Dinosaur physiology

Dinosaurs dominated terrestrial ecosystems for more than 160 million years, yet understanding how these animals actually functioned as living organisms — how they breathed, grew, regulated their body temperature, reproduced, and perceived their environment — remained largely speculative until the latter decades of the twentieth century. The breakthrough came not from new skeletons but from new methods: thin-sectioning bones under polarised light, scanning skulls with computed tomography, measuring stable isotope ratios in fossilised enamel and eggshell, and identifying molecular biomarkers preserved in ancient tissues. Together these approaches have transformed dinosaur physiology from a domain of educated guesswork into one of the most data-rich frontiers in palaeobiology.

The results overturn the Victorian image of dinosaurs as sluggish, cold-blooded reptiles. Most lineages maintained elevated body temperatures, grew at rates that rival or exceed those of modern mammals, breathed through lungs more efficient than any mammal's, and exhibited reproductive strategies that bridge the gap between reptilian egg-laying and avian brooding. This article surveys the major lines of physiological evidence, from bone microstructure to brain endocasts, and traces the scientific debates that have shaped current understanding.

The endothermy debate

When Richard Owen coined the name Dinosauria in 1842, he envisioned its members as advanced reptiles — large and powerful but fundamentally ectothermic, relying on external heat sources to regulate body temperature much as modern lizards and crocodilians do. This assumption went largely unchallenged for over a century. Dinosaurs were depicted in art and museum displays as lumbering, tail-dragging creatures, their metabolic capabilities implicitly equated with those of living reptiles. The paradigm began to shift in the late 1960s and early 1970s, during what has come to be called the "dinosaur renaissance."

The most influential voice of the renaissance was Robert T. Bakker, who in 1972 published a landmark paper in Nature arguing that anatomical and ecological evidence pointed to endothermy — internally generated metabolic heat — in dinosaurs.¹ Bakker marshalled several lines of reasoning. First, he noted that the predator-to-prey ratios preserved in Mesozoic fossil assemblages resembled those of modern mammalian communities (roughly 3% predator biomass) rather than ectothermic ones (which can support far higher predator fractions, because cold-blooded predators require less food). Second, he observed that dinosaurs exhibited erect, parasagittal limb postures — a stance associated in living animals with sustained, high-energy locomotion. Third, he pointed to the widespread distribution of dinosaurs at high palaeolatitudes, including regions with pronounced seasonality, environments that would pose severe challenges for large ectotherms reliant on ambient warmth.

Bakker's arguments ignited a debate that persisted for decades. Critics countered that predator-prey ratios in the fossil record are heavily influenced by taphonomic bias, that erect posture need not require endothermy, and that Mesozoic climates were warmer globally, potentially permitting large ectotherms at high latitudes. The debate could not be resolved by skeletal morphology and ecological inference alone. Resolution required direct evidence of metabolic rate, growth speed, and body temperature — evidence that would eventually come from bone histology, isotope geochemistry, and molecular palaeontology.

Bone histology and growth rates

The interior microstructure of fossil bone, accessible through thin-sectioning and examination under polarised light, provides one of the most powerful windows into dinosaur physiology. The pioneering work of Armand de Ricqlès in the 1960s and 1970s established that dinosaur bones are predominantly composed of fibrolamellar tissue — a type of bone in which rapidly deposited, poorly organised woven-fibred matrix is penetrated by dense vascular canals and subsequently infilled by lamellar bone. In living animals, fibrolamellar bone is characteristic of birds and mammals — organisms with high metabolic rates and fast growth — while slower-growing ectotherms such as crocodilians and lizards typically deposit lamellar-zonal bone, a slower-forming tissue with fewer vascular canals and prominent growth rings (lines of arrested growth, or LAGs).

