Saber-toothed cats, especially the best-studied species Smilodon fatalis, were ambush predators that fed on large herbivores and killed with a distinctive slashing bite powered more by neck muscles than jaw strength. Their jaw muscles produced only about a third the bite force of a similarly sized lion, which sounds like a disadvantage until you realize the entire skull was engineered for a different kind of attack. The way these animals brought down prey was so unlike anything alive today that researchers spent decades arguing about it, and new findings are still reshaping the picture.
What Tooth Wear Tells Us About Their Diet
The most direct evidence of what saber-toothed cats ate comes from microscopic wear patterns on their teeth. Scanning electron microscopy of the lower carnassial (the main shearing tooth) in Smilodon fatalis reveals narrow, elongated scratch marks and an extremely low frequency of pits compared to modern large carnivores. That pattern is a strong indicator that Smilodon consumed very little bone. Even cheetahs, which are considered light feeders among big cats, show more pitting on their teeth than Smilodon does.1Journal of Zoology. Molar microwear and diet in large carnivores: inferences concerning diet in the sabretooth cat, Smilodon fatalis
The likely explanation is that Smilodon deliberately avoided crunching bones to protect its canines. Those elongated upper canines were impressive weapons, but they were also structurally vulnerable. Biting into a thick leg bone could snap or fracture them, which would have been a death sentence for an animal that depended on those teeth to kill. So Smilodon probably targeted soft tissue, slicing into the throat or belly of prey, and left behind a carcass with plenty of meat and bone still attached. That leftover material would have been a windfall for scavengers.
More recent analysis using dental microwear texture analysis largely confirmed this picture but added nuance. When researchers compared Smilodon’s tooth surface complexity and texture fill volume against modern lions, cheetahs, and spotted hyenas, Smilodon fell between the cheetah and the lion. Its teeth showed more wear than a cheetah’s but less than a lion’s or hyena’s, suggesting it consumed somewhat tougher food than a pure soft-tissue specialist but still stayed well away from the bone-crushing end of the spectrum.2PLoS ONE. Implications of Diet for the Extinction of Saber-Toothed Cats and American Lions
Interestingly, the South American species Smilodon populator, which was larger than its North American cousin, tells a somewhat different story. Researchers studying tooth mark patterns on megaherbivore bones from the Pleistocene of South America found that S. populator canines actually penetrated and processed cortical bone, pushing back against the assumption that all sabertooth canines were functionally fragile.3PubMed Central. Smilodon sabertooth bite marks, new perspectives on prey-predator models in South American Pleistocene This suggests that different species within the same genus may have had meaningfully different feeding behaviors, or that individual animals occasionally used their canines more aggressively than the “avoid bone at all costs” model implies.
A Weak Jaw Backed by a Powerful Neck
The single most surprising thing about Smilodon’s skull is how feeble its jaw muscles were relative to its size. High-resolution computer simulations of the skull showed that the bite force generated by jaw muscles alone was roughly a third of what a lion of comparable size could produce.4PubMed Central. Supermodeled sabercat, predatory behavior in Smilodon fatalis revealed by high-resolution 3D computer simulation That finding initially puzzled researchers, because it seems counterintuitive for a predator that hunted animals much larger than itself.
The resolution came from looking at the neck. Smilodon had well-developed mastoid processes, bony attachment points for neck muscles, positioned unusually far forward on the skull. This anatomy gave the animal tremendous leverage for driving its head downward. The prevailing model, known as the canine-shear bite hypothesis, proposes that Smilodon used both jaw-closing muscles and powerful neck flexors together to drive the upper canines into prey. The jaw held the mouth open and positioned the canines; the neck muscles provided much of the force to sink them in.5PLoS ONE. Comparative Biomechanical Modeling of Metatherian and Placental Saber-Tooths: A Different Kind of Bite for an Extreme Pouched Predator
Recent anatomical bite force estimates put more precise numbers on this. At the canine tip, Smilodon produced about 1,284 newtons of force from jaw muscles alone, and about 4,671 newtons at the carnassial. Those figures are comparable not to the largest modern cats but to a jaguar, which is a much smaller animal.6PubMed Central. Myological and osteological approaches to gape and bite force reconstruction in Smilodon fatalis The implication is clear: without the neck-driven component of the bite, Smilodon’s jaw alone could not have dispatched large prey. The canine-shear bite was a whole-body effort.
An alternative model pushes this even further. Some researchers have proposed a Class 1 lever mechanism in which the forelimbs pressed down on the prey while the body hunched forward, effectively using the prey’s body as a fulcrum. In this scenario, the neck muscles would flex the body upward toward the head rather than nodding the head downward. The mandible would be braced against the prey, and the upper canines would be driven in by the combined action of forelimb and trunk muscles.7PLoS ONE. Jaw Function in Smilodon fatalis: A Reevaluation of the Canine Shear-Bite and a Proposal for a New Forelimb-Powered Class 1 Lever Model Whether the standard canine-shear bite or this lever model is closer to reality, both agree on the key point: Smilodon’s killing bite depended on muscles far beyond the jaw.
