The Miocene is often described as the age when forests gave way to savannas. That summary is too simple. Grasslands, woodlands and forests shifted in different ways across continents, and the expansion of grasses did not happen at one time everywhere. Fossil teeth and plant remains preserve a more complicated history: regional ecosystems changed as climate, atmospheric carbon dioxide, fire and geography interacted.
What counts as a Miocene grassland?
The Miocene Epoch lasted from about 23 million to 5.3 million years ago. It included warm intervals, long-term cooling and major changes in rainfall. A savanna is not simply a treeless plain. It is a landscape where grasses coexist with shrubs and trees, often under seasonal rainfall and recurring disturbance. Fossil deposits can preserve evidence for wooded habitats, open patches or a mosaic of both, rather than a single uniform biome.
Grasses using the C4 photosynthetic pathway became more important in many ecosystems during the later Miocene, but their history was staggered. Carbon-isotope measurements from fossil mammal enamel can distinguish diets rich in C3 plants from those containing more C4 plants. Enamel is a record of what an animal ate over time, not a direct photograph of the vegetation. It can show that an animal fed in a landscape with C4 resources, while pollen, phytoliths, soil carbonates and plant fossils help identify what grew nearby.
A spread with different regional clocks
Some East African records preserve evidence of C4 grasses earlier than the familiar late Miocene expansion. A multiproxy study of fossil localities found evidence for C4-rich habitats in parts of eastern Africa during the Early Miocene, challenging the older idea that open grasslands appeared there only very late. Other records show that C4 plants became a major food source for many herbivores later, and the timing varies among sites. The findings are compatible when local presence is separated from regional dominance.
Across several continents, tooth enamel and paleosols record a wider rise in C4 vegetation around the late Miocene, especially between roughly 8 and 6 million years ago. Yet not every animal switched diets at the same time. Some mammals continued browsing on leaves and shrubs while grazers expanded into open habitats. A regional fossil sequence is therefore more informative than a single global date.
Climate mattered, but no single cause explains every site
C4 grasses are often efficient in warm, sunny and seasonally dry settings, and their expansion has been linked to lower atmospheric carbon dioxide. But aridity alone cannot explain every record: some grass-rich ecosystems emerged without a matching increase in local dryness. Fire, herbivory, soil conditions, monsoon systems and the rise of mountain ranges could all influence the balance between grasses and woody plants. Their effects varied with latitude and local geology.
Grassland animals also changed, but the familiar picture of a single evolutionary race between fast grazers and fast predators should be treated cautiously. High-crowned teeth can help resist wear from abrasive diets and dust, yet tooth shape does not identify one plant with certainty. The same landscape could support grazers, browsers and mixed feeders. See how the Miocene drylands of Central Asia preserved a different regional ecological story.
Why the modern savanna is not a Miocene snapshot
Modern African savannas are useful comparisons, not exact replicas of Miocene habitats. Their species, fire regimes and human influences are different, and fossil sites are filtered by preservation. A tooth may come from an animal that travelled from another habitat; plant microfossils can be carried by wind or water; and a missing grass fossil does not prove grass was absent. Researchers combine independent records and state how each one can bias the picture.
The safest conclusion is that the Miocene saw major, uneven expansion of grass-rich ecosystems. It changed the foods available to mammals and the structure of many landscapes, but it did not replace forests everywhere or create today's savannas in a single event. The broader Cenozoic climate and continental history helps explain why neighboring regions could follow different ecological paths.
Frequently asked questions
When did Miocene grasslands first appear?
Grass-rich habitats developed at different times in different regions. Some records show C4 vegetation in parts of East Africa during the Early Miocene, while wider expansion occurred later.
How do fossil teeth reveal ancient grasslands?
Carbon isotopes in enamel can indicate whether an animal ate mostly C3 or included C4 plants. They record diet, so researchers compare them with plant fossils and soil evidence.
Did savannas replace forests across the whole planet?
No. Forests, woodlands and open habitats persisted in different regions and often formed mosaics. The change was uneven rather than a single global replacement.
Did climate alone cause the rise of grasslands?
No single cause explains every site. Carbon dioxide, rainfall, fire, soils, herbivory and mountain building could interact, with different effects in different places.