cave

Wonderwerk Cave Entrance. (Credit: Wonderwerk Cave Project)

Scientists Used a Crime-Scene Trick to Find Evidence of Human Fire Use Nearly 2 Million Years Ago

In A Nutshell

  • A South African cave has yielded evidence of fire use dating back somewhere between 1.07 and 1.79 million years ago, pushing back one of the oldest known records of early human fire use
  • Researchers used a forensic technique that makes ancient burned bones glow red under special filter glasses, allowing fast, non-destructive detection in the field
  • Tiny owl-regurgitated mouse bones, accidentally scorched when cave fires ignited pellet piles, served as an independent marker of ancient flames
  • Evidence points to early Homo erectus carrying fire into the cave from outside, not making it on demand

A cave in South Africa has become one of the most important sites for studying early human fire use, and now scientists have pushed that record back even further by making ancient bones glow.

Wonderwerk Cave, tucked into South Africa’s Northern Cape Province, has been a treasure trove for researchers studying early human behavior. Previous research had already established that our ancestors were using fire in this cave roughly a million years ago. But a new study published in the journal PLoS ONE has found evidence that fire use at Wonderwerk goes back even further, in a deeper layer dated somewhere between 1.07 and 1.79 million years ago, possibly near the older end of that range. That is a significant addition to one of the most important chapters in human history.

Fire changed everything for early humans: warmth, light, protection from predators, and food that was easier to digest, which scientists think helped fuel growth of the human brain over time. When exactly our ancestors first started using fire remains one of the most contested questions in the study of human evolution. Wonderwerk Cave, with its deep interior and largely undisturbed sediment layers, is one of the best places on Earth to look for those answers.

A Crime-Scene Technique Detects Fire in Ancient Bones

What made this study stand out was not just what researchers found, but how they found it. Identifying ancient fire use is notoriously difficult. Bones can turn dark not because they were burned, but because a mineral called manganese seeps in during fossilization and stains them black. Telling those apart from genuinely scorched bones has traditionally required slow, expensive equipment.

Scientists introduced a new approach borrowed from forensic science, the kind used at crime scenes. By shining specialized blue light onto bone fragments, they were able to make burned bones glow red under special filter glasses. Bones that were never burned, or that changed color through natural processes, produced no glow at all. Fast, non-destructive, and portable, the technique can also be deployed directly in the field.

Researchers confirmed the technique by comparing results against a well-established lab method that identifies heat-related changes in bone at the molecular level. Both approaches agreed closely, and tests on bones from a Spanish archaeological site and experimentally burned bones confirmed the method.

stages of burning
White (#5 on right) is the most burnt while yellow-beige (on left #1) is unburnt (Credit: Wonderwerk Cave Project)

Owl Pellets Left Behind an Accidental Record of Ancient Fire

At the center of this study were not the remains of animals that early humans hunted, but the bones of tiny mammals, mice and similar creatures, swallowed and regurgitated as pellets by barn owls roosting inside the cave. Those pellets piled up on the cave floor over millennia. When fire was present, the pellet piles ignited, scorching the tiny bones inside. Because these small animals were not part of the human diet, their burned remains serve as an independent indicator that fire was present, an accidental record of flames moving through the cave.

In total, 161 small mammal fossil bones from two separate cave layers were examined. All 32 of the white and grey bones from the deeper layer showed signs of burning under both methods. Crucially, the burned bones were not scattered randomly across the cave floor. Instead, they were concentrated in two specific excavation areas located five meters apart, both containing only burned bones, while surrounding areas held a mix of burned and unburned remains. That kind of clustering is consistent with fire lit in particular spots, not with a wildfire sweeping through indiscriminately.

Owl Pellet
Modern Barn Owl Pellet from Floor of Wonderwerk. (Credit: Wonderwerk Cave Project)

Fire Found 30 Meters Deep in a Cave Points Away From Natural Wildfire

Proving the fire was set by human ancestors rather than by lightning or wildfires is one of this research’s biggest hurdles. During the time period in question, the area of the cave where these fossils were found would have been located roughly 30 meters from the cave entrance. Natural wildfires are not a likely explanation for fire reaching that far into the interior, though the researchers stop short of ruling it out entirely. Both burned layers are also separated by more than 80 centimeters of sediment, representing tens of thousands of years of time, which argues against a single burning event contaminating both.

Homo erectus, an early relative of modern humans, was most likely responsible. As the paper notes, “early hominins were not passive recipients of natural fires. They introduced fire, probably acquired from wildfires outside the cave, into the shelter and maintained it until it burnt out.” No evidence of cooking has been found at the site. What Wonderwerk shows is an earlier, simpler relationship with fire: not mastery, but use.


Paper Notes

Limitations

Neither the new luminescence method nor the established molecular technique it was validated against can detect burning that occurred below approximately 537 degrees Celsius, meaning lightly charred bones may go undetected. Luminescence also cannot be applied to black-colored bones, since black materials absorb all light. Chemical changes that occur to fossils over time, particularly a process in which fluoride ions are absorbed into bone mineral, can interfere with both methods, though luminescence showed some ability to distinguish these altered bones from genuinely burned ones. The dating of the deeper cave layer is expressed as a range of 1.07 to 1.79 million years ago, so a precise age for the fire evidence has not been established. The authors also note that the evidence is consistent with opportunistic fire use rather than controlled fire-making, and that the site has yielded no evidence of cooking.

Funding and Disclosures

Fieldwork at Wonderwerk Cave was funded by the Canadian Social Sciences and Humanities Research Council, granted to co-author Michael Chazan. Additional support came from a Leakey Foundation grant awarded to co-author Yolanda Fernández-Jalvo. No competing interests were declared. Funders had no role in study design, data collection, analysis, publication decisions, or manuscript preparation.

Publication Details

Authors: M. Dolores Marin-Monfort, Candice L. Shaw, Filipe Natalio, Liron Grossman, Peter Andrews, Joaquín Campos, Sara García-Morato, José M. Pereira, Alicia Pons, Michael Chazan, Liora Kolska Horwitz, Yolanda Fernández-Jalvo | Journal: PLoS ONE | Title: “New evidence for Early Pleistocene use of fire at Wonderwerk Cave (South Africa)” | Published: June 1, 2026 | DOI: 10.1371/journal.pone.0347480

About StudyFinds Analysis

Called "brilliant," "fantastic," and "spot on" by scientists and researchers, our acclaimed StudyFinds Analysis articles are created using an exclusive AI-based model with complete human oversight by the StudyFinds Editorial Team. For these articles, we use an unparalleled LLM process across multiple systems to analyze entire journal papers, extract data, and create accurate, accessible content. Our writing and editing team proofreads and polishes each and every article before publishing. With recent studies showing that artificial intelligence can interpret scientific research as well as (or even better) than field experts and specialists, StudyFinds was among the earliest to adopt and test this technology before approving its widespread use on our site. We stand by our practice and continuously update our processes to ensure the very highest level of accuracy. Read our AI Policy (link below) for more information.

Our Editorial Process

StudyFinds publishes digestible, agenda-free, transparent research summaries that are intended to inform the reader as well as stir civil, educated debate. We do not agree nor disagree with any of the studies we post, rather, we encourage our readers to debate the veracity of the findings themselves. All articles published on StudyFinds are vetted by our editors prior to publication and include links back to the source or corresponding journal article, if possible.

Our Editorial Team

Steve Fink

Editor-in-Chief

John Anderer

Associate Editor

Leave a Comment