The prevalence of fibrolamellar bone across Dinosauria, from basal sauropodomorphs to derived coelurosaurs and ornithischians, indicates that rapid growth was ancestral to the clade, not a derived condition limited to the bird lineage. In 2001, Gregory Erickson and colleagues published the first comprehensive set of dinosaur growth curves, constructed by counting LAGs in histological sections of long bones from multiple ontogenetic stages of the same species.² Their results demonstrated that dinosaurs exhibited sigmoidal growth trajectories — slow juvenile growth, a rapid adolescent acceleration, and a plateau at skeletal maturity — similar in shape to the growth curves of birds and mammals. Small dinosaurs grew at rates comparable to those of marsupials, while large species attained growth rates rivalling those of eutherian mammals and precocial birds. The giant sauropods grew at rates comparable to those of baleen whales of similar body mass.²

The most dramatic growth data came from Tyrannosaurus rex. Histological analysis of a growth series of specimens revealed that T. rex underwent a remarkable adolescent growth spurt between approximately 14 and 18 years of age, during which it gained an estimated 2.1 kg per day, reaching skeletal maturity at roughly 20 years and a maximum lifespan of perhaps 28 years.² Such rapid growth requires sustained high metabolic rates to fuel the biosynthesis of bone, muscle, and other tissues, and it is difficult to reconcile with a purely ectothermic physiology. The presence of LAGs in dinosaur bone does not contradict this inference; LAGs indicate seasonal pauses in growth (possibly during resource-scarce periods) and are found in the bones of some living endotherms, including high-latitude birds and ruminant mammals.

Growth rates and life-history parameters for selected dinosaurs^{2, 15}

Isotopic and molecular evidence for body temperature

While bone histology reveals growth rate, it does not directly measure body temperature. That capability arrived with the development of isotopic thermometry techniques, most notably clumped isotope palaeothermometry. In standard oxygen isotope analysis, the ratio of ¹⁸O to ¹⁶O in biogenic apatite reflects a combination of body temperature and the isotopic composition of ingested water, making it difficult to isolate temperature alone. In 2006, Romain Amiot and colleagues addressed this problem by comparing the oxygen isotope compositions of dinosaur phosphatic tissues with those of co-occurring ectothermic vertebrates (crocodilians, turtles) from the same localities, thereby controlling for local water composition. Across eleven Cretaceous sites spanning a range of palaeolatitudes, dinosaur body temperatures clustered between 36 and 38 °C — a range consistent with modern endotherms — while co-occurring ectotherms tracked ambient environmental temperature and varied accordingly.⁶

A more precise approach came in 2011, when Robert Eagle and colleagues applied clumped isotope thermometry to sauropod dinosaur teeth. This technique measures the preferential bonding of heavy isotopes ¹³C and ¹⁸O within carbonate minerals, a thermodynamically controlled process that depends solely on the temperature at which the mineral formed, independent of the isotopic composition of body water. Eagle's team analysed tooth enamel from large Jurassic sauropods including Brachiosaurus and Camarasaurus and obtained body temperatures of 36–38 °C, closely matching those of modern mammals.¹⁴ The technique was subsequently extended to dinosaur eggshells. In 2020, Robin Dawson and colleagues used clumped isotope thermometry on eggshell carbonates from representatives of all three major dinosaur clades — Ornithischia, Sauropodomorpha, and Theropoda — and found that all three groups maintained warm body temperatures significantly above the estimated ambient environmental temperatures at their respective fossil sites.¹⁸ This result strongly suggests that metabolic thermoregulation was ancestral to Dinosauria as a whole, not a derived feature of any single lineage.

The most recent advance has come from molecular proxies for metabolic rate. In 2022, Jasmina Wiemann and colleagues developed a technique using Raman and Fourier-transform infrared spectroscopy to quantify lipid peroxidation byproducts (advanced lipoxidation end products) preserved in fossil bone. Because these waste molecules accumulate in proportion to oxygen consumption — a direct proxy for metabolic rate — their abundance can distinguish endotherms from ectotherms. Wiemann's team analysed bone from 55 genera of extant and extinct animals and found that dinosaurs, including even early forms such as Herrerasaurus, showed metabolic signals in the endothermic range. Pterosaurs likewise scored as endothermic, while ornithischians showed somewhat lower but still elevated metabolic rates. The ancestral ornithodiran condition, the study concluded, was already endothermic, with metabolic rates increasing further along the avian stem lineage.²⁰

Mesothermy and gigantothermy

Not all researchers accept a simple endothermic model for all dinosaurs. In 2014, John Grady and colleagues proposed a middle ground by analysing the relationship between growth rate, body mass, and metabolic rate across a large dataset of both living and extinct vertebrates. After controlling for body size and estimated body temperature, they found that dinosaur metabolic rates fell between those of modern endotherms and ectotherms, clustering instead with a handful of living animals — such as the leatherback sea turtle, great white shark, and tuna — that maintain elevated body temperatures through metabolic heat production but do not defend a precise thermal set point as rigorously as birds and mammals do. Grady termed this intermediate metabolic strategy "mesothermy."¹⁶

