If Ardi’s skeleton was an unexpected mosaic, her habitat was even more curious. The Middle Awash scientists analyzed more than 150,000 vertebrate fossils from the site, from rats to foxes to saber-toothed cats, along with hundreds of geologic samples, to arrive at a detailed understanding of Ardi’s habitat. “It was like a whole series of snapshots across an ancient landscape,” White says.

For decades anthropologists have argued the “savanna hypothesis”: that bipedalism evolved on the savannas of Africa as spreading grasslands forced our ancestors to walk increasing distances across open territory. As White and his team analyzed their evidence, they realized that Ardi must have lived in the woods. In that case, bipedalism must have emerged for different reasons. “Since her species was already bipedal and already had reduced canines, those characteristics were not the result of adaptation to savanna,” White says.

But Ardi’s most important legacy could be the light she sheds on our last common ancestor, that mysterious creature that ultimately gave rise to both today’s humans and our closest living relatives, the chimpanzees. “There’s a big gap in our knowledge about our past,” White says, “and it lies down there somewhere about 7 million years ago in the form of a last common ancestor. It may never be found. But Ardi tells us what that creature looked like, and it’s something we never expected.”

The long-favored view is that the last common ancestor must have been similar to a chimp, with more evolutionary change occurring subsequently on the human branch of the family. But Ardi’s anatomy suggests that our last common ancestor was like neither a human nor a chimpanzee. The shape of Ardi’s hands makes the point. Their anatomy contains bony structures that would have allowed her to walk comfortably on her palms, more like monkeys than like living apes. “The Ardi wrist is wholly unlike a modern ape wrist,” Lovejoy says. “Apes can’t bend their wrist backward, and it’s the bending of the wrist backward that allowed Ardi to walk on the palms.” In contrast, modern apes like gorillas, chimpanzees, and bonobos walk on their knuckles, an adaptation that was always assumed to be ancient.

Not everyone likes these surprises. “People are tightly invested in the chimp as a model for our ancestors,” White says. “The idea that the chimpanzee is basically a living missing link is deeply embedded in paleoanthropology. Ardi is not particularly chimpanzee-like, and we’ve gotten a lot of extreme pushback on that.”

University of Toronto paleoanthropologist David Begun is one of the skeptics. “Ardi lived at least 2½ or 3 million years after the split of chimps and humans,” Begun says. “The idea that this fossil tells us what the last common ancestor looks like is unfounded. Ardi is a spectacular discovery, but it may actually be an early side branch of hominids that is not even directly related to Lucy or humans. It’s naive to think every fossil you find is directly on the line leading to humans. And if we evolved from a monkeylike quadruped,” as Lovejoy’s analysis of Ardi’s hands suggests, “then all our extensive anatomy related to suspension and hanging would have evolved in parallel to the great apes. That’s possible but unlikely.”

Others question whether Ardi was truly a biped. William Jungers, the Stony Brook paleoanthropologist, and a group of colleagues spent several days in White’s lab last year. After examining the casts and digital images, Jungers decided that “Ardi was at best a facultative biped,” a creature that is capable of walking but somewhat inefficiently. That description also fits modern chimps, gibbons, and even capuchin monkeys. Jungers also questions the idea that male food-gathering turned A. ramidus into upright walkers. “Owen’s provisioning theory is untestable,” he says. “Striding bipedalism is obviously a dandy adaptation, but there is no shortage of colorful and plausible speculations as to why it occurred, from thermoregulation to sexual displays, from looking over tall grasses to wading through the water.”

Even the savanna hypothesis is not dead yet. Geochemist Thure Cerling of the University of Utah and seven other geologists and anthropologists recently questioned the ecological reconstruction from White’s team. Cerling reexamined the soil and tooth enamel data provided by White and concluded instead that Ardi lived in a bush-savanna, with less than a quarter of the area providing canopy cover. “I believe their data indicate a significant savanna influence,” Cerling says. White heatedly disagrees. The point is that Ardi’s particular habitat was woodland, he says, even if savanna was nearby.

Ardi has been hit with one potshot after another since she was unveiled to the world. Tim White’s answer to all the objections? “Some people have a hard time wrapping their minds around the anatomical mosaic that Ardi represents and its implications for human origins,” he says. “You’re able, as you take those sand grains away from the fossil, to see a creature that nobody has seen for the last four and a half million years.”

