Dental geochemistry reveals thermoregulation in the Neogene ocean’s most infamous superpredator

The extinct megalodon shark, Otodus megalodon, has captured the imagination of the public and paleontologists for over a century. This ancient shark, estimated to be the size of a modern whale shark or larger (~15 m, ref. 1), resembled white sharks—made famous by the movie “Jaws”—and marauded the Miocene seas of North Carolina, California, Japan, and even Malta. This has led to paleoecological questions about the ocean ecosystems that supported such enormous marine carnivores. Research has also sought a definitive answer as to why these iconic sharks went extinct, leaving “The Meg” to exist only on Hollywood movie screens and in the imaginations of everyone from elementary school students and amateur fossil hunters to stratigraphers and museum curators. The remarkably large fossil teeth of O. megalodon (Fig. 1) demand attention, similar to dinosaur skeletons and hominid skulls in their ability to transport us to the mysterious and unfamiliar worlds that Earth once hosted. Using new geochemical measurements of megalodon dental enamel, Griffiths et al. (2) add a new dimension by reconstructing body temperatures of these ancient sharks. Their new data prompt consideration of physiological factors as contributors to the prominence and eventual demise of O. megalodon.

Fig. 1.

A 15-cm O. megalodon tooth. Gray material on the cusp of the tooth shows that the enamel is distinct from the dark gray fossilized dentine underneath. Extant shark teeth, and well-preserved fossil ones, are covered entirely by enamel, but this partially delaminated one illustrates the thin enamel veneer subsampled for isotopic analysis by Griffiths et al. (2). Image of VPPU.011995 Courtesy of the Yale Peabody Museum.

The skeletal system of sharks is primarily composed of cartilage, a flexible yet strong proteinaceous connective tissue. This distinguishes them from bony fish, which are well represented in the fossil record because bone readily fossilizes, whereas cartilage does not. Consequently, shark paleontology relies on teeth (and occasionally vertebrae), which are made of porous dentine and covered by a thin layer of dense enamel. The fluorapatite that forms enamel is resistant to physical and chemical degradation, allowing for the preservation of dental records of shark anatomy and feeding habits. Stable isotopes of carbon found in rare carbonate groups within fluorapatite and, more recently, nitrogen isotopes of intercrystalline organic material in enamel, provide glimpses into the diets of ancient sharks, including megalodon (3). This research has revealed that O. megalodon occupied an extremely high trophic position, surpassing any living shark and rivalling only modern marine carnivores like orca whales and polar bears. It also argues against Late Neogene changes in baleen whales, hypothesized to be a main prey item for megalodon, as causal for their extinction.

With no clear trophic driver identified for megalodon’s disappearance, alternative hypotheses include ecophysiological explanations for their extinction. In today's oceans, various taxonomic groups, including mackerel sharks (Lamnidae) and tunas (Thunnus spp.), are known to exhibit a form of endothermy. Lamnidae, which includes the great white shark, possess a unique arrangement of their slow-twitch red muscles, enabling countercurrent heat exchange, metabolic heat conservation, and elevation of their core body temperatures above the surrounding seawater (4). Numerous evolutionary factors contribute to this convergent evolution phenomenon termed “regional endothermy,” granting laminid sharks and tunas the ability to undertake excursions well below the thermocline and enhancing energy efficiency during burst swimming speeds (5). Although marine biologists employ comparative anatomy and electronic tagging to investigate the consequences of regional endothermic adaptations, these approaches lack a historical perspective. Therefore, the availability of a paleontological record of shark thermoregulation would be highly beneficial. Development of such a record has been challenging because of the limitations of existing geochemical thermometry methods applied to fossil shark teeth, but such barriers have now been overcome by Griffiths et al. (2)

One widely applied paleotemperature method involves measuring the stable isotopes of oxygen (e.g., ¹⁸O vs. ¹⁶O) within mineralized hard parts like teeth. Oxygen isotope paleothermometry, which examines the oxygen isotopes in carbonate (CO₃²⁻) or phosphate (PO₄⁻) bound within fossil teeth, relies on the temperature-dependent fractionation of ¹⁸O into tooth enamel. Modern oxygen isotope ratio mass spectrometry can provide temperature estimates with a precision of ±1 °C or better. However, to obtain an accurate value, one must also know the oxygen stable isotope ratio (δ¹⁸O) of the water in which the shark inhabited. In the ocean, there is substantial latitudinal variation in δ¹⁸O (6), which is problematic considering the long migrations characteristic of laminid sharks. Additionally, decades of paleoclimate research have revealed changes in the average ocean δ¹⁸O value over evolutionary timescales (ranging from millions to tens of millions of years) as well as during glacial-interglacial shifts in Earth's climate (spanning tens to hundreds of thousands of years). This variability has limited the application of oxygen isotope thermometry for deriving ancient shark body temperatures (7).

By utilizing new geochemical measurements of megalodon dental enamel, Griffiths et al. contribute by reconstructing body temperatures of these ancient sharks.