The mesothermy hypothesis generated vigorous discussion. Several technical comments challenged the statistical methods and the assignment of metabolic categories, noting that the boundaries between endothermy, mesothermy, and ectothermy are not discrete but form a continuum. The subsequent molecular data from Wiemann and colleagues, which placed most dinosaurs firmly in the endothermic range using an entirely independent proxy, has shifted the consensus away from a strictly mesothermic interpretation.²⁰ Nonetheless, the Grady study made an important conceptual contribution by emphasising that the endothermy-ectothermy dichotomy oversimplifies a spectrum of thermoregulatory strategies.

A related concept is gigantothermy (also called inertial homeothermy), which applies specifically to very large animals. Because heat loss scales with surface area while heat production scales with volume, a sufficiently large animal — even one with a low mass-specific metabolic rate — will retain metabolic heat effectively enough to maintain a stable, elevated body temperature. Biophysical models have demonstrated that a sauropod the size of Brachiosaurus (approximately 30–50 tonnes) would experience negligible diurnal temperature fluctuation even with a metabolic rate only modestly above that of a crocodilian, simply because its thermal inertia is enormous.¹⁵ Gigantothermy may have been particularly relevant for the largest sauropods, permitting them to maintain mammal-like body temperatures without necessarily requiring mammal-like metabolic rates. For smaller dinosaurs, however, gigantothermy cannot account for the elevated and stable body temperatures indicated by isotopic data, making some form of internally driven thermogenesis necessary.

Respiratory system

The lungs of modern birds are uniquely efficient among vertebrates. Rather than the tidal, bellows-like breathing of mammals (in which air flows into and out of blind-ended alveoli), birds employ a flow-through system in which air passes unidirectionally across the gas-exchange surfaces (parabronchi) during both inhalation and exhalation. This design, maintained by a series of air sacs that act as bellows anterior and posterior to the rigid lung, extracts oxygen far more efficiently than the mammalian plan and is a key factor enabling sustained powered flight and the high metabolic demands of avian endothermy.

Fossil evidence demonstrates that this respiratory architecture did not originate with birds but evolved deep within the saurischian dinosaur lineage. The vertebrae of theropod and sauropod dinosaurs are permeated by pneumatic cavities — chambers, fossae, and foramina formed by the invasion of soft-tissue diverticula extending from the pulmonary air sac system. Mathew Wedel's detailed studies of sauropod vertebral pneumaticity showed that the distribution of these pneumatic features maps closely onto the air sac system of modern birds, with cervical vertebrae pneumatised by diverticula of cervical air sacs, dorsal vertebrae by thoracic and abdominal sacs, and in derived forms such as diplodocids and titanosaurs, even sacral and proximal caudal vertebrae invaded by postcervical diverticula.³ The degree of skeletal pneumatisation in large sauropods exceeds that seen in any living bird, consistent with the hypothesis that air sacs reduced skeletal density and facilitated respiration in these enormous animals.

In 2005, Patrick O'Connor and Leon Claessens extended this analysis to theropod dinosaurs, demonstrating that non-avian theropods possessed both anterior (cervical and clavicular) and posterior (abdominal) air sacs — the complete complement needed to drive unidirectional airflow through a parabronchial lung.⁴ Their work established that the basic avian pulmonary design was in place by at least the Late Jurassic, well before the origin of flight. The finding was dramatically reinforced in 2010 when Colleen Farmer and Kent Sanders demonstrated unidirectional airflow in the lungs of the American alligator, an archosaur that lacks air sacs and is separated from dinosaurs by more than 240 million years of independent evolution.¹³ The implication is that unidirectional pulmonary airflow is ancestral to Archosauria as a whole and may have been a key innovation enabling archosaurs to thrive in the low-oxygen conditions of the Middle to Late Triassic. Dinosaurs inherited and elaborated this respiratory system, coupling it with extensive pneumatic air sacs to achieve gas exchange efficiencies unmatched by any other terrestrial vertebrate group.

