Florisbad’s strange ancient landscapes and early Homo sapiens 200,000 years ago
Around 200,000 years ago, early Homo sapiens in what is now South Africa lived in landscapes that no longer exist anywhere. New chemical evidence from two sites, Florisbad and Lombard’s Drift, suggests the plant communities around them mixed vegetation types in ways that have no close modern match. The finding gives researchers a sharper picture of the world our earliest ancestors had to handle.
Why Florisbad matters
Florisbad, in the Free State region of South Africa, is a famous spring site. It produced the Florisbad cranium, a fossil often discussed in debates about the origins of our species. Lombard’s Drift, also in South Africa, adds a second record to compare. Both sites hold sediment layers that built up over many thousands of years, and those layers keep traces of the plants that once grew nearby.
A team led by researchers including Michaela Ecker, Maïlys Richard and Michael Toffolo of the University of the Free State published the work on 30 October 2026 in Communications Earth & Environment. Their paper is titled “Non-analog environments supported early Homo sapiens in southern Africa 200 000 years ago.”
Reading ancient plants from wax
Plants coat their leaves in wax made of long-chain molecules called n-alkanes. These molecules survive in sediment for a very long time. By measuring their abundance and their stable carbon and hydrogen isotopes, scientists can estimate which kinds of plants grew in a place and how much water was available.
The team analysed 21 samples from Florisbad and five from Lombard’s Drift. They also produced new luminescence ages, which measure when mineral grains were last exposed to light. Those ages place key layers at Florisbad at about 205,000 years (Unit O), about 198,000 years (Unit N) and about 168,000 years (lower Unit M), each with an uncertainty of roughly 14,000 to 19,000 years. The new dates also revise the age of the Florisbad cranium.
Landscapes with no modern match
The biomarker values for the Pleistocene layers fall outside the ranges seen in modern South African biomes, especially for two warm-and-cool phases known as MIS 5 and MIS 6. Researchers call such settings “non-analog.” The layers from around 200,000 years ago show a strong share of C3 plants, which include many shrubs and trees, and the pollen record includes fynbos-type plants. The authors estimate that C4 plants, mostly warm-season grasses, covered about 30 percent of the ground in Units O and N. In lower Unit M, that share rose to about 47 percent, and in upper Unit M it fell to about 25 percent.
By contrast, samples from the Holocene, the last 11,700 years, sit close to the modern Grassland Biome. So the world early Homo sapiens knew at Florisbad was not just a wetter or drier copy of today’s grassland. It was a different mix altogether.
What it may mean for tools and food
Press coverage of the study links these shifting ecosystems to how people used stone tools. According to that reading, groups at Florisbad may have resharpened and reworked their tools more heavily when food was harder to find, and made quicker, lightly worked flakes when resources came easily. That idea fits a wider pattern in which tool care tracks risk. We could not check this tool-use link against the paper’s full text, so it is best treated as a working interpretation rather than a settled result.
The paper’s own headline message is more direct: unusual, varied plant environments supported early Homo sapiens in southern Africa. A mosaic of shrubs, grasses and other plants could offer different foods across seasons, which may have helped people adapt.
Caveats and open questions
The authors are open about the limits. Plant wax preserved poorly in several layers, so some important units have no isotope data at all. Many samples held only low concentrations of the molecules, and the signal from some wax types can come from either succulent plants or C3 grasses, which makes interpretation harder. Pollen and biomarker records do not always agree, because they sample different areas and time spans. Some luminescence methods may overestimate ages if grains were not fully bleached, so the team prefers the conventional quartz results. At Lombard’s Drift, only one sample is dated.
Future work that ties tool assemblages to each dated layer should show whether tool reworking truly followed food supply. For now, Florisbad shows that our species emerged in landscapes we can no longer visit, and that reading their chemistry is one way to rebuild them.