Toothless Beak

A toothless beak, in paleontological usage, refers to a jaw apparatus in which all teeth have been lost (a condition termed edentulism) and the jawbones are covered by a rhamphotheca—a continuous keratinous sheath composed of β-keratin, the same protein family found in claws, horns, and feathers. In living birds and turtles, the rhamphotheca covers both the rostral (outer, snout-facing) surface and part of the oral surface of the premaxilla, maxilla, and dentary, providing a hard, self-sharpening edge for food processing.

Because the rhamphotheca is composed of organic material, it rarely preserves in the fossil record. Paleontologists therefore rely on osteological correlates on the jawbone surface to infer its presence. These include dense vascular grooves, small foramina arranged in specific patterns, and a rugose or pitted bone texture distinct from the smooth surfaces associated with lips or the linearly arranged foramina associated with labial scales in squamates (Hieronymus et al., 2009; Carr et al., 2017; Cullen et al., 2023). Aguilar-Pedrayes et al. (2024) systematically coded these rostral bone surface textures as proxies for keratin covering across 93 species of dinosaurs in their phylogenetic comparative analysis.

The evolution of toothless beaks was not a single event but a remarkably convergent phenomenon that occurred independently in numerous dinosaur lineages.

Theropoda: Edentulous beaks evolved at least seven times independently within theropod dinosaurs (Wang et al., 2017; Louchart & Viriot, 2011). The major lineages include Oviraptorosauria, where basal forms such as Incisivosaurus and Caudipteryx retained some teeth but derived groups (Caenagnathidae, Oviraptoridae) were fully edentulous with robust, parrot-like beaks; Ornithomimosauria, where derived "ostrich dinosaurs" such as Ornithomimus, Gallimimus, and Struthiomimus had completely toothless jaws, while basal forms like Harpymimus and Pelecanimimus still possessed teeth; Therizinosauria, where the premaxilla was edentulous and covered by a rhamphotheca while the posterior maxilla and dentary retained teeth; Ceratosauria, most notably Limusaurus inextricabilis, which demonstrates ontogenetic edentulism—the gradual loss of teeth during individual growth—providing a living window into the evolutionary mechanism of beak acquisition (Wang et al., 2017); and multiple avian lineages, including the Early Cretaceous Confuciusornis and members of Ornithuromorpha.

Ornithischia: All ornithischians possessed a rostral bone (the predentary) at the mandibular tip that was covered in keratin, forming a beak. However, nearly all ornithischians retained teeth posterior to the beak. Ceratopsians had a sharp, parrot-like rostral beak combined with extensive tooth batteries for shearing. Hadrosaurs had a broad, "duck-bill" beak anteriorly and complex dental batteries posteriorly for grinding plant material. Complete jaw toothlessness was rare in this group.

Sauropodomorpha: Some early sauropodomorphs (e.g., Leyesaurus, Adeopapposaurus) show evidence of small keratinous plates at the rostral tips of the jaws, but no sauropodomorph achieved complete edentulism. Some sauropods like Camarasaurus and Diplodocus had rostral bone textures consistent with limited keratin covering, but teeth were always retained.

The functional significance of replacing teeth with a keratinous beak was rigorously tested by Lautenschlager et al. (2013) using finite element analysis (FEA) on a detailed biomechanical model of the therizinosaur Erlikosaurus andrewsi. By comparing skull models with full dentition, partial dentition, and a complete keratinous rhamphotheca, the researchers demonstrated that the beak configuration significantly reduced stress and strain in the rostral skull during simulated feeding loads. The rhamphotheca effectively dissipated forces across a broader surface area, making the rostral portion of the skull less susceptible to bending and displacement. The authors concluded that keratinous beaks represent an evolutionary innovation that enhanced cranial stability, a benefit distinct from and potentially more fundamental than the traditionally postulated weight-saving advantages associated with flight.

Additional functional advantages of the toothless beak include the following. First, the continuous growth of keratin eliminates the metabolic cost of dental replacement cycles, which in some dinosaurs like hadrosaurs involved maintaining hundreds of replacement teeth simultaneously. Second, keratin is substantially lighter than dentin and enamel, contributing to overall skull mass reduction. Third, the beak's shape can be honed through wear into a self-sharpening edge, enabling efficient cutting, clipping, shelling, and crushing of diverse food items.

A landmark phylogenetic comparative study by Aguilar-Pedrayes et al. (2024), published in Proceedings of the Royal Society B, tested whether the evolution of rostral keratin covering drove tooth loss across Dinosauria. Using a multinomial regression model across 93 dinosaur species, they found that rostral keratin cover is strongly associated with partial toothrow reduction—when keratin was present, the odds of partial tooth loss were approximately 9 times higher in the upper jaw and 31 times higher in the lower jaw compared to full dentition. However, the presence of keratin covering did not significantly predict complete jaw toothlessness in the upper jaw, and did not increase the evolutionary rate of tooth loss.

Critically, the study revealed different evolutionary trajectories across the three major dinosaur lineages. In theropods, partial toothrow reduction preceded the evolution of rostral keratin covering, suggesting that dietary shifts drove initial tooth loss, with keratin beaks evolving subsequently. In ornithischians and sauropodomorphs, by contrast, rostral keratin covers evolved while full or nearly full dentitions were maintained, and some lineages even re-evolved complete toothrows from partial dentition despite having keratin covers. These findings demonstrate that the relationship between beaks and teeth is far more complex than a simple antagonistic replacement.

