According to J.R.R. Tolkien, “The beginning of hobbits lies far back in the Elder Days that are now lost and forgotten.” But what of the origins of the puzzling, meter-tall hominid that walked our own Earth a mere 17,000 years ago? Among scientists, three hypotheses are still in play for how this creature came to live on the Indonesian island of Flores. One idea is that members of Homo erectus landed on the island and the species' brain and body shrank over time in an evolutionary process called insular dwarfing. Another hypothesis suggests that the hobbit’s ancestor was even more primitive—and therefore smaller brained—than H. erectus. And some researchers still argue that the most complete hobbit specimen and associated skull are merely those of a diseased modern human. The first and second hypotheses get independent support in this week’s issue of Nature on topics ranging from feet to hippopotamuses. (Skeptics continue to argue for pathology, although no recent papers from them have appeared.)

In the first Nature paper, William Jungers of Stony Brook University in New York and co-authors describe the surprisingly primitive feet of the creature they call Homo floresiensis; this work was also presented at a meeting last year (see Science, 25 April 2008, p. 433). Jungers found that H. floresiensis had big feet compared to its short legs (see foot and tibia, left), a short big toe and flat feet, and was probably a poor runner who walked with a high-stepping gait. Jungers says the foot and other aspects of the skeleton hark back to a very early Homo species or even the australopithecines who lived 2 million or 3 million years ago in Africa. Some of these features are more primitive than those of H. erectus, suggesting either a more ancient ancestor or unusual evolutionary reversals to primitive traits, he says. “If it’s an insular dwarf, then it’s got reversals from head to toe,” Jungers said at a recent meeting.

Not so fast, say Eleanor Weston and Adrian Lister of the Natural History Museum in London, authors of the second Nature paper. The problem with the island-dwarfing hypothesis has been H. floresiensis's tiny brain of about 417 cubic centimeters, roughly the size of a chimpanzee’s. Past studies have suggested that when mammals shrink in body size, brain size is only moderately reduced. Weston challenges that view with a hefty chunk of data from hippos. She studied the brains and sizes of fossils of two species of dwarf hippo on the island of Madagascar and compared them to their normal-sized hippo ancestor on the mainland (see ancestor, above left, and a dwarf descendant shown at roughly the same scale). The brains of the Malagasy hippos were quite small—about 30% smaller than predicted if the mainland ancestor had shrunk proportionately. “The hippos are reducing their brain size in a way we never predicted,” says Weston. “It’s mechanistically possible to reduce brain size much more than we thought.”

The pair then turned to H. floresiensis and found that if the Flores skull followed the hippo scaling rules, it could be descended from several smaller H. erectus individuals, including those from Dmanisi, Georgia, and from Africa. “You don’t need to rule out Homo erectus as an ancestor,” said Weston. She theorizes that in addition to reducing size, island dwarfing could also lead to the appearance of primitive traits. There are examples of young mammals retaining primitive traits, she says, and adult insular dwarfs may incorporate the juvenile stages of their ancestors and so end up looking primitive. She urged paleoanthropologists to explore whether H. floresiensis's primitive traits are found in early human development. “When you scale things, you get correlated shape variation, and something small might look primitive,” she said.

Jungers calls Weston’s work a “nice, provocative paper” but points out that young mammals have relatively big heads and brains—not at all like those of H. floresiensis. Others remain skeptical of the paper’s methodology, even after delving deep into the Supplemental Information. For example, says John de Vos of Naturalis Museum in Leiden, Netherlands, in comparing the hippo’s brain and body sizes, Weston and Lister used skull volume as a proxy for body size because it was hard to find associated hippo bones. But he was not convinced that the proxy relation holds. Robert D. Martin of the Field Museum in Chicago, Illinois, who has argued that H. floresiensis's brain is too small to have evolved through dwarfing, was also skeptical about whether skull volume can be used to represent body size, noting that if not, the paper’s findings are an “an artifact.” Even if the findings hold up, Martin says, hippos may be a special case, one not necessarily transferable to hobbits.