Reproductive biology

Dinosaurs were egg-laying animals, and their reproductive biology has been illuminated by an increasingly rich record of fossil eggs, nests, embryos, and associated adult specimens. The earliest dinosaurs likely laid soft-shelled, parchment-like eggs resembling those of modern turtles and many lizards. In 2020, Mark Norell and colleagues used phylogenetic ancestral-state reconstruction combined with chemical and ultrastructural analysis of eggshells from Protoceratops and Mussaurus to demonstrate that the ancestral dinosaurian eggshell was soft and uncalcified. Hard, calcified eggshells — which resist desiccation and permit the eggs to be incubated in open nests — evolved independently at least three times within Dinosauria: in theropods (including birds), in hadrosaurs/other derived ornithischians, and in sauropods.¹⁹

The most informative nesting sites come from derived theropods, particularly maniraptorans. David Varricchio and colleagues described clutches of the troodontid Troodon formosus from the Late Cretaceous Two Medicine Formation in Montana, revealing that the animal produced two eggs at a time (consistent with paired oviducts), arranged them in a circular or semicircular pattern within a sediment mound, and incubated them using a combination of sediment cover and direct body contact.⁸ Multiple specimens of the oviraptorid Citipati from the Djadokhta Formation of Mongolia have been found preserved in a brooding posture atop their clutches, with the forelimbs draped over the nest periphery and the hindlimbs crouched symmetrically on either side — a posture strikingly similar to that of modern ground-nesting birds. These discoveries provide direct evidence that at least some non-avian dinosaurs engaged in avian-style contact incubation.

Incubation duration has been estimated by counting daily growth lines (von Ebner increments) in the teeth of dinosaur embryos still within their eggs. In 2017, Erickson and colleagues used this method on embryos of the small ceratopsian Protoceratops andrewsi and the large hadrosaur Hypacrosaurus stebingeri, finding incubation periods of approximately three and six months, respectively.¹⁷ These durations are far longer than expected for birds of comparable egg size (which typically incubate for only 40–80 days) and instead resemble the incubation times of reptiles. The finding has profound implications: it suggests that non-avian dinosaurs had not yet evolved the rapid embryonic development that characterises modern birds, and that prolonged incubation may have been a significant vulnerability, tying adults to nesting sites for months at a time and exposing eggs to predation and environmental hazards.

Molecular evidence has also revealed reproductive physiology. In 2005, Mary Schweitzer and colleagues identified medullary bone — a calcium-rich, highly vascularised tissue deposited inside the long-bone marrow cavities of female birds in the weeks before egg-laying — in a Tyrannosaurus rex femur (MOR 1125).⁵ Medullary bone serves as a labile calcium reservoir for eggshell formation and is found in all living birds during their reproductive cycle. Its presence in T. rex not only confirms that the specimen was a reproductively active female but also indicates that the physiological mechanism of calcium mobilisation for eggshell production was shared between non-avian theropods and modern birds.

Digestive systems

The digestive physiology of dinosaurs cannot be observed directly but can be inferred from coprolites (fossilised faeces), tooth morphology, gut contents, and the presence or absence of gastroliths (stomach stones). The most celebrated dinosaur coprolite is a massive specimen from the Late Cretaceous Frenchman Formation of Saskatchewan, described in 1998 by Karen Chin and colleagues. At over 2.4 litres in volume, it is the largest known carnivore coprolite and contains 30–50% fragmented bone, including identifiable pieces from a juvenile ornithischian dinosaur. Its stratigraphic and geographic context, combined with its enormous size, strongly implicate a tyrannosaurid — most likely Tyrannosaurus rex — as the producer.⁹ The abundance of crushed bone indicates that the animal's digestive system was capable of processing skeletal material, a capacity consistent with the massive jaw musculature and robust dentition of tyrannosaurids. However, the bone fragments were only partially dissolved, suggesting that digestion was less thorough than in modern mammalian bone-crushers such as hyenas, which reduce bone to fine powder.

