When Was Victoria Falls Formed and How Did It Happen?

Victoria Falls, straddling the border between Zambia and Zimbabwe, owes its existence to a combination of ancient volcanic rock laid down roughly 180 million years ago and a river system that only reached its current form within the last few hundred thousand years. The falls did not appear in a single dramatic event. Instead, they are the product of the Zambezi River slowly exploiting cracks in a basalt plateau, carving a succession of gorges that record where the waterfall stood at earlier points in time. The story of how Victoria Falls came to be is really two stories: one about very old rock, and one about a surprisingly young river.

The Basalt Foundation

Everything about Victoria Falls starts with the rock the water falls over. The Zambezi gorge system downstream of the falls cuts through a thick sequence of flood basalts known as the Batoka Basalts. These lavas were erupted during the breakup of the ancient supercontinent Gondwana, as part of a massive volcanic episode called the Karoo Large Igneous Province. Argon-argon dating of the Batoka Basalts places their age tightly at around 179 to 180 million years old, squarely in the early Jurassic period.1Geochemistry, Geophysics, Geosystems. Age of the Batoka basalts, northern Zimbabwe, and the duration of Karoo Large Igneous Province magmatism These are dense, hard volcanic rocks that resisted erosion for tens of millions of years. But they are not uniformly solid. The basalt plateau is crisscrossed by networks of joints, fractures, and faults. Some of these fractures were filled over time with softer material. That contrast between hard basalt and weak fracture zones is the key to how a waterfall can exist in such tough rock.

The basalt forms a broad, relatively flat plateau across this part of southern Africa. Water flowing across the surface encounters the fracture zones, and over long periods, erosion preferentially removes the weaker material along those cracks. This selective erosion is the fundamental mechanism behind the falls and the gorge system that stretches downstream from them.

How the Zambezi River Reached Victoria Falls

The rock may be 180 million years old, but the river flowing over it is far younger. The Zambezi as we know it today is a patchwork of formerly separate river systems stitched together by tectonic forces and drainage reorganizations over millions of years. The Upper Zambezi, which drains the highlands of Angola and western Zambia, and the Middle Zambezi, which runs through the gorges below Victoria Falls and on toward Mozambique, were once entirely independent rivers flowing in different directions.

One influential model proposes that the Upper Zambezi originally drained not eastward to the Indian Ocean, as it does today, but southward into a vast inland basin in what is now Botswana, feeding a system of ancient lakes known as the Makgadikgadi palaeo-lakes. According to this model, tectonic uplift along the Chobe Fault, which runs roughly east-west across the path of the ancestral Zambezi, blocked the river’s original course and diverted water into the Makgadikgadi Basin.2South African Journal of Geology. THE EVOLUTION AND AGES OF MAKGADIKGADI PALAEO-LAKES: CONSILIENT EVIDENCE FROM KALAHARI DRAINAGE EVOLUTION SOUTH-CENTRAL AFRICA Over time, this basin filled to become a lake that, at its largest, functioned much like modern Lake Victoria: a shallow, broad body of water that influenced local climate by increasing rainfall and reducing evaporation.

The critical event for Victoria Falls came when this lake overtopped the barrier created by the Chobe Horst, spilling water eastward into the course of the proto-Middle Zambezi. This river capture event effectively connected the Upper and Middle Zambezi into a single system for the first time. Archaeological evidence associated with the deposits from this capture places it at around the peak of the last interglacial, near the end of the Middle Pleistocene, which puts the event in the range of roughly 100,000 to 200,000 years ago.3Palaeogeography, Palaeoclimatology, Palaeoecology. The Zambezi River: tectonism, climatic change and drainage evolution That capture event was apparently dramatic. Sedimentary deposits downstream are interpreted as the record of a catastrophic flood, the enormous rush of water released when the lake breached its dam.

This means the Zambezi in its current form, with the full volume of the Upper Zambezi feeding into the gorge system at Victoria Falls, is geologically very young. The waterfall process likely predates the capture event, since the Middle Zambezi already flowed through the area, but the sheer scale of Victoria Falls as we see it today depends on the combined flow of the joined river. Before capture, a smaller river was working on the basalt; after capture, the process accelerated dramatically.

