Reptile Phylogeny: The Family Tree of Reptiles and Birds

Birds are not just relatives of reptiles; in evolutionary terms, they are reptiles. The family tree that connects lizards, snakes, turtles, crocodilians, and birds is a single branching structure rooted in animals that lived over 300 million years ago, and modern genetic and fossil evidence has made the boundaries of that tree far clearer than they were even two decades ago. The result is a picture that overturns old classroom charts and reveals some genuinely surprising relationships, including one involving turtles that caught even specialists off guard.

Where the Tree Begins

The story starts with amniotes, the group of vertebrates whose embryos develop inside a protective membrane (the amnion) rather than depending on open water. Amniotes first appeared roughly 320 million years ago, and their emergence was quickly followed by a deep fork: one branch led toward mammals (the synapsid lineage) and the other toward reptiles (the reptile or sauropsid lineage).¹ Every living reptile and every bird descends from that second branch.

Within the reptile branch, the most important early development was the rise of diapsids, animals whose skulls had two openings behind the eye socket. During the Permian period, diapsids underwent a burst of evolutionary change, with body shapes, skull structures, and ecological roles diversifying at rates well above average for the lineage.² That Permian radiation seeded the two great lineages that would dominate reptile evolution from then on.

Two Great Branches

Almost every reptile and bird alive today belongs to one of two major lineages that diverged early in diapsid history. One is Lepidosauromorpha, the branch that gave rise to lizards, snakes, and the tuatara. The other is Archosauromorpha, which produced crocodilians, pterosaurs, dinosaurs, and birds. Understanding which branch a given animal sits on is the single most useful thing for reading the reptile family tree.

Lepidosaurs split internally into two groups: Rhynchocephalia, represented today only by the tuatara of New Zealand, and Squamata, the spectacularly diverse group containing all lizards and snakes. Fossil evidence now places that split at roughly 245 to 241 million years ago, in the Middle Triassic.³ Rhynchocephalians were once widespread and varied, but today the tuatara is their sole survivor, making it one of the most evolutionarily isolated reptiles on Earth.

On the archosaur side, the lineage divided into two paths of its own. One led to crocodilians and their many extinct relatives. The other, called Avemetatarsalia or the bird-line archosaurs, led to dinosaurs, pterosaurs, and eventually birds. An early member of this bird-line branch, a Triassic animal called Teleocrater, looked nothing like a bird: it was a four-legged, roughly crocodile-shaped predator. Its anatomy shows that the bird-line body plan was assembled gradually, starting from very un-birdlike ancestors.⁴

Where Turtles Actually Belong

For most of the twentieth century, turtles were considered primitive reptiles, possibly not even diapsids, placed at the base of the reptile tree or grouped with lizards and the tuatara. Genomic data demolished that view. Multiple independent studies using large datasets of nuclear genes have converged on the same conclusion: turtles are the sister group of archosaurs, meaning their closest living relatives are birds and crocodilians, not lizards.⁵⁶

An analysis of over a thousand ultraconserved DNA elements from across the genomes of all major reptile groups, including the tuatara, provided what researchers described as overwhelming support for this placement, firmly rejecting the old hypothesis that turtles were close to lepidosaurs.⁷ The reason earlier morphological studies got it wrong is partly that turtles lost the two skull openings characteristic of diapsids, making their skulls look superficially “primitive.” Genetics told a different story: turtles are modified diapsids whose skull openings closed up over evolutionary time.

This reshuffling has real consequences for how biologists interpret shared traits. Turtles and crocodilians share a class of structural skin proteins (corneous beta-proteins) that form a separate evolutionary cluster from the equivalent proteins in lizards and snakes, consistent with the idea that the two groups diverged from each other back in the Permian.⁸ Features like the turtle shell and the bird feather turn out to be elaborations on shared molecular building blocks that their common ancestor already possessed.