Flo, the Hobbit Person


“Big” has been the sine qua non of our success as humans. Relative to our ancestors and most of our primate cousins, we have large bodies, long limbs, and oversize brains. It seemed that only in our bigness could we stride out of Africa and across the planet. But maybe bigness was unnecessary. That is the message from a strange Indonesian fossil belonging to a previously unknown species of the human family: Homo floresiensis, the hobbit people. If Ardipithecus has utterly upset our notions about early human origins, the hobbits have altered our thinking about late human evolution by showing us, among other things, that small might be just as adept.

The ancestors of hobbits probably left the rift area of Africa (Ardi’s home) for Southeast Asia on foot about 2 million years ago, ultimately crossing treacherous ocean waters to land on the narrow, 230-mile-long Indonesian island of Flores. More amazing, hobbits seem to have survived into modern times alongside modern humans; they fashioned stone tools, hunted cooperatively, and even cooked with fire—all with a brain just one-third the size of that of a typical Homo sapiens adult.

The key hobbit skeleton is an adult female named LB1 for the place where it was found: a vast, open, sun-drenched limestone cave called Liang Bua, on Flores. In the tradition of giving notable hominid fossils familiar names, LB1 was nicknamed Flo. In addition to Flo, archaeologist Michael Morwood of the University of Wollongong in Australia found partial remains of as many as 14 other individuals in the same cave, all of them presented to the world in the journal Nature in 2004.

After the publication of Morwood’s article, the hobbit immediately became a scientific and media sensation. Flo is “one of the most complete fossils found anywhere until you get to true burials, like in Neanderthals and early modern humans,” says Jungers, who has been closely involved in Homo floresiensis research.

Flo and her species lived on Flores from about 90,000 years ago until about 14,000 years ago, when they were wiped out—perhaps by a volcanic eruption, or perhaps by competition with modern humans. If they did interact with humans, the hobbits may have inspired the local legends of a small, hairy, humanlike creature that some Flores natives call the Ebu Gogo (which loosely translates to the “grandparent who eats anything”), anthropologist Gregory Forth of the University of Alberta speculates.

Flo’s body shape was truly unexpected: she was just 3.3 feet tall and weighed about 60 pounds. She walked upright on large, flat feet unsuited to running and had a prominent brow, primitive teeth, no chin, short legs, and mysteriously long arms.

But it was her small head and brain that prompted the most fascination, fury, and often derision. Critics swiftly objected, saying that the specimen was not what it seemed. Some suggested Flo was a diseased modern human, related to pygmies but suffering from a condition like microcephaly, which causes the brain and head to be pathologically small. Others proposed she was simply a late form of Homo erectus, a tall, strong human ancestor that spread into Southeast Asia at least 1.5 million years ago but, in this instance, “dwarfed,” as sometimes happens to species isolated on an island. Or maybe hobbits had descended from Australopithecus afarensis—Lucy’s kin—since that species was a highly adaptable biped that spread over great masses of African land.

Morwood and other experts have dismissed each of these explanations in turn. “If our interpretation is right, we’re dealing with hominids that came out of Africa more than 1.8 million years ago, before the emergence of erectus,” Morwood says.

“The face and teeth are all wrong for australopiths,” adds Jungers. “As for dwarfing, if Homo erectus were its ancestor, it would have had to do more than dwarf; it would have had to re-evolve a more primitive body design from head to toe.”

Such arguments quickly established who Flo is not. Establishing exactly who she is has taken a lot longer, but slowly a consensus has emerged. Using CAT scans, digital imaging, statistical analysis, and computer reconstructions of the brain, anthropologists have determined that the little hobbit is most likely a normal, nondiseased human, albeit one with a very unusual form.

If so, homo floresiensis crushes our cherished notions about the key human trait of bigness, both in body and in brain. Jungers is not surprised. “Ask yourself why something would have to be large and evolve long hind limbs to get out of Africa in the first place,” he says. “That idea is crazy. We now think Flo’s ancestor was possibly an isolated remnant of an early human species that left Africa almost 2 million years ago. We know for certain that Flo’s ancestors were on Flores at least a million years ago, because we’ve found stone tools on the island that are that old.” Morwood has begun hunting for stone tools on the nearby and much larger Indonesian island of Sulawesi, where he hopes to find fossils ancestral to the Flores group.