Why the Prey Had to Be Pinned Down First
One consistent finding across biomechanical studies is that Smilodon’s skull was poorly suited to handle the unpredictable forces generated by struggling prey. The same computer simulation that revealed the weak jaw bite also showed that the skull experienced high stress when lateral or twisting loads were applied, the kind of forces you’d expect if a bison or horse were thrashing its head while being bitten.4PubMed Central. Supermodeled sabercat, predatory behavior in Smilodon fatalis revealed by high-resolution 3D computer simulation A lion can afford to clamp onto a struggling animal’s muzzle and wait it out. Smilodon could not.
This tells us something fundamental about the hunting sequence. Smilodon almost certainly brought its prey to the ground and restrained it before attempting the killing bite. The forelimbs were heavily muscled and the claws were retractable, consistent with grappling. Think of it as a two-phase attack: wrestle the animal down and hold it still, then deliver a precise slash to the throat or belly. The loads at the lower canine were better constrained in the front-to-back plane than in modern cats with conical teeth, suggesting the bite was delivered with precision against an immobilized target.8Zoological Journal of the Linnean Society. Feeding behaviour and bite force of sabretoothed predators
Smilodon’s gape adds another dimension to this picture. While the angular gape, the maximum angle the jaw could open, was significantly larger than in modern cats, the actual linear distance between the canine tips when the mouth was fully open was not dramatically different from living felids.6PubMed Central. Myological and osteological approaches to gape and bite force reconstruction in Smilodon fatalis In other words, Smilodon’s jaw opened to an unusually wide angle to accommodate its long canines, but it didn’t have a freakishly large gap between tooth tips. It needed that wide angle just to clear the canines past each other and get the mouth around prey.
More strikingly, in Smilodon the upper canines extended well beyond the animal’s ability to gape around them. Gape and canine clearance, which are tightly coupled in most sabertooths, appear decoupled in Smilodon and its close relative Megantereon. This means the canines were longer than what could be fully deployed in a simple point-to-point stabbing bite, which is itself strong evidence for the shear-bite mechanism, where the canines were dragged through tissue rather than simply plunged in.9PLoS ONE. Sabretoothed Carnivores and the Killing of Large Prey
Growing Into Those Fangs
Smilodon’s enormous upper canines didn’t appear overnight. Recent research on the timing of skull development shows the adult canines took years to fully erupt. During this prolonged eruption period, the animal carried both its deciduous (baby) canines and the partially emerged adult canines simultaneously for more than half of the eruption process. This dual-canine arrangement effectively widened the functional tooth profile and brought the bending strength and stiffness of the canine apparatus closer to what the animal would need for effective hunting.10PubMed Central. Bending performance changes during prolonged canine eruption in saber-toothed carnivores: A case study of Smilodon fatalis
This has real implications for how young Smilodon survived. A juvenile with half-erupted canines was not yet equipped for the full canine-shear bite that adults relied on. As the canine emerged further, its cross-sectional shape was optimized for increased bending strength, but its stiffness against lateral deflection actually worsened. Constraints from neighboring sensory structures in the upper jaw limited how thick the root could grow, creating a structural trade-off that only resolved once the canine was fully erupted and the deciduous tooth was gone. During the vulnerable transition period, young animals would have depended heavily on parental support for food.
The Sociality Debate
Whether Smilodon hunted in groups has been one of the more heated questions in paleontology. The case for social behavior rests partly on the Rancho La Brea tar pits in Los Angeles, where Smilodon fossils are astonishingly abundant. Some specimens show partially healed injuries, including bone lesions that would have impaired hunting, yet the animals survived long enough for the bones to begin healing. The argument goes that a solitary predator with a serious injury couldn’t have fed itself and would have starved, so other members of a group must have helped.
That logic has been challenged. A detailed rebuttal pointed out that the assumption, that injured animals would inevitably die without help, is not well supported. Solitary predators in the modern world sometimes survive significant injuries by scavenging, resting, or simply being tough enough to hunt at reduced efficiency.11Brain, Behavior and Evolution. Assessing Behavior in Extinct Animals: Was Smilodon Social? The healed injuries alone don’t prove sociality.
More recently, other lines of evidence have kept the question alive. Fossils from Ecuador suggest that subadult Smilodon stayed with their mothers for extended periods, which is common behavior in large cats today. Researchers studying joint defects in Smilodon bones, similar to a condition called osteochondrosis found in modern snow leopards, noted that most defects were small enough to heal on their own, but the survival of animals with these lesions would have been supported by maternal hunting and protection.12PLOS ONE. Subchondral defects resembling osteochondrosis dissecans in joint surfaces of the extinct saber-toothed cat Smilodon fatalis and dire wolf Aenocyon dirus Extended maternal care isn’t the same as cooperative group hunting, though, and the question of whether adults hunted together as prides do in lions remains genuinely unresolved.
