A replica of "Karabo," an adolescent of the species Australopithecus sediba who lived 2 million years ago, on display at the Smithsonian National Museum of Natural History. Researchers are debating whether the species was an ancestor of humans. (Photo courtesy of the Smithsonian Institution)

He was built to climb, and yet he strode upright.

His arms hung low like an orangutan’s. Yet with his long thumbs and curved fingers he could grasp sticks and rocks, like a man.

His brain was not much larger than a chimpanzee’s. Yet his widened pelvis implied his kind gave birth to children with much bigger brains.

And so a fossilized adolescent named Karabo — which means “answer” in a South African dialect — is raising a lot of questions about human evolution.

Researchers found his skeleton, and much of an adult female’s, in a cave about 25 miles north of Johannesburg in 2008 and announced the discovery in 2010. They coined a new species, Australopithecus sediba , and launched an intensive multi­national effort to study the find.

This 2-million-year-old fossilized skull of a young male Australopithecus sediba represents a putative human ancestor, researchers say in the Sept. 9 issue of Science magazine. (Photograph by Brett Eloff/Brett Eloff, University of Witswatersrand)

In the journal Science, the team provides detailed descriptions of the creatures’ heads, hands, feet and hips. It also dates the fossils to 1.98 million years ago, smack in the middle of an era famous for its lack of evidence of possible human relatives.

The mash-up of humanlike and apelike traits resemble a “a stop-action snapshot of evolution in action,” said Richard Potts, director of the Smithsonian Institution’s Human Origins Program. He was not involved in the research.

The researchers stop just short of calling the creatures an ancestor to the human lineage known as Homo. But they place A. sediba squarely in the running for that coveted title.

The species is “possibly the best candidate” yet for a Homo ancestor, said Lee R. Berger of the University of Witwatersrand in Johannesburg. Berger, along with his then-9-year-old son Matthew, discovered Karabo in a fossil-rich region known as the Cradle of Humankind.

Over the past four decades, scientists have sketched in branches on the human evolutionary tree from nearly a dozen species of creatures that roamed East and South Africa stretching back 4 million years. The emerging picture is not a linear march from ape to man but rather a profusion of cousin species overlapping in time and space. Some died out, others continued to evolve — but divining which were which has proved challenging.

The new fossils’ feet and ankles are “mostly human,” said Bernhard Zipfel, also of the University of Witwatersrand. And yet, the heel looks more apelike — one of many such anomalies.

The knee lined up above the ankle, not angled out as in apes, a giveaway that the creatures walked upright. And yet, bumps on the inside of the ankle and other features mark the ankle as more chimplike, the “type of ankle you need to climb a tree,” Zipfel said.

The thumb was long relative to the fingers — even longer than ours. The fingers curved, suggesting a powerful grasp. Toolmaking “was possible,” said Tracy Kivell of the Max Planck Institute for Evolutionary Anthropology in Germany, although the scientists have found no stone tools nearby.

In France, scientists scanned Karabo’s skull using high-powered X-rays. The imprint of the brain on the inner skull suggests that certain areas had begun to evolve and grow into those of more humanlike species, the researchers said.

The pelvis is another puzzler. It is bowl-shaped, like that of the human lineage, and widened compared with that of earlier Australopithecines. It looks designed to give birth to big-brained offspring. And yet, the brain is small.

“This is what evolutionary theory would predict, exactly this confusion, this mix of features of Australopithecines and Homo,” said Darryl de Ruiter, an anthropologist at Texas A&M University who contributed to the research.

In the famously contentious arena of human origins research, Berger’s assertion of A. sediba as a putative Homo ancestor is already under attack.

“Even if it does share more features of Homo than other species, it could be a close sister to us rather than an ancestor,” said Carol Ward, a paleoanthropologist at the University of Missouri who was not involved in the work. “It has some features that are more like Homo, like the pelvis and the hand. That’s interesting right there. They don’t need to overstate the importance of their fossils.”

And then there’s the timing. The rock encasing the fossils yielded a date of death of 1.98 million years ago. But by that time, a few fragments of Homo-like fossils had already been buried in East Africa for 300,000 years. That makes it difficult to call A. sediba an ancestor to the Homo lineage, which eventually evolved into us.

Berger deals with this problem by dismissing the older fossils as not sufficiently Homo-like, particularly a piece of jaw Donald C. Johanson and colleagues found in Ethi­o­pia in the mid-1990s.

Johanson, the director of the Institute of Human Origins at Arizona State University who discovered the most famous pre-human of them all, Lucy, in 1974, complained, “They are going out of their way to sweep all of these [older] fossils under the rug.” The disputed jaw’s co-discoverer, William Kimbel, also of Arizona State, said that Berger and his team declined an offer to view that fossil. “How they can make any kind of informed judgment . . . is beyond my comprehension,” Kimbel wrote in an e-mail.

Johanson suggested that Berger’s fossils could be “another branch of the family tree” that died out instead of evolving into humans.

Whatever the case, the hunt for a deep human ancestor will continue. Said Johanson: “There seems to be some magic with our direct ancestor. If it gave rise to Homo, that somehow makes it very, very special.”