One pressing question hanging over the discovery: How did Flo’s ancestors cross the deep, seemingly impassable waters from mainland Southeast Asia to their island home? Jungers speculates that a giant tsunami like the one that hit the region in 2004 swept them out to sea. Survivors clinging to trees could have been washed ashore on Sulawesi, only to migrate to nearby Flores after that. This early hominid species was small, and on Flores it might have become even smaller in response to the limited resources.

In their extreme focus on early human evolution in Africa, scientists may have missed major clues about our ancestry still buried in other parts of the world. That is another message from Flo. Asia in particular could be full of surprises, Jungers believes. “Perhaps hominids spread throughout the Southeast Asian archipelagos earlier and more extensively than we’ve realized. What about early man in the Indian subcontinent? In remote areas of China? There’s so much yet to be discovered,” he says.

Still, the biggest shock is the fact that Flo’s puny brain—no bigger than a chimpanzee’s—was so capable. “The hobbit discovery challenges the idea that intelligence is directly proportional to brain size,” Morwood says.

“We’re talking about a creature that was fairly well advanced,” adds archaeologist Carol Lentfer of the University of Queensland in Australia. “It was able to use stone tools to make other tools.” Relics unearthed in Flores indicate that the hobbits used large stones as hammers to knap and chip away at stone flakes, shaping them into cutting tools. The fabrication methods did not change significantly over time, however. Flakes were created with an average of about nine blows per tool from as far back as 100,000 years ago right up to the time of the hobbits’ extinction.

Those simplest of stone tools had far-ranging consequences. “With a chip of stone you had a hammer; you could crush things more effectively than with an elephant’s molar,” says anthropologist Rick Potts, director of the Smithsonian’s Human Origins Program. “With a sharp flake you could cut more effectively than with a carnivore’s canine. A whole world opened up for humanity with the use of simple stone tools.” Evidence of butchery and fire on animal bones found near hobbit remains suggests that these early humans enjoyed a good barbecue, usually of baby elephants, huge rats, and deadly Komodo dragons that they hunted and killed.

So how did Flo and her kin manage such big achievements using their little heads? Through resculpting the brain itself, it seems. The first evidence of this trend came as long ago as 1925, when South African anthropologist Raymond Dart published controversial findings in Nature on the first known australopithecus, called the Taung child. Dart argued that a brain structure called the lunate sulcus had been thrust back into a human position and that parts of the brain linked with higher cognitive functions had expanded. In his interpretation, Lucy’s relatives had already started reorganizing their little brains a long time ago.

Biological anthropologist Dean Falk did not expect to see anything like that when she got to work on models of hobbit skulls at Florida State University in Tallahassee. To create a virtual version of the hobbit brain, Falk’s colleague, engineer Kirk Smith, of the Mallinckrodt Institute of Radiology in St. Louis, used three-dimensional CAT scans that Morwood’s team had taken of its fossilized skull and braincase. Smith’s replica could be sliced, diced, twirled, and viewed in picturesque detail by Falk and her team. The scans show that the hobbit brain was uniquely folded and unusually complex. “It was beautiful, and the temporal lobes were really wide, which is an advanced feature,” says Falk, who has since moved her lab to the School for Advanced Research in Santa Fe, New Mexico. “At the very front were two enormous convolutions in an area associated with executive functions like planning ahead, again a complex feature.”

In the size-matters world of anthropology, these studies were a revelation. Flo’s brain was “globally reorganized in comparison with the brains of apes,” Jungers says. “That means that brain architecture and function are not always tightly constrained by size.”

Neanderthal, the Rival and Mate


“There’s a mantra in paleoanthropology,” Jungers says, “and it goes like this: We need more fossils.” Even with profound and important surprises like Ardi and Flo, scientists argue endlessly over their meaning. Because there are still so few complete fossils and so many gaps in the overall fossil record, interpretations abound. But now that increasingly powerful genomic technology can definitively identify a species from a fragment of bone or uncover Neanderthal genes embedded in the DNA of modern humans, there is less room for debate.