For herbivorous dinosaurs, a key question concerns how plant material was processed without mastication. Sauropod dinosaurs, despite their enormous body sizes, had relatively small, simple teeth incapable of chewing. It has long been speculated that sauropods compensated by using gastroliths to grind plant matter in a muscular gizzard, analogous to the gastric mill of modern herbivorous birds such as ostriches and gallinaceous fowl. However, Oliver Wings and P. Martin Sander tested this hypothesis in 2007 by comparing the mass of gastroliths found associated with sauropod skeletons to those experimentally measured in ostriches. In birds with a functioning gastric mill, gastroliths constitute roughly 1% of body mass, whereas in sauropods the associated stones represent less than 0.1% — an order of magnitude too low to indicate an effective grinding function.⁷ Wings and Sander concluded that sauropods likely lacked a bird-style gastric mill and instead may have compensated through greatly extended gut retention times, permitting microbial fermentation to break down plant material over days or weeks. This strategy, combined with the enormous gut volume available in an animal the size of Brachiosaurus or Diplodocus, may have been sufficient to extract adequate nutrition from even low-quality forage.¹⁵

Ornithischian dinosaurs, in contrast, evolved sophisticated oral processing. Hadrosaurs possessed dental batteries containing hundreds of closely packed teeth arranged in columns, with worn teeth continuously replaced from below. The grinding surfaces of these batteries produced complex shearing and crushing motions during jaw closure, rivalling the efficiency of mammalian mastication. Ceratopsians developed similar dental batteries with a predominantly shearing function, suited to processing tough, fibrous vegetation. These innovations in oral processing, absent in sauropods, likely reduced the digestive burden on the gut and may have enabled ornithischians to exploit food sources that sauropods could not.

Sensory capabilities and brain structure

Advances in computed tomography (CT) scanning have allowed palaeontologists to produce detailed three-dimensional reconstructions of the cranial endocast — the internal mould of the braincase — and the inner ear of dinosaurs, providing unprecedented insight into their neurological and sensory capabilities. The endocast does not represent the brain itself (the brain does not fill the entire cranial cavity in most reptiles), but it preserves the relative sizes and proportions of major brain regions, from which functional inferences can be drawn.

Lawrence Witmer and Ryan Ridgely produced landmark CT-based reconstructions of the braincase and inner ear of multiple tyrannosaur specimens, including Tyrannosaurus rex. Their analyses revealed that T. rex possessed unusually large olfactory bulbs relative to cerebral hemisphere size, indicating an acute sense of smell — a finding quantified independently by Darla Zelenitsky, François Therrien, and Yoshitsugu Kobayashi, who developed an olfactory ratio (the greatest diameter of the olfactory bulb divided by the greatest diameter of the cerebral hemisphere) and applied it across Theropoda.^{10, 11} Their results demonstrated that tyrannosaurids and dromaeosaurids had significantly enlarged olfactory bulbs, suggesting that smell played a major role in their ecology — perhaps in locating carrion, tracking prey, or navigating territories. Ornithomimosaurs and oviraptorids, by contrast, had reduced olfactory ratios, consistent with a lesser reliance on olfaction and potentially greater emphasis on visual or other sensory modalities.

The inner ear, reconstructed from the bony labyrinth visible in CT scans, provides information about hearing acuity and vestibular function (balance and head movement coordination). Witmer and Ridgely showed that T. rex had a long cochlear duct, indicating sensitivity to a broad range of sound frequencies, with an estimated best hearing range at relatively low frequencies — consistent with communication over long distances or detection of large-bodied prey.¹¹ The semicircular canals of the inner ear reveal habitual head posture: in T. rex, the orientation of the lateral semicircular canal indicates that the animal held its head with the snout tilted slightly downward, roughly horizontal, providing a wide binocular visual field. Forward-facing eyes with overlapping visual fields, as inferred from orbital morphology, would have given T. rex stereoscopic depth perception — an advantage for a predator that needed to judge distances accurately when striking at prey.

Endocast studies of other dinosaur groups have revealed contrasting sensory specialisations. Hadrosaurs possessed relatively large cerebral hemispheres and well-developed optic lobes, consistent with complex social behaviour and reliance on vision. Troodontids had among the largest brain-to-body-mass ratios of any non-avian dinosaur, with expanded cerebral hemispheres, large optic lobes, and relatively large eyes — features that have been interpreted as adaptations for nocturnal or crepuscular activity. Ankylosaurs, by contrast, had small brains relative to body size, with modest cerebral hemispheres and large olfactory bulbs, suggesting a slow-paced lifestyle in which smell was more important than rapid cognitive processing.

