Australopithecus sediba: The Mosaic That Split a Field

Few hominin fossils are dated as precisely as the skeletons from Malapa cave: 1.977 million years old, give or take about 2,000 years. That figure, published by Robyn Pickering and colleagues in Science in 2011, came from uranium–lead dating of flowstone layers above and below the bones. It belongs to Australopithecus sediba, one of the most complete and most argued-over early hominins ever found.

The first bone was spotted on 15 August 2008 by Matthew Berger, the young son of palaeoanthropologist Lee Berger of the University of the Witwatersrand. It was a right collarbone. Excavation at Malapa, in the Cradle of Humankind north-west of Johannesburg, went on to produce two partial skeletons.

Lee Berger’s team named the species in Science in April 2010. Sediba means “fountain” or “wellspring” in the Sesotho language.

How do you date a cave to 2,000 years?

Caves are notoriously hard to date, because bones can wash or fall into older or younger deposits. At Malapa, the fossils lay in sediment sandwiched between sheets of flowstone, a layer of calcite left by water running across a cave floor. Flowstone contains traces of uranium, which decays into lead at a known rate, so the layers above and below the bones give minimum and maximum ages.

Pickering’s team then narrowed the window using the magnetic signal locked in the sediments. They record a brief episode when Earth’s magnetic field flipped to “normal” polarity, an event that can be matched to a known date. That is how a 2-million-year-old deposit gets an uncertainty of only a few thousand years. For more on these clocks, see how scientists date fossils.

Two skeletons, two life stages

The two individuals are known by their site labels:

• MH1 (Malapa Hominin 1) is a juvenile male, estimated at around 12–13 years old. He had a nearly complete skull and stood no more than about 1.3 metres tall.
• MH2 is an adult female, represented by teeth, part of a jaw and much of the skeleton, including a remarkably complete hand.

Having both a youngster and an adult from one moment in time is rare. It lets researchers see how the species grew. It also creates a trap: some of MH1’s Homo-like traits might have changed by adulthood, a point critics raise when judging how close the species is to Homo.

What makes Australopithecus sediba a mosaic

Palaeoanthropologists use the word mosaic for a species that mixes old and new features. Few fossils fit the word better.

The brain measurement comes from a 2011 Science study by Kristian Carlson and colleagues, who scanned MH1’s skull with synchrotron X-rays to build a virtual cast of the inside. At about 420 cc, the brain was only slightly larger than an average chimpanzee’s. Yet the team reported a frontal region whose shape looked closer to humans than other australopiths. If that is right, the brain began to reorganise before it began to grow.

The hand is just as striking. In 2011, Tracy Kivell and colleagues described MH2’s right hand, one of the most complete early hominin hands known. It pairs a long thumb and short fingers, the proportions needed for a precise pad-to-pad grip, with strong, curved finger bones of the kind used for climbing. Whether sediba made stone tools is unknown; the hand would have allowed it.

That conclusion carries a caveat. A brain cast records the inside of the skull, not the brain itself, and reading fine anatomy from bumps on bone is notoriously difficult.

A strange way to walk

Australopithecus sediba walked upright, but not quite like Lucy or like us. In 2013, Jeremy DeSilva and colleagues reconstructed its gait from the leg and foot bones. They argued that it landed on the outer edge of the foot and then rolled sharply inward, a motion called hyperpronation. The reconstruction implies extra rotation at the knee and hip with each step.

Fresh evidence arrived in 2021, when Scott Williams and colleagues described newly recovered lower-back vertebrae of MH2 in eLife. The spine had a human-like inward curve, a sign of regular upright walking, but also large attachment points for trunk muscles that would have helped with climbing.

The picture is of an animal at home on the ground and in the trees. It also suggests that bipedalism evolved along more than one path, rather than as a single march toward the human stride.

The ancestor problem

Berger’s team proposed that A. sediba was a good candidate for the ancestor of Homo, or something very close to it. The trouble is timing.

In 2015, Brian Villmoare and colleagues described LD 350-1, a jaw from Ledi-Geraru in Ethiopia dated to about 2.8 million years ago and assigned to Homo. That is roughly 800,000 years older than Malapa. In 2019, Andrew Du and Zeresenay Alemseged tested the problem statistically in Science Advances and concluded that, given the known fossil record, A. sediba is unlikely to be the ancestor of Homo.

Defenders have a reply. A species can survive long after it gives rise to a descendant, just as an ancestral population can persist beside its offshoots. On that view, Malapa could be a late snapshot of a lineage that began much earlier. The difficulty is that no older sediba fossils have been found to support it.

The Smithsonian’s Human Origins Program lists the main options:

1. Ancestor of Homo, despite the timing, as originally proposed.
2. Close relative of an unknown ancestor, a sister lineage that survived after Homo had already appeared.
3. Not especially close to Homo at all, if its Homo-like traits turn up in other australopiths or reflect MH1’s youth.
4. Early Homo itself, which would mean redrawing where the genus begins.

What sediba changed, whatever its place

Even if A. sediba is a side branch, it has already done something important. It shows that traits once treated as a package — small teeth, a precision hand, a human-like pelvis, a big brain — did not evolve together. Different parts of the body changed at different times, and some may have evolved more than once.

That makes defining the genus Homo harder. Is it the hand, the teeth, the body proportions or the brain that counts? Malapa forces the question without answering it. For the species that most researchers see near the root of Homo, read Homo habilis vs Homo erectus. For the South African species sediba probably descended from, see the Australopithecus africanus profile and the Taung Child.

You can set sediba against its neighbours on the interactive family tree, or line up brain sizes and dates for different species with the species comparison tool.