—Elizabeth Culotta

PHOTO CREDITS: Foot/tibia: William Jungers/ARKENAS; >Hippos: E. Weston, Natural History Museum. Specimens from University Museum of Zoology, Cambridge, U.K.

... Long Island, that is, where researchers studying the puzzling little people of the Indonesian island of Flores gathered for a public symposium at Stony Brook University on 21 April. The biggest news was archaeologist Mark Moore's detailed report on the stone tools throughout the Liang Bua, the cave where hobbit bones were found. See ScienceNOW and this Friday's Science for more on Moore’s surprising conclusion: that the modern humans who arrived 11,000 years ago at the cave made tools in the same way as hobbits, who lived there from 95,000 years ago to perhaps 17,000 years ago. Moore, of the University of New England in Armidale, Australia, even suggested the possibility of contact between the species, with Homo sapiens learning from hobbits. His work is also in press at the Journal of Human Evolution. (Pictured above, from left: Mark Moore, Mike Morwood, Susan Larson, William Jungers, and Thomas Sutikna with a cast of the hobbit and a modern human skull and limb bones for comparison.)

The meeting was a rare chance for U.S. researchers to hear from the team that discovered the hobbits, which they officially call H. floresiensis. Lead excavator Thomas Sutikna of the National Research and Development Centre for Archaeology in Jakarta and Mike Morwood, now of the University of Wollongong in Australia, flew across the globe for the meeting, which gathered only those researchers who already accept H. floresiensis as a new species. The lingering skeptics, who think the fossilized bones represent a diseased modern human, were not invited to give talks. (In 2007, a meeting in Indonesia chiefly featured critics.)

The public talks at the Long Island meeting reprised some previously published information. For example, researchers detailed features of the shoulder, wrist, and brain, suggesting that the 17,000-year-old hobbit skeleton resembled much older African hominids rather than a diseased modern human.

But bits of important news emerged: Hand specialist Matthew Tocheri of the Smithsonian Institution in Washington, D.C., reported that last summer he went through hundreds of bags of bone fragments recovered from Liang Bua, including those from a layer that has already yielded a second hobbit jawbone, known as specimen LB6. Tocheri was seeking one of the four fingernail-sized bones of the wrist, and he got lucky: He found one, called the capitate, presumably from LB6. The bone has the same peculiar and primitive configuration seen in the capitate of the main skeleton, suggesting that at least two individuals from Liang Bua have this oddly shaped wrist bone. Two individuals with such a primitive trait makes the disease argument less likely, Tocheri said.

A couple of relatively neutral observers praised the detailed skeletal analyses presented at the meeting but retained skepticism on some points. Archaeologist Paul Mellars of the University of Cambridge in the United Kingdom, who is at Stony Brook for the semester, said he started the day thinking there had been a bit too much fuss about the little people but was impressed by the data by the day's end—although he remains skeptical that humans learned from hobbits. Meanwhile, famed fossil hunter Richard Leakey, chair of the Turkana Basin Institute at Stony Brook, which hosted the conference, described himself as originally "needing a nudge" to be convinced that the hobbit was a new species. One of the few present with an African perspective, he remained quite doubtful about some presenters' ideas that H. floresiensis represents an extremely early migration out of Africa but said he was impressed by several talks and satisfied that the skeleton was not diseased.

Also at the meeting, researchers unveiled a model of the hobbit skeleton, made by materials scientists at Stony Brook who transformed CT scans of the original fossils into three-dimensional casts. The new model will be given to the Indonesian National Research and Development Centre for Archaeology for display, according to symposium organizer William Jungers of Stony Brook.

And finally, Sutikna and Morwood reported that they are still looking for more bones: This summer, they will excavate again at Liang Bua and also plan digs at other sites on Flores. 

—Elizabeth Culotta

PHOTO CREDIT: John Griffin/SBU