Wang et al. (2017) provided compelling evidence that the macroevolution of toothless beaks was driven by heterochronic truncation of odontogenesis—the progressive shifting of tooth development cessation to earlier ontogenetic stages. In Limusaurus, hatchlings possess functional teeth that are gradually lost during postnatal growth as the beak expands, representing ontogenetic edentulism. Similar vestigial alveoli (tooth sockets) were identified in caenagnathid oviraptorosaurs and the Early Cretaceous bird Sapeornis, indicating that postnatal tooth loss was a transitional evolutionary stage.

The molecular mechanism linking beak expansion and tooth suppression centers on BMP4 (bone morphogenetic protein 4). In bird embryos, BMP4 regulates the growth and keratinization of the caruncle (egg tooth) and the expanding rhamphotheca. Overexpression of BMP4 produces peramorphic (enlarged) beak phenotypes, while its antagonist noggin (NOG) produces paedomorphic (reduced) beaks. BMP4 overexpression also appears to disrupt the agonist/antagonist balance necessary for normal odontogenesis, effectively suppressing tooth development. This dual role of BMP4—promoting keratin growth while inhibiting tooth formation—provides a mechanistic explanation for the coupling of beak expansion and tooth loss observed in the fossil record.

In modern birds, the key genes required for enamel formation (enamelin, ENAM; ameloblastin, AMBN) have been lost or pseudogenized, and the gene for dentin formation (dentin sialophosphoprotein, DSPP) is inactivated. These genetic losses represent the final, irreversible stage of the evolutionary sequence from toothed jaws to toothless beaks.

Yang & Sander (2018) proposed an additional selective pressure favoring tooth loss: the acceleration of embryonic development. Erickson et al. (2017) used von Ebner lines (daily growth increments in embryonic tooth dentin) to demonstrate that toothed ornithischian dinosaurs—specifically Protoceratops and Hypacrosaurus—had incubation periods of 3 to 6 months, comparable to modern reptiles and far longer than previously assumed based on bird models. Since tooth dentin formation rate is constrained to a maximum daily increment of approximately 30 μm, and teeth do not begin forming until roughly 40% through incubation, tooth development may have imposed a lower limit on incubation duration.

Under this hypothesis, selection for faster incubation—which reduces vulnerability to predation, disease, and environmental catastrophe—would have favored the elimination of embryonic tooth development. This would explain the repeated independent evolution of toothless beaks in theropod lineages more derived than Tyrannosaurus, which are also the lineages that show evidence of open nesting and brooding behavior rather than burying eggs. Although this hypothesis awaits direct experimental testing, it provides an elegant explanation for why edentulism appeared convergently in so many dinosaur groups.

The end-Cretaceous mass extinction 66 million years ago represents a critical juncture in the history of beaks. All surviving birds (Neornithes) are edentulous, while all toothed bird lineages, including Enantiornithes and Hesperornithiformes, went extinct. Larson et al. (2016) suggested that the K-Pg event acted as an ecological filter favoring edentulous birds, potentially because beaked, seed-eating lineages could exploit the limited food resources available in the post-impact environment, whereas specialized toothed birds dependent on animal prey could not.

However, Brocklehurst & Field (2021) found no overarching macroevolutionary trend toward edentulism in Mesozoic birds (Avialae) prior to the extinction event. This indicates that the universal toothlessness of modern birds was not the culmination of a directional evolutionary trend but was instead shaped by the selective bottleneck of mass extinction, combined with the irreversibility of tooth loss (Dollo's Law) once the edentulous condition was achieved.

The transition to a toothless beak appears to be effectively irreversible once complete—a classic example of Dollo's Law. After a lineage achieves full edentulism and the rhamphotheca covers the entire jaw, relaxation of selection on dental genes leads to the accumulation of loss-of-function mutations in genes essential for odontogenesis. In modern birds, the pseudogenization of ENAM, AMBN, and DSPP (Meredith et al., 2014) means that even if selection were to favor teeth, the genetic machinery for producing them no longer exists. Among approximately 10,000 extant bird species, not a single one has re-evolved true teeth, though some species (e.g., the "tooth-billed" bowerbird, certain geese) have evolved beak serrations that are functionally analogous to teeth but are entirely keratinous.

Recent research continues to refine understanding of beak evolution in dinosaurs. A 2025 study (Navalón et al., iScience) demonstrated that beaks evolved at least six times independently in theropods and that beak shapes across these independently evolved lineages follow a common mathematical growth rule (power cascade model), suggesting that developmental constraints channel beak morphology into predictable forms regardless of phylogenetic background.

The Aguilar-Pedrayes et al. (2024) study opened new avenues by showing that the spatial dynamics of keratin cover and tooth distribution differ between the upper and lower jaws, with the upper jaw showing greater evolutionary lability in toothrow variation when a keratin cover is present. Future research directions include expanding phylogenetic comparative analyses to include pterosaurs and crocodylomorphs (other archosaur lineages that independently evolved beaks), applying morphometric and biomechanical methods to analyze the spatial influence of rhamphotheca on toothrow evolution, and using evo-devo experimental approaches to directly test whether tooth loss accelerates incubation periods.