Strange Holes in Ancient Sloth Bones May Rewrite Human History

5 min readResearchers found human-modified giant sloth bones dating to 25,000–27,000 years ago in Brazil, potentially pushing back South America’s earliest human presence by 12,000 years.Scientists used advanced microscopy and scanning techniques to argue that perforations in sloth bones were deliberately drilled by humans and worn as pendants.Genetic evidence from Indigenous American populations suggests humans first entered South America only 16,000–23,000 years ago.Figuring out when exactly humans first reached South America increasingly depends on how you—read the bones of a giant ground sloth?The most provocative candidate for the continent’s oldest evidence of human activity comes from the Santa Elina rock shelter in Mato Grosso, central Brazil. There, researchers reported the presence of human-modified giant sloth bones (osteoderms) found alongside stone artifacts in a layer dated to roughly 25,000–27,000 years ago.“The three giant sloth osteoderms were intentionally modified into artifacts before fossilization of the bones,” researchers wrote in a 2023 study in Proceedings of The Royal Society B. “This provides additional evidence for the contemporaneity of humans and megafauna, and for the human manufacturing of personal artefacts on bone remains of ground sloths.”The bones, from the giant ground sloth Glossotherium phoenesis, not only carry perforations that the team argues could only have been drilled by people, but those holes display evidence of smoothing consistent with the objects having been strung and worn as pendants. The holes preserved stone-tool striations, scraping marks, and polishing, and comparison with bite marks let the team rule out both rodent gnawing and passage through a digestive tract. Establishing that knowledge required painstaking analysis using advanced microscopy and optical scanning techniques.“These observations show that these three osteoderms were modified by humans into artefacts,” the team wrote, “probably personal ornaments.” Mirian Pacheco, a study author and paleontologist at the Federal University of São Carlos, told Live Science that while the meaning of the objects to the people of Santa Elina can’t be recovered, the physical signs point to use as pendants.Thaís Pansani—then a doctoral researcher at the same university and later based at the Smithsonian Institution—said the findings reinforce dating arguments made at Santa Elina for decades. “We believe that there should be more evidence waiting to be found in the rock shelters and caves of Brazil, in places little or not explored.” Her follow-up work, which included radiocarbon dating and isotopic analysis of the site’s G. phoenesis remains published in the Journal of Quaternary Science in 2024, and ongoing analysis of whether sloth bones were charred at human-made fire temperatures rather than in wildfires, is aimed at strengthening the case that fresh sloth carcasses were present at human campsites.The long-standing anchor for human presence in South America has been the Monte Verde II site in southern Chile, which was radiocarbon dated to about 14,500 years ago and validated by a panel of outside experts in 1997. Santa Elina’s oldest layer would therefore be some 12,000 years older than the continent’s only widely accepted Late Pleistocene site. It would also be older than the human footprints at White Sands, New Mexico, which were dated to roughly 21,000–23,000 years ago and independently corroborated in 2023 and again in 2025 by radiocarbon dating of seeds, pollen, and mud (plus optically stimulated luminescence on quartz).Critics don’t dispute that the Santa Elina deposit itself is around 27,000 years old. The question is what made the holes. Archaeologist André Costopoulos, who reviewed the study, pointed out that three of the site’s 7,069 recovered osteoderms were identified as human-modified (roughly one in 2,350), and that work on bioerosion in extinct armadillo osteoderms published in 2021 documented insect and microbial lesions that can visually resemble the Santa Elina perforations.The recurring problem at very early American sites is that natural processes can mimic man-made ones. Frost, fire, rockfall, and water transport, for instance, can fracture stone in ways that mimic flaked tools. As a result, Santa Elina’s chronology rests on analyses of sediment, wood, and bone rather than the act of tool-making itself. The same is true of other Brazilian claims in the 30,000–20,000-year range, including Vale da Pedra Furada, Boqueirão da Pedra Furada, Toca da Tira Peia, and Sítio do Meio—all of which remain contested for similar reasons.The debate shifted sharply in March 2026, when Todd Surovell (from the University of Wyoming), Claudio Latorre (from the Pontifical Catholic University of Chile), and colleagues published a reassessment in Science concluding that Monte Verde II is only 4,200–8,200 years old, rather than 14,500. They argued that the site sits on a landform along Chinchihuapi Creek that formed about 8,600 years ago and that the ancient radiocarbon dates come from Ice Age wood redeposited by the creek. “It’s not an ice age site,” Surovell said, adding that the finding “supports an initial interior migration into continental North America as a viable colonization hypothesis.”The research was soon met with highly negative reactions from other experts. In three eLetters published in Science in early May 2026, 30 researchers—led separately by Tom Dillehay (from Vanderbilt University), geoarchaeologist Michael Waters (from Texas A&M), and archaeologist David Meltzer (from Southern Methodist University)—called the study’s claims “categorically false and found to be unsupported.” Dillehay maintains that there is no Lepué Tephra beneath MV-II, and that the samples came from a different context nearby. “After a few hours of fieldwork with no excavation, Surovell’s team proposed that the MV-II site dates to the mid-Holocene [...]. We uphold the original interpretation of MV-II as a late Pleistocene human occupation.” Waters called the geological work “egregiously poor.” Meltzer’s group accused the authors of trying to resurrect the theory that humans reached deep into the interior of North America before they entered South America. But Surovell said his team sees “little that raises serious concern,” and welcomes replication.Ancient DNA may provide a way to resolve these conflicting views that doesn’t depend on any single excavation—all ancient and present-day Indigenous American populations trace to lineages whose divergences can be dated genomically. The evidence shows that ancient Beringians split from other Native American ancestors around 20,900 years ago, and the northern and southern Native American branches separated around 15,700 years ago. That timeline doesn’t currently show a founding population established in South America 27,000 years ago.An April 2026 Nature study of 128 Indigenous American genomes deepened the picture without extending the timeline, describing several distinct South American dispersals, a faint Australasian ancestry signal in some Amazonian groups, and a previously unrecognized migration from central and southern Mexico into South America and the Caribbean beginning roughly 1,300 years ago. The continent’s population history, in other words, is far more layered than a single founding event, but the genomic clock still places first entry in the range of about 16,000 to 23,000 years ago, not 27,000.The Santa Elina team frames its own findings as an invitation rather than a verdict. “Questions regarding the timing and routes for human dispersal into the Americas remain open to debate, but human dispersal probably followed multiple routes and time frames, including Pacific coastal and inland (ice-free corridor) routes,” the authors wrote. “In this scenario, Late Pleistocene sites containing evidence of early human occupation in South America should be closely scrutinized with interest.”Tim Newcomb is a journalist based in the Pacific Northwest. He covers stadiums, sneakers, gear, infrastructure, and more for a variety of publications, including Popular Mechanics. His favorite interviews have included sit-downs with Roger Federer in Switzerland, Kobe Bryant in Los Angeles, and Tinker Hatfield in Portland.