When paleontologists study fossils through bone shape alone, they can only broadly infer the relationship between two hominids, no matter how many fossils they collect. By going inside those bones and siphoning DNA—the genetic essence of long-dead ancestors—scientists can now use sequencing technology to make exact measurements of the similarities between groups. Using these techniques, says biologist Svante Pääbo of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, “you can determine, quantitatively, how much extinct human forms have contributed to the human form today.”

Hominid paleogenomics has advanced quickly since biologists started analyzing the Neanderthal genome in 1997. Today, DNA can be sequenced from bone bits that a few years ago were regarded as far too eroded and contaminated to yield relevant results. Although some bones are still out of reach (Ardi is far too old, and Flo’s bones too poorly preserved), others are yielding phenomenal insights. The latest gene-sequencing technologies developed by biotech companies 454 Life Sciences and Illumina can analyze several million DNA fragments at a time.

Using the same techniques being applied in medicine and forensics, Pääbo’s team is filling in the fossil record in formerly inconceivable ways. Last year they announced that modern humans outside Africa carry 1 to 4 percent Neanderthal genes. To reach this conclusion, Pääbo and his team spent years sequencing the complete genome of three Neanderthal bones from the Vindija Cave in Croatia and compared the results with the genomes of five modern humans from southern Africa, West Africa, Papua New Guinea, China, and Western Europe. They found that the Neanderthal genome shows more similarity with non-African modern humans throughout Europe and Asia than with African modern humans, suggesting that the gene flow between us and Neanderthals most likely occurred outside Africa as humans were en route to Europe, Asia, and New Guinea.

This pattern of gene flow means humans and Neanderthals must have mated at some point. The notion of such interspecies trysts had long been only a theory. “We thought if we did interbreed, it might have been when modern humans came to Europe, about 30,000 to 40,000 years ago,” Pääbo says. But when placed under genetic scrutiny, the bones tell a different tale.

The likeliest place for human-Neanderthal romance was the Middle East, where bones of both humans and Neanderthals have been found. “Modern humans appeared in the Middle East before 100,000 years ago, and Neanderthals were there at least 60,000 years ago—providing a likely 30,000-year window of opportunity for interbreeding before Neanderthals disappeared,” Pääbo says.

The more we learn about Neanderthals, the more we stand to learn about ourselves. By comparing our DNA with that of our big-boned relatives, Pääbo has already found spots in the modern human genome that appeared after we diverged from our Neanderthal cousins and evolved apart. These may be the very genes that enhanced our survival. Examples include modern human gene variants for cognitive development. Unique genes for skin morphology and physiology could be other examples. “We will ultimately catalog everything that has changed in our genome in the last 300,000 years since we shared a common ancestor with the Neanderthals,” Pääbo says.

And there is more to the story. In what could be the biggest triumph of paleogenomics, Pääbo and his colleagues have just extracted another evolutionary secret from especially sparse fossil remains. In 2010 the team discovered a new kind of human, cousins to Neanderthals called Denisovans, by sequencing DNA from a 50,000-year-old pinkie finger found in a high-altitude Siberian cave in Denisova. Based on the genetic evidence, the Denisovans lived in Asia from about 400,000 to 50,000 years ago and also interbred with the ancestors of modern-day humans—in this case, ones living in Asia. “We were examining mitochondrial DNA [found within the energy factories of cells and transmitted only by the mother] from that pinkie finger to find out if it was from a Neanderthal,” Pääbo says. “Instead it turned out to be something completely different.” Subsequent sequencing of the nuclear genome 
followed, revealing that the pinkie came from a previously unknown hominid group, similar to Neanderthals, that migrated east toward Asia while Neanderthals migrated west. Modern humans in New Guinea still carry genomes with nearly 5 percent 
Denisovan DNA.

