A single bone fragment, unearthed more than thirty years ago and then forgotten in South African reserves, has reshuffled the cards of human evolution. The cranial fragment DNH 127, discovered in Drimolen (South Africa) and long left unstudied, confirms the presence of Homo erectus about two million years ago, 200,000 years before any other known trace of the species. Out of 18 new hominin fossils described at Drimolen, 14 belong to Paranthropus robustus and only one to Homo erectus, the site totaling 119 specimens, some of which remain to be studied. The story begins at the Drimolen Main Quarry site in South Africa, nestled in the UNESCO World Heritage site known as the "Cradle of Humankind," located about forty kilometers north of Johannesburg. It is here that research teams uncovered no fewer than 18 hominin fossils between 1992 and 2019, none of which had previously been the subject of in-depth scientific publication. This delayed description is not trivial. It is the result of meticulous work conducted by Dr. Jesse Martin from La Trobe University, in collaboration with South African and American researchers from the University of Johannesburg and the University of the Witwatersrand. Their study, published in the journal Annals of Human Biology, has brought to light pieces that had literally been lying in boxes for nearly thirty years. Among these fossils, one specimen stands out: DNH 127, a partial fragment of the temporal bone, that is, the part of the base of the skull that houses the inner ear. This piece of bone, although modest in appearance, was sufficient to raise a doubt that had been lingering for several years about the true identity of the site's earliest inhabitants. Before the discovery of DNH 127, scientists already had an intriguing clue with DNH 134, a fossil discovered in 2015 belonging to a child aged two to three years. Problem: at that age, anatomical characteristics do not always allow for certainty in determining the species concerned. Doubt was therefore permissible. With this new cranial fragment belonging this time to an adult, the uncertainty is lifted. DNH 127 confirms that it is indeed Homo erectus, and its dating is unequivocal: about two million years. A date that pushes back by 200,000 years the known presence of this species compared to any other documented site, including that of Dmanisi in Georgia, where comparable fossils had been dated to 1.8 million years. Why does this chronological precision matter so much? Because Homo erectus is considered the first species of the genus Homo, likely the direct ancestor of Homo sapiens. This species has existed for about two million years and has spread across Africa and a large part of Eurasia, a much wider range than that of its predecessors. This discovery also reignites a scientific debate that had been simmering for some time between researchers from East Africa and South Africa, each claiming in a way the geographical paternity of our origins. The Georgian fossil had even temporarily tilted the balance towards an origin outside of Africa, a hypothesis that is losing ground today. According to Jesse Martin, the East African fossils usually attributed to Homo erectus would actually be less ancient than previously thought, with a confirmed dating around 1.6 million years and an uncertain estimate at 1.9 million years, without presenting the ancestral characteristics expected of a founding population. The researcher presents a bold hypothesis: Homo erectus may have evolved elsewhere, perhaps in West or Central Africa, regions still very little explored on the paleontological front, before migrating to southern Africa. He also mentions a bias inherited from the history of archaeological research, which has long focused excavations on certain areas at the expense of others. Another fascinating lesson from this study: Homo erectus was not alone at the Drimolen site. Fossils of Paranthropus robustus, another hominin lineage, were found in the same geological layers, covering a period of approximately 100,000 years. The two species therefore truly coexisted on the same territory for several tens of thousands of years. These two evolutionary cousins, however, took radically different paths. Homo erectus developed an increasingly voluminous brain and increasingly small molars, a transformation that researchers associate with an increasing consumption of meat. On the other hand, Paranthropus robustus retained a brain barely more developed than that of a chimpanzee, but bet on massive molars, true tools comparable to a mortar and pestle, shaped to crush particularly hard foods. And the power balance between the two species was far from balanced. Two million years ago, Paranthropus robustus was more than ten times more numerous than Homo erectus at the Drimolen site. A figure that is staggering when one knows that it was actually the minority lineage that would, millions of years later, give rise to Homo sapiens. Nothing guaranteed at the time that our lineage would survive. The Paranthropus robustus fossils found at Drimolen, two million years old, also present notable, although subtle, differences with those unearthed in the neighboring cave of Swartkrans, dated to 1.8 million years. This progressive evolution of teeth and masticatory muscles over time illustrates a phenomenon of micro-evolution, a fine and continuous adaptation of the species to its environment. What makes Drimolen Main Quarry truly unique does not reduce to a single spectacular discovery. The site constitutes the largest concentration of primitive human remains ever discovered, with no fewer than 119 hominin specimens cataloged since its discovery in 1992. A considerable number, but which hides a surprising reality: many specimens have yet to be described in detail by the scientific community. Before this new study, only part of the paleontological treasure of Drimolen had been studied in depth: 105 dental elements, several emblematic skulls referenced DNH 7, DNH 134, DNH 152, and DNH 155, a pelvis called DNH 43, as well as a thirty of post-cranial elements. The rest was patiently waiting its turn. Thanks to the meticulous description work conducted by Jesse Martin's team, 18 new cranial specimens could finally be classified. The verdict: one fossil attributed to Homo erectus, 14 to Paranthropus robustus, and three remaining for now impossible to conclusively attribute to a specific species, classified in the general category of hominins. The researchers have also identified a new trait at the level of the mandible that could, in the long term, allow distinguishing the Drimolen population from other fossil populations compared in the study. Among the masterpieces of this collection, the skull DNH 7 remains to this day the most complete known skull of Paranthropus robustus, while DNH 155 stands out for its exceptional state of preservation. The mandible DNH 8, also remarkably preserved, has allowed enriching the data already available on the species' dentition. All these elements show that beyond the spectacular discovery around Homo erectus, an entire part of the history of Paranthropus robustus continues to be clarified, fossil by fossil. This abundance of material still to be studied suggests an exciting outlook for the years to come. If a single bone fragment, forgotten for thirty years, was sufficient to push back the emergence of Homo erectus by 200,000 years and to revive a scientific debate spanning several decades, one can legitimately wonder what the future analyses of the still undeciphered Drimolen specimens will reveal. The site clearly has not finished surprising us.