Squamates and the Many Reinventions of the Snake Body

Squamata is the largest order of living reptiles by a wide margin, containing over 11,000 species of lizards, snakes, and amphisbaenians (worm-lizards). They are also the group that has most enthusiastically experimented with body shape. The elongated, limbless body plan we associate with snakes has evolved independently dozens of times among squamates, in lineages as different as glass lizards, skinks, and legless geckos.⁹ Squamates are also the only vertebrate group with true parthenogenesis (reproduction without males), having evolved it at least 40 separate times, and they hold the record for the most independent origins of live birth among vertebrates.¹⁰

The origin of snakes themselves has been a long-running debate. One camp proposed that snakes descended from marine lizards related to mosasaurs, losing their limbs while swimming. The competing hypothesis argued for a terrestrial, burrowing origin. Molecular evidence weighs against a close relationship between snakes and monitor lizards (the living relatives supposedly linking snakes to mosasaurs), undercutting the marine scenario.¹¹ Meanwhile, CT scans of the inner ear of Dinilysia patagonica, a fossil snake closely related to the ancestors of modern snakes, revealed a spherical vestibule shared specifically with living burrowing squamates. Predictive models built on vestibular shape estimated both Dinilysia and the hypothetical ancestor of modern snakes as burrowers with high probability.¹²

The genomic basis of limb loss is itself a rich story. Researchers comparing the genomes of limbless reptile lineages have found both shared and lineage-specific changes in the regulatory regions that control limb development, suggesting that different groups arrived at similar body plans through partly different genetic routes.¹³ Limb loss in snakes, for example, involves changes not just to the limbs themselves but to the developmental program that elongated the trunk, adding hundreds of vertebrae.¹⁴

From Dinosaurs to Birds

On the archosaur side of the tree, the path to birds runs through theropod dinosaurs: the bipedal, often carnivorous group that includes Tyrannosaurus and Velociraptor. Fossil discoveries over the past few decades have made the case overwhelming that birds descended from maniraptoran theropods, a subgroup within the theropods. Feathers, a wishbone, hollow bones, rapid growth rates, and a specialized respiratory system all appeared in various theropod lineages before birds existed, meaning these traits were not invented for flight but were repurposed for it.¹⁵ The classic bird body plan was assembled piece by piece over tens of millions of years during the Jurassic, roughly 165 to 150 million years ago.¹⁶

Pterosaurs, the flying reptiles of the Mesozoic, were not on the line leading to birds, but they were closer relatives of dinosaurs than many people assume. Recent analyses of well-preserved skulls and braincases show that lagerpetids, a group of small, cursorial, non-flying animals previously classified as dinosaur precursors, are actually the closest known relatives of pterosaurs.¹⁷ Both pterosaurs and lagerpetids sit on the bird-line side of the archosaur tree, in a close relationship with dinosauriforms.¹⁸ So while pterosaurs and birds evolved flight independently, they were part of the same broad evolutionary neighborhood.

The Post-Extinction Burst of Bird Diversity

The asteroid impact at the end of the Cretaceous, 66 million years ago, wiped out all non-avian dinosaurs and pterosaurs. But some lineages of birds survived. The deepest splits within living birds likely occurred just before the extinction: the divide between Palaeognathae (ostriches, emus, kiwis) and Neognathae (everything else), and within Neognathae, the split between Galloanserae (chickens and ducks) and Neoaves (the vast remainder).¹⁹ The earliest well-supported crown-bird fossil is barely older than the extinction event, at roughly 67 million years.

After the extinction, birds radiated rapidly. An initial burst produced the basal orders of Neoaves, from grebes to hummingbirds. Two subsequent waves gave rise to “core waterbirds” (penguins, pelicans, loons) and “core landbirds” (hawks, woodpeckers, parrots, and the enormous clade of songbirds).²⁰ The more than 10,000 bird species alive today are the product of those cascading radiations, each filling ecological niches left empty by the mass extinction.

Crocodilians, interestingly, did not experience the same kind of explosive diversification after the extinction. Their fossil record shows that several lineages crossed the Cretaceous boundary, including marine dyrosaurids and terrestrial sebecids, but no major diversification event followed. The decline of Mesozoic crocodyliform lineages was spread over a long period, with some disappearing well before the asteroid and others persisting long after.²¹ The roughly two dozen species of living crocodilians are a thin remnant of a once far more diverse group.

What the Tree Reveals About Shared Biology

Reading the family tree correctly changes how you interpret shared traits across reptiles and birds. Take hearts. Most living reptiles have a three-chambered heart with an incompletely divided ventricle. Crocodilians and birds, however, have a fully four-chambered heart, with a complete wall (septum) separating the left and right ventricles, much like mammals.²² Researchers studying the gene Tbx5, which patterns the ventricles during embryonic development, found that in turtles (close relatives of archosaurs, as we now know), Tbx5 expression starts out uniform across the ventricle and then gradually concentrates in the left side during development, forming a gradient. In birds and mammals, that gradient is steep and established early, producing a complete septum. In lizards, the gradient never develops. Experimentally disrupting the Tbx5 gradient in mice produces a single, undivided ventricle resembling the reptilian condition.²³ The turtle heart, in other words, looks like a snapshot of the evolutionary transition from three chambers to four.