The Drainage Network Kept Changing

The connection of the Upper Zambezi to the Middle Zambezi was not the only drainage reorganization in the region. The Makgadikgadi lake system went through a series of contractions as successive river captures stole tributaries away from the basin. The loss of the Upper Chambeshi catchment caused the lake to drop from one shoreline elevation to a lower one. The severing of a former link between the Kafue and Zambezi rivers caused a further drop. The impoundment of a large lake on the Upper Zambezi itself, Palaeo-Lake Bulozi, reduced inflow still further.2South African Journal of Geology. THE EVOLUTION AND AGES OF MAKGADIKGADI PALAEO-LAKES: CONSILIENT EVIDENCE FROM KALAHARI DRAINAGE EVOLUTION SOUTH-CENTRAL AFRICA Each of these events was driven by tectonic activity, specifically fault movements that redirected rivers into new channels. The Makgadikgadi palaeo-lakes eventually desiccated entirely as their water sources were progressively removed.

What this means for Victoria Falls is that the river’s volume and erosive power have fluctuated over time. Periods when more tributaries fed into the system would have accelerated gorge cutting; periods of drought or river diversion would have slowed it. The falls are not just a product of rock and water, but of a tectonic chess game that rearranged the entire drainage network of south-central Africa over hundreds of thousands of years.

How the Gorges Record Former Waterfall Positions

One of the most striking features of Victoria Falls is the system of zigzagging gorges that stretches downstream for several kilometers. These are not random ravines. Each gorge represents a former position of the waterfall itself. The falls have been retreating upstream through the basalt plateau, and every time the river eroded through one fracture line and reached the next, the waterfall jumped to a new position, leaving the old gorge behind as an empty channel.

The zigzag pattern reflects the geometry of the fracture network in the basalt. The joints and faults do not all run in the same direction; they intersect at angles. As the waterfall erodes along one fracture, it eventually breaches a perpendicular or oblique fracture, causing the lip of the falls to shift direction. The current Victoria Falls occupies the most recent of these fracture lines, with the spray rising from a chasm that runs roughly east-west. Downstream, the abandoned gorges angle off in different directions, each recording the orientation of the fracture the river exploited at that stage.

At least seven or eight former waterfall positions have been identified in the gorge system. The oldest gorges, farthest downstream, are the most eroded and hardest to trace. The youngest gorge, closest to the current falls, still has sharp walls and a clear profile. By studying the spacing and size of these gorges, geologists can estimate how fast the falls have been retreating. Estimates vary, but the recession rate appears to be on the order of a few centimeters to perhaps a meter per century, depending on rock conditions and river flow. That pace is slow enough that the current position of the falls has been stable for thousands of years in human terms, but fast enough that the gorge system has grown substantially over the last hundred thousand years or so.

There is also evidence that the next retreat is already beginning. At the eastern end of the current falls, a notch known as the Devil’s Cataract is cut somewhat deeper into the lip than the rest of the crest. This notch follows a fracture that extends upstream, and geologists have long suspected it marks the point where the falls will eventually break through to their next position. When that happens, probably thousands of years from now, the current falls will become one more abandoned gorge in the series.

Livingstone’s Theory and Why It Was Wrong

When David Livingstone became the first European to see the falls in 1855, he proposed that the chasm was a crack or “rent” in the earth caused by some tectonic upheaval, as if the ground had simply split open and the river had fallen into the gap. This explanation had an intuitive appeal. The gorge below the falls is so narrow and deep, with such sheer walls, that it does look like a crack in the landscape. Livingstone’s theory went largely unchallenged for decades and was repeated in virtually every subsequent description of the falls.4Geological Magazine. Notes on the Geological History of the Victoria Falls

The problem with the rent theory is that it does not account for the multiple gorges downstream. If the chasm were simply a tectonic crack, there should be one gorge, not a whole series of them stretching for kilometers. Geologists working in the early twentieth century recognized this and proposed the erosion model that is now standard: the gorge was carved by the river itself, working along pre-existing fractures in the basalt over long periods of time. The rock was not split by an earthquake; it was worn away by water, one weak zone at a time. Livingstone can be forgiven. He saw the falls from a limited vantage and had no aerial perspective on the gorge system. It took later exploration, and eventually aerial photography, to reveal the full zigzag pattern that makes the erosive origin clear.

