothesis remains difficult to test rigorously.

Hutchinson (2021) placed these considerations within the broader context of the evolutionary biomechanics of giant land animals. Giant tetrapods—whether mammalian or dinosaurian—face nonlinear constraints as mass increases. Limb bone geometry becomes more robust (shorter, thicker), effective mechanical advantage (EMA) plateaus beyond approximately 300 kg in mammals, and locomotor abilities are progressively reduced. Sauropod dinosaurs demonstrate that body plans quite different from mammalian models (e.g., lacking the capacity for active pronation, possessing columnar metacarpal and metatarsal arrangements) can also support extreme masses, indicating multiple evolutionary solutions to the biomechanical constraints of gigantism.

Not all dinosaurs were strictly obligate bipeds or obligate quadrupeds. Several taxa exhibit evidence of facultative quadrupedality or bipedality—the capacity to shift between two-legged and four-legged locomotion depending on circumstance. Iguanodon has long been interpreted as a facultative biped that habitually walked on all fours but could rear onto its hind limbs for running or defense (Norman 1980). Hadrosaurs similarly may have used both gaits.

Ontogenetic shifts in locomotor mode have been proposed for several taxa. Zhao et al. (2013) used bone histology to suggest that Psittacosaurus lujiatunensis underwent a postural shift during growth, potentially from quadrupedal juveniles to bipedal adults—the reverse of the pattern proposed for some ornithopods such as Maiasaura, where juveniles may have been bipedal and adults predominantly quadrupedal (Dilkes 2001). In sauropodomorphs, ontogenetic changes in limb proportions in Massospondylus suggest a possible shift from bipedal to quadrupedal posture during growth (Reisz et al. 2005). These ontogenetic transitions echo the phylogenetic transitions within the broader clade and suggest that the developmental flexibility to shift between stances may have facilitated the evolutionary transition itself.

Trackway (ichnofossil) evidence provides direct confirmation of quadrupedal locomotion in dinosaurs. Quadrupedal trackways preserve both manus (hand) and pes (foot) impressions in a characteristic pattern. Sauropod trackways are among the best-studied and are classified as narrow-gauge (with manus prints close to the midline) or wide-gauge (with manus prints placed laterally), with titanosaurs generally associated with wide-gauge trackways. Ceratopsian and thyreophoran trackways also preserve four-limbed locomotion patterns. Ichnological evidence complements skeletal reconstructions and helps constrain gait parameters such as stride length, gauge width, and speed estimates, providing a more complete picture of how quadrupedal dinosaurs actually moved through their environments.