A solution to this problem emerged with the advent of a stable isotope thermometry method reliant on so-called “clumped isotopes.” This method offered a temperature-only isotope fractionation signal found in paired arrangements of the rare, heavy stable isotope of carbon, ¹³C, and ¹⁸O in CO₃²⁻ (8). And better still, this CO₃²⁻ could be harvested for clumped isotope analysis from the same samples previously targeted for δ¹⁸O. In a collaborative effort centered around clumped isotope paleothermometry, Griffiths et al. (2) conducted a careful subsampling of megalodon teeth enamel, along with laminid and nonlaminid fossil shark teeth obtained from five globally distributed locations. They were in search of elevated temperatures indicative of regional endothermy. Additionally, at some of these sites, samples were taken from fossil mollusks, which lack the ability to regulate body temperature, for baseline temperature comparisons.

The new isotope results are clear and unequivocal: The clumped isotope temperatures (T(Δ₄₇)) derived from O. megalodon (and its related species O. chubutensis) teeth are significantly warmer than those of ectothermic taxa, including nonendothermic shark teeth, and exceed T(Δ₄₇) of most regional endotherms and contemporaneous white shark teeth. Conversely, Otodus T(Δ₄₇) values are cooler than co-occurring whales, whose ear bones, curiously, also provided mineralized material for clumped isotope analysis. Griffiths et al. (2) qualify these findings by noting that fossil shark tooth T(Δ₄₇) is unlikely to match the maximum internal body temperature sustained by regional endothermy. In modern white sharks, temperature gradients of up to 13 °C exist from the core to the epidermis in contact with seawater, predominantly determined by the ambient water temperature (9). It is uncertain how the new Otodus tooth results correlate with similar temperature gradients, although this may be addressed by modeling and further morphometric analysis. Nonetheless, this finding solidifies the presence of endothermy within Lamnidae.

An obvious outcome of this study is the need for more shark tooth T(Δ₄₇) data. Expanding datasets to include a variety of fossil Galean sharks could shed light on the prevalence of endothermy within this superorder. Clumped isotope measurements, due to the rarity of double heavy isotope substitution, are known to require large sample sizes relative to thin enamel. However, recent innovations that have reduced sample requirements and increased throughput may prove fruitful in this regard (10). Griffiths et al. (2) also recognized the advantages of δ¹⁸O analysis in relation to the potential of clumped isotope thermometry. In parallel with T(Δ₄₇) analysis, they used a Bayesian framework, with priors for temperature and seawater δ¹⁸O, to derive δ¹⁸O-based shark tooth paleotemperatures. This clever statistical approach offers a workaround for the limitations of δ¹⁸O thermometry, and perceptive readers may appreciate its potential for reevaluating other applications of marine δ¹⁸O thermometry (e.g., ref. 11). Both the Bayesian temperatures and T(Δ₄₇) converge, highlighting the same temperature differences between Otodus teeth and coexisting fossil sharks (as well as all extant sharks). It appears that megalodon had the warmest body temperatures of all sharks, a subtle observation that Griffiths et al. (2) attribute to its exceptionally large body mass.

It seem likely that both benefits and costs were associated with regional endothermy in megalodon. A perceived benefit comes from modern whale sharks. Unlike megalodon, the size-equivalent whale shark is an ectotherm. Their planktivorous lifestyle also differs from the carnivorous laminids, and in order to reduce energetic demands, their average swimming speed is many times slower (1). Also, whale shark gigantism and its accompanying large thermal inertia result in slower fluctuations of body temperature compared to the more dynamic ambient environment in which it swims (12). Megalodons also likely benefited from similar metabolic stability in addition to their physiological endothermy. Any advantages conferred by their large size and homeothermy must have been balanced by increased energetic demands. This relationship is a double-edged sword as it may have enabled but also necessitated larger ranges to enhance the chances of encountering prey. This is true of modern fin whales which, when combined with the habitual depth migrations of extant laminids (13), suggest that megalodon may have been a transoceanic presence in the Miocene and Pliocene.

The results of Griffith et al. represent a significant milestone in unveiling the life history and ecophysiology of megalodon. However, an intuitive, yet unresolved question remains: How was ancient ocean paleoproductivity elevated enough to support the undisputed apex marine superpredator of the Late Neogene? Sustaining a large population of top-level marine consumers like megalodon (3) would have required a higher baseline of marine primary productivity, modulated by the efficiency of assimilation (digestion) and transfer (trophic consumption) up the food chain (14). There is paleoceanographic evidence of increased primary production during the Miocene and early Pliocene, which ended approximately 3.5 to 4.5 Mya (15). Some records suggest a peak in productivity during the late Miocene that was up to twice as high as present-day ocean productivity (16). A return to modern-like proxy values coincides with the estimated median extinction age of O. megalodon at ~3.5 Mya (17). Could the changing early Pliocene paleoclimate simply have rendered the conditions in global oceans unsuitable for sustaining the large food webs required by megalodon? If so, the complete story of megalodon, including its physiology, trophic ecology, and demise, may hold valuable lessons for ongoing shark conservation as the modern ocean warms (18). Nevertheless, there is a real juxtaposition between the size and fragility of such a high-level marine predator, whose very existence reflects a zenith of Neogene global ocean productivity. And for those captivated by the sight of impressively large megalodon teeth in a museum display case, Griffith et al. (2) once again demonstrate that there is more to the stories they tell than what meets the eye.