Mosi-oa-Tunya and the People Who Were Already There

Long before Livingstone arrived, the falls were central to the lives of the Toka-Leya people, an indigenous group of the Tonga language family who have inhabited the area surrounding Victoria Falls for hundreds of years. They know the falls by two names: Mosi-oa-Tunya, meaning “the smoke that thunders,” and Shungu Namutitima, meaning “boiling water.” Both names capture the physical reality of the falls with an immediacy that the European name, honoring a distant queen, does not.5Nomina Africana: Journal of African Onomastics. “The Devil’s Cataract”: Demonising the Tonga people’s spirituality through colonial naming

The Toka-Leya comprise two related groups, the Toka (Batoka) and the Leya (Baleya), who speak dialects of Tonga. For them, the falls are not a geological curiosity but a sacred place woven into spiritual and cultural life. The name “Batoka” also attaches to the basalt gorge system itself, a reminder that the geological and cultural histories of this landscape are entangled. Colonial-era naming practices often overwrote indigenous names and spiritual associations, sometimes deliberately. The Devil’s Cataract, the deepest notch at the eastern end of the falls, received its English name in a way that reframed Tonga spiritual significance as something sinister, a pattern common across colonial Africa.5Nomina Africana: Journal of African Onomastics. “The Devil’s Cataract”: Demonising the Tonga people’s spirituality through colonial naming

Why the Falls Are Where They Are

Given everything above, the location of Victoria Falls is not random. It sits at a geologically specific spot: the point where the Zambezi River crosses a basalt plateau riddled with fractures, in a region where tectonic activity has been actively reshaping drainage patterns for millions of years. If the basalt were uniformly solid with no fracture network, the river would flow over it as a broad, shallow cascade with no deep gorge. If the rock were uniformly soft, the river would have carved a wide valley rather than a narrow slot. It is the contrast between hard basalt and weak fracture zones that creates the dramatic vertical drop and confined gorge.

The falls are also positioned at a natural break in the landscape’s gradient. The area upstream is a broad, flat floodplain where the river spreads out to a width of more than a kilometer during the wet season. At the lip of the falls, this wide, shallow flow is suddenly funneled into a narrow chasm barely 100 meters wide. That abrupt transition from flat plateau to deep gorge is the signature of waterfall recession through fractured basalt: each new fracture zone the river reaches creates a new cliff face, and the flat stretch upstream simply represents the surface of the next basalt layer waiting to be undermined.

The Ongoing Tectonic Context

Southern Africa is not tectonically dead. The region sits near the southern extension of the East African Rift System, which is slowly pulling the African continent apart. The faults and fractures that control the gorge pattern at Victoria Falls are related to this broader tectonic framework. The Chobe Fault, whose uplift originally diverted the Upper Zambezi into the Makgadikgadi Basin, is part of this rift-related system.2South African Journal of Geology. THE EVOLUTION AND AGES OF MAKGADIKGADI PALAEO-LAKES: CONSILIENT EVIDENCE FROM KALAHARI DRAINAGE EVOLUTION SOUTH-CENTRAL AFRICA Future tectonic movements could, in principle, redirect drainage yet again, altering the Zambezi’s course or flow volume in ways that would change the falls.

For now, the falls continue their slow upstream retreat. Each wet season, the full force of the Zambezi hammers at the lip and walls of the gorge, scouring out weaker rock along fractures. Spray saturates the basalt, and cycles of wetting and drying weaken the rock further. Blocks of basalt periodically collapse from the overhanging lip, widening existing cracks and extending the gorge by small increments. The process is imperceptible on a human timescale. Visitors returning to Victoria Falls a century from now would see essentially the same cataract in essentially the same place. But over thousands of years, the lip will retreat, the current gorge will be abandoned, and the falls will occupy a new position upstream. The geological clock that has been ticking for at least a hundred thousand years shows no sign of stopping.