Skin coverings tell a parallel story. Scales, feathers, claws, beaks, and turtle scutes are all built from beta-keratins (more precisely called corneous beta-proteins), a class of structural proteins found only in reptiles and birds. These proteins evolved within a genomic region called the epidermal differentiation complex, and their gene families have expanded and specialized along different branches of the tree. In birds, feather beta-keratins make up the largest proportion of the family, but birds of prey have proportionally more claw beta-keratin genes, and aquatic birds have expanded a different subfamily, reflecting adaptations to different lifestyles.²⁴ Feathers likely evolved independently of the overlapping scales seen in lizards, each originating from an ancestral rough, beta-keratinized skin.²⁵

Color Vision Across the Tree

One of the more striking examples of how phylogeny shapes biology is color vision. The ancestral vertebrate had five visual pigments: four types of cone pigments for color vision in bright light and one rod pigment for dim light. Where a given reptile group sits on the family tree predicts, with surprising accuracy, how many of those pigments it retained.

Lizards generally still express all five ancestral pigments. Some turtles do as well, giving them the potential for tetrachromatic vision: they can distinguish four independent channels of color, compared to the three most humans use. Birds also have four cone types and are well-known tetrachromats.²⁶ The spectral sensitivities of cones in birds, lizards, and turtles are remarkably similar despite hundreds of millions of years of independent evolution, suggesting that the tetrachromatic retina sits at a stable adaptive optimum: once you have it, there is little pressure to change it.²⁷

Snakes are the exception that proves the rule. They lost two of the five ancestral pigments, probably during the dim-light, possibly burrowing phase of early snake evolution that inner-ear evidence also points to. Some snake lineages later reinvaded bright, open habitats and partially compensated by tuning the pigments they retained: mutations at key spectral-tuning sites in the rod pigment gene, for example, have arisen independently in multiple snake families.²⁸ Crocodilians lost two pigments as well, though not the same two that snakes did, reflecting a separate reduction event on a different branch of the tree.²⁶

How Continents Sculpted Distribution

A family tree tells you who is related to whom, but geography tells you why certain relatives ended up on opposite sides of the planet. Continental drift has been a major sculptor of reptile and bird biogeography. Analyses of turtle phylogeny matched against plate-tectonic timelines find multiple well-supported range shifts and vicariance events (populations split by physical barriers like opening oceans) that line up with episodes of continental fragmentation.²⁹

The same pattern holds for dinosaurs. Statistical analyses of dinosaur distributions from the Middle Jurassic through the mid-Cretaceous reveal biogeographic patterns that closely track the breakup of Pangaea, providing robust evidence that tectonic events determined where and when particular dinosaur groups flourished.³⁰ When South America separated from Africa, for instance, the dinosaur faunas on each continent began evolving independently. The same separations shaped the distribution of crocodyliform, lizard, and turtle lineages, which is why closely related species sometimes turn up on continents that are now thousands of miles apart.

Reproductive Strategies and Their Evolutionary Flexibility

Reproduction is another domain where the reptile family tree reveals unexpected flexibility. Most reptiles lay eggs, but live birth has evolved independently over a hundred times among squamates alone.¹⁰ That transition may have been smoothed by a pre-existing behavioral trait. Research on gravid female lizards and snakes shows that they adjust their body temperatures to better match the thermal optimum for their developing embryos, cooling down if their normal preferred temperature is too warm and warming up if it is too cool. This behavioral thermoregulation occurs in both egg-laying and live-bearing species, and ancestral reconstructions suggest it was present before live birth evolved, effectively pre-adapting certain lineages for the shift to retaining eggs internally.³¹

Birds and crocodilians, by contrast, are locked into egg-laying. No bird or crocodilian has ever evolved live birth, which may reflect constraints imposed by their rigid-shelled eggs and the physiology of flight (in birds) or the deeply conserved developmental program of archosaurs. The contrast with squamates, where the transition has happened again and again, underlines how different branches of the same family tree can have very different evolutionary degrees of freedom for a given trait.