Sequencing technology has advanced so far that, these days, fresh evolutionary insights do not necessarily require any fossils at all: Within our DNA, we modern humans provide a genomic window onto what came before. At the University of New Mexico in Albuquerque, genetic anthropologists Jeffrey Long, Keith Hunley, and Sarah Joyce used computers to analyze genetic data from 2,000 people in 100 modern human populations in Africa, Europe, Asia, Oceania, and the Americas. In much the same way that forensic scientists compare DNA samples to catch criminals, the New Mexico researchers compared 619 “microsatellite” positions on genomes, creating a digital evolutionary tree of the groups. There seemed to be two periods of interbreeding between modern and ancient humans (such as Neanderthals, perhaps Denisovans, and other large-brained hominid cousins).

Those two periods of interbreeding occurred after humans left Africa—first about 60,000 years ago in the eastern Mediterranean, and more recently about 45,000 years ago in eastern Asia. Offspring from the first interbreeding went on to migrate to Europe, Asia, and North America. The second mating in eastern Asia further altered the genetic makeup of people in New Guinea and possibly Australia. The findings mean our species did not branch off from all the others as sharply and irrevocably as we like to think, Long says. “This is giving us a window on human evolution where we see that there were these periods of flux and ebbs and flows in our gene pool.”

The Future of Our Past

With each discovery come new questions about our identity. “Over the next 10 years, projects like the Neanderthal genome will lead to a contentious debate about what it means to be human,” Rick Potts says. “To my mind, if you had to crystallize the human essence in one word, it would be adaptability. That goes all the way back to Lucy.” William Jungers calls members of Lucy’s genus, the australopiths, the “ultimate morphological generalists.” They accommodated huge amounts of climate change, were able to survive where there were lots of trees and very few trees, and expanded across Africa for millions of years in the face of overwhelming challenges. We are the bearers of Lucy’s legacy, able to live in igloos or space shuttles and interconnect instantaneously through technology that confounds our own imagination.


Ardi’s kin launched that adaptability with their complex, versatile anatomy. Flo’s relatives, swept unceremoniously onto an isolated island, adjusted and thrived. Even the Neanderthals, themselves doomed, managed to share space with Homo sapiens long enough to spread their genes. They still live in us. Yet out of all these species, only our species became a global success. As much as we resemble our relatives, there clearly is something special about us.



“Imagine a long-lived alien observing Earth through geologic time,” White says. “It would be pretty darn monotonous for a million years. We showed up anatomically about 150,000 years ago, and we are a very bizarre primate compared with all the others. Our feet don’t grasp. We walk only on two legs. Our braincase is massive. Our face and front teeth are almost infantile. Human breasts don’t cycle with lactation. They remain ‘enlarged’ throughout adulthood. We are different in every possible way.”


And then, over just the past two centuries or so, came a far more epochal change. The number of Homo sapiens on the planet skyrocketed. Our technology altered every aspect of the global ecosystem. Was all of this potential locked away in our brains and in our genes? Is it tethered to the same traits that allowed us to survive competing with and mating with our cousins, the Neanderthals and Denisovans? What will a few more generations bring?


 
“In art, in exploration, in technology, modern humans seem to have it over these other forms,” Pääbo says. “The million-dollar question seeks the genetic reason for some of that.” But he is cautious as well. “We are more numerous and spread over more parts of the globe, but we haven’t lived on this planet very long.” At the very least, we can expect our talent to render more astounding discoveries about our past—once as invisible and mute as a long-buried fossil in some deep cleft of rock but now exhumed and decoded by that unique brain that sets our species apart.

The Human Family Bush

In the new view, the human family tree looks more like a thickly branched bush. Ardipithecus ramidus, or "Ardi," one of the earliest human relatives, appears near the base. Likely human relatives predating, Ardi—Sahelanthropus tchadensis, Orrorin tugensis,and Ardipithecus kadabba—are known only from fragmentary remains; all probably walked upright. About 4 million years ago, Ardi may have given rise to the chimp-size australopithecines, including A. anamensis. Like Ardi, the australopithecines were adapted to life both in trees and on the ground. One of their descendants, the Paranthropus genus, had massive chewing muscles that enabled them to eat tough foods, including nuts. Two million years ago, the most recent of these rugged relatives (P. robustus and P. boisei) lived contemporaneously with the first members of the Homo genus—our own. Although today we are the only human species as recently as 30,000 years ago we shared the planet with at least three others: H. floresienis, H. neanderthalensis (Neanderthals), and the recently discovered Denisovans; evidence suggests we interbred with at least the latter two.