Sexual dimorphism offers another indirect clue. Analysis of skull measurements in Smilodon shows low to moderate size differences between males and females, comparable to the range seen in modern big cats. Some cranial proportions show higher dimorphism, particularly the relative sizes of certain premolars and the height of the mastoid process.13PLoS ONE. Variation in Craniomandibular Morphology and Sexual Dimorphism in Pantherines and the Sabercat Smilodon fatalis In modern cats, the degree of sexual dimorphism loosely correlates with social structure, with more dimorphic species tending toward polygynous mating and sometimes group living. But the correlation isn’t tight enough to draw firm conclusions for an extinct species.
Not All Sabertooths Hunted the Same Way
A common mistake is to treat “saber-toothed cat” as if it describes a single animal. The sabertooth body plan evolved independently multiple times across different mammalian lineages, and even within the cat family, there was real diversity in hunting strategies. The two broad categories are dirk-toothed sabertooths like Smilodon, which had long, relatively narrow canines, and scimitar-toothed sabertooths like Homotherium serum, which had shorter, broader, more coarsely serrated canines.
CT scanning of skulls from both types reveals fundamentally different internal architectures. Smilodon had an extremely thick skull with relatively little trabecular (spongy) bone inside, consistent with a skull built to absorb the concentrated stress of a stabbing canine-shear bite. Homotherium, by contrast, had more trabecular bone and a skull structure more consistent with a clamp-and-hold killing technique, something more like what a lion does when it suffocates prey by clamping down on the muzzle or throat.14Current Biology. Distinct Predatory Behaviors in Scimitar- and Dirk-Toothed Sabertooth Cats These two species lived alongside each other in Late Pleistocene North America, and their different killing methods allowed them to partition the predator guild rather than directly competing for the same prey in the same way.
The convergent evolution of the sabertooth form is itself remarkable. The elongated canine and wide-gape skull architecture appeared not only in multiple lineages of true cats but also in nimravids (sometimes called “false sabertooths”), which split from the cat lineage tens of millions of years ago, and even in marsupial predators like Thylacosmilus from South America. The fact that natural selection independently arrived at the same basic design in such distant lineages speaks to how effective the sabertooth approach was for killing large prey in the ecosystems where it appeared.
Sharing the Landscape with Dire Wolves and American Lions
Smilodon didn’t hunt in a vacuum. Late Pleistocene North America hosted a crowded predator guild that included dire wolves, American lions, pumas, and coyotes. How all these carnivores divided up the available prey without driving each other to extinction is an active area of research. Stable carbon isotope analysis of tooth enamel from Rancho La Brea specimens has been used to reconstruct the diets of each species, and some studies have argued that there was minimal food competition between Smilodon and dire wolves specifically, because the isotope signatures are distinct enough to suggest different prey bases or habitat preferences.
That conclusion isn’t universally accepted. Other researchers analyzing stable carbon isotopes from bone collagen, a different tissue that records dietary information over a different time window, found extensive dietary overlap among the large carnivores in these deposits.15Current Biology. Dietary ecology of Rancho La Brea carnivores The disagreement partly reflects the different information captured by enamel versus collagen, and partly reflects genuinely different interpretive frameworks. What’s clear is that the picture of neat, tidy niche partitioning is probably too simple. These animals likely overlapped in diet to some degree, with competition fluctuating depending on prey availability and season.
The dental microwear evidence offers a complementary perspective. The American lion (Panthera atrox) showed tooth wear similar to the cheetah and distinct from Smilodon, with both species showing significantly lower texture fill volume than lions or hyenas. This suggests the American lion was also a relatively delicate feeder that avoided heavy bone processing, which would have put it in more direct competition with Smilodon for the same soft-tissue-rich carcass resources than either was competing with the more omnivorous dire wolf.2PLoS ONE. Implications of Diet for the Extinction of Saber-Toothed Cats and American Lions
What a Living Sabertooth Looked Like
Museum reconstructions tend to show saber-toothed cats with their canines dramatically exposed, and recent anatomical work suggests this is probably accurate, at least for the more derived species. In Megantereon, a predecessor of Smilodon, the narrow chin and protruding canine tips made it unlikely that soft tissue sheaths could have covered the teeth without dangling loosely and risking puncture. The lips instead fitted between the upper canines and the lower jaw, leaving the canine crowns largely exposed. Smilodon, with even longer canines relative to its lower jaw flanges, would have had even more tooth showing in life. These cats really did walk around with visible fangs.
Brain anatomy offers glimpses into how sabertooths perceived the world. An endocast study of Megantereon cultridens, a close relative in the sabertooth lineage, found an expanded occipital cortex, the brain region associated with visual processing, alongside a reduced auditory cortex. The cerebellum was also expanded, which researchers interpreted as consistent with life in closed, forested environments where balance and spatial coordination mattered more than long-distance hearing.16PubMed. In the brain of a sabertooth: First neuroanatomical description and ethological insights on Megantereon cultridens (Felidae, Machairodontinae) If Smilodon’s sensory profile was similar, and its close evolutionary relationship with Megantereon makes that plausible, then these were primarily visual hunters. An ambush predator in partially wooded terrain, stalking by sight, wrestling prey to the ground, and killing with a precisely aimed slash, fits neatly with everything the biomechanics tell us about how the body worked.