Locomotion and energetics

The biomechanics of dinosaur locomotion offer an independent line of evidence bearing on metabolic physiology. In 2009, Herman Pontzer, Vivian Allen, and John Hutchinson developed biomechanical models to estimate the metabolic cost of walking and running in 14 extinct bipedal dinosauriforms, using relationships between limb dimensions, body mass, and locomotor energetics established from living animals.¹² Their approach exploited the fact that the cost of transport in legged animals scales predictably with hip height and body mass, and that the minimum metabolic rate required to sustain locomotion at ecologically relevant speeds provides a lower bound on resting metabolic rate.

Pontzer and colleagues found that even at modest walking speeds, the estimated metabolic costs for virtually all bipedal dinosaurs fell within the range of modern endotherms and well above the aerobic capacity of comparably sized ectotherms. A 6-tonne Tyrannosaurus walking at a moderate pace, for instance, would have required a sustained metabolic output that a crocodilian-grade metabolism could not support. Only the very smallest dinosauriforms in the analysis approached the ectothermic range, and even these were borderline.¹² The biomechanical data thus independently corroborate the histological, isotopic, and molecular evidence for elevated metabolism across Dinosauria. The erect, parasagittal limb posture that Bakker had identified as circumstantial evidence for endothermy in 1972 turns out, when subjected to rigorous biomechanical analysis, to require endothermic-grade metabolic rates for sustained function in animals of dinosaurian body sizes.

Quadrupedal dinosaurs present additional biomechanical questions. The columnar, graviportal limb design of sauropods — with straight, pillar-like legs, reduced distal musculature, and minimal joint flexion — minimised the energetic cost of supporting and moving immense body mass but limited maximum speed. Trackway evidence and biomechanical modelling suggest that the largest sauropods were restricted to walking gaits with maximum speeds of roughly 5–7 km/h, comparable to a brisk human walk.¹⁵ Ornithischian quadrupeds such as ceratopsians and ankylosaurs likely had somewhat greater speed capabilities, as their limb proportions suggest more flexed, less strictly columnar postures. The long-standing question of whether any large dinosaur was capable of sustained running (as opposed to walking with a suspended phase) remains contested, with most current analyses concluding that animals above approximately 5–6 tonnes were restricted to walking gaits.

Synthesis and the emerging picture

The convergence of evidence from bone histology, isotope geochemistry, molecular biomarkers, respiratory anatomy, biomechanics, and reproductive biology paints a coherent portrait of dinosaur physiology that would have astonished Owen, and even surprised the early proponents of the dinosaur renaissance. The ancestral dinosaur was almost certainly an endothermic animal, maintaining a body temperature in the range of 36–38 °C through internal metabolic heat production.^{18, 20} It breathed through parabronchial lungs with unidirectional airflow, an architecture inherited from archosaurian ancestors and supported by an air sac system that left tell-tale pneumatic cavities in the skeleton.^{4, 13} It grew rapidly, reaching adult size in years to a couple of decades depending on final body mass, and its bones recorded this speed in their fibrolamellar microstructure.²

Variation existed within this general framework. The very largest sauropods may have relied partly on gigantothermy to maintain thermal stability, requiring somewhat lower mass-specific metabolic rates than a strict endotherm of equivalent size.¹⁵ Some ornithischians may have had metabolic rates modestly lower than those of theropods, as suggested by both the Grady mesothermy analysis and the Wiemann molecular data, though still well above the ectothermic range.^{16, 20} Incubation biology was slower than that of modern birds, suggesting that the final acceleration of embryonic development occurred only within the avian crown group.¹⁷ And sensory equipment varied dramatically by ecology: tyrannosaurs relied heavily on olfaction, troodontids on vision in low-light conditions, and hadrosaurs on a combination of vision and auditory communication.^{10, 11}

What emerges is not a monolithic "warm-blooded dinosaur" hypothesis but a more nuanced picture of a physiologically diverse clade that shared a common endothermic foundation. The debate initiated by Bakker in 1972 is not over — the precise metabolic rates, thermoregulatory mechanisms, and physiological strategies of individual dinosaur lineages remain active areas of research — but the broad question of whether dinosaurs were metabolically closer to birds or to lizards has been resolved decisively in favour of the former. Dinosaurs were not "terrible lizards" in any physiological sense. They were, in every measurable respect, the active, warm, fast-growing predecessors and relatives of the birds that surround us today.