Homo floresiensis: Two Years Out

View Images

Two years ago this month, I was taken aback by some explosive news. A team of Indonesian and Australian scientists reported that they had discovered fossils of what they claimed was a new species of hominid. It lived on the island of Flores in Indonesia, it stood three feet tall, and it had a brain about the size of a chimp’s. Making the report particularly remarkable was the fact that this hominid, which the scientists dubbed Homo floresiensis, lived as recently as 18,000 years ago. I wrote up a post on the paper, and took note of some strong skepticism from some quarters. And since then, I’ve found myself devoting a number of posts to the new papers from the discoverers of Homo floresiensis, and the emerging responses from the skeptics–so many that I gave them their own category. Recently there’s been so much stuff coming out pro and con that I have had to skip a couple opportunities to blog on Homo floresiensis–mainly because I’ve been frantically deep in the first draft of my current book on a very different topic: Escherichia coli. (I assume Homo floresiensis carried Escherichia coli in its gut, but the overlap stops there.)

Fortunately the first draft is now done, so I can let my mind drift back from the microbial world, to Homo floresiensis. And it just so happens that a big new paper has come out today which is a good topic on which to blog.


This paper, published in the Anatomical Record, actually builds on a much shorter one that appeared a few months ago in Science. To keep the strands of this story from tangling up with each other, let me lay out a timeline. (If you feel I’ve left something important out, remind me in the comments and I may insert it below…)

October 2004: Homo floresiensis makes its debut. The bones include only one brain-case, dubbed LB1. Along with the bones are lots of stone tools, raising the question of whether a small-brained hominid could have made or used them. Flores is also home to dwarf elephants, which illustrate the fact that many mammals evolve to smaller sizes on islands. Perhaps Homo floresiensis evolved from a bigger hominid. The best candidate, according to the authors, is Homo erectus, which spread from Africa about 1.8 million years ago and existed in southeast Asia perhaps as late as 50,000 years ago. Homo erectus was tall, could make simple tools, and had a brain about two-thirds the size our own. One line of evidence that may support this claim is the presence of stone tools on Flores dating back 840,000 years ago. They might have been left by Homo erectus migrants, whose descendants later evolved to tiny proportions.

November 2004: Things get weird. A prominent Indonesian paleoanthropologist named Teuku Jacob gets hold of the Flores bones and studies them for himself. He tells the press that Homo floresiensis is not a separate species, but a human pygmy, perhaps with a birth defect called microcephaly that causes small brains. (This is a line of argument taken by other skeptics.) The discoverers of the fossils cry foul, and three months later, when the bones are returned, they complain that some bones have been permanently damaged.

March 2005: Homo floresiensis gets a brain scan. The fossil discoverers team up with Dean Falk, a hominid brain expert, to give the LB1 brain case a CT scan. They reconstruct its brain and compare it to the brain of a human microcephalic, from a skull kept at the American Museum of Natural History. They argue that the brain is significantly different from the microcephalic, and most resembles that of Homo erectus.

June 2005: The site where Homo floresiensis was found is sealed off from any further investigation, reportedly due to the conflicts among the rival scientists. (I have not heard if it has been opened since then.)

October 2005: More bones. The discoverers of Homo floresiensis publish descriptions of additional fossils. These bones, including material from the lower jaw, arms, and legs, show strong similarities to the original fossils. They lack a chin, they have long arms relative to their legs, their teeth have some odd double roots, and so on. What’s more, they come from several ages. The bones that the authors assign to Homo floresiensis now range from 97,000 to 12,000 years. If LB1’s small skull belonged to a human with a genetic disorder, then why would all of these other individuals show so many similarities? Robert Martin, a primatologist at the Field Museum in Chicago who has expressed skepticism about Homo floresiensis, says he’s writing up a critical paper.

October 2005: After more bones come more brains. A few days after the new fossil paper is published, a German team of scientists publishes a comparison of LB1 to skulls from a different sample of microcephalics. The scientists argue that the brains are quite similar. They point out that microcephalics are quite variable in the shape and size of their brains, and say it’s premature to rule out the possibility that LB1 was microcephalic too. Dean Falk and her colleagues come right back at the Germans, arguing that they tipped the brains at the wrong angle before comparing them. If all the brains are lined up at the same angle, they look less similar.

May 2006: More skeptics weigh in. After a long winter without more news to take in, Robert Martin publishes his first attack on Homo floresiensis, teaming up with experts in other relevant areas such as William Dobyns, an expert on microcephaly at the University of Chicago. They present evidence that the microcephalic Falk chose was a nineteenth-century child named Jakob Moegele. An adult would have been a better comparison. The scientists present a couple sketches which they claim are more similar to LB1. The scientists also argue that LB1 is too small to be the result of evolutionary dwarfing. If Homo floresiensis followed the same trend that other mammals have, it should have had a much bigger brain relative to its body. Falk and her colleagues responded by asking how they could judge sketches (as opposed to detailed scans). They also hinted that perhaps Homo floresiensis might have evolved from an older, smaller branch of hominids.

June 2006: The discoverers of Homo floresiensis now publish new details on the stone tools. They argue that the tools from 840,000 years ago and the more recent ones found alongside the fossils probably represent a continuous technology made by the same lineage of hominids. To say that a small-brained hominid could not make such impressive tools is an assumption cloaked as a conclusion.

June 2006: A new idea emerges: Homo floresiensis is a healthy human. Gary Richards of the University of California at Berkeley argues that humans could have settled on the island of Flores and rapidly evolved into pygmies with small brains. Many of the traits that seem to set Homo floresiensis apart are also found sprinkled among living humans–particularly among pygmies.

August 2006: Teuku Jacob and an international team of colleagues take on Homo floresiensis. Their criticisms come from many different directions. They complain, for example, that the discoverers compared the hominid to European individuals. The proper comparison is between Homo floresiensis to people from southeast Asia and the Pacific (Australomelanesians)–especially pygmy Australomelanesians. They would have discovered many traits from Homo floresiensis in modern humans that they claimed were not found in our species. Jacob’s team also presents evidence that LB1 had an asymmetrical face–which is sometimes seen in microcephaly. The original discoverers hit back, telling reporters that the asymmetry might have come after death, as the skull was squeezed under sediment.

October 2006: A second vote for Homo floresiensis. The Journal of Human Evolution publishes an analysis of the Flores bones from a different team of scientists from Australia. I was too busy to write this one up, so let me just quote from the abstract now:

“We explore the affinities of LB1 using cranial and postcranial metric and non-metric analyses. LB1 is compared to early Homo, two microcephalic humans, a ‘pygmoid’ excavated from another cave on Flores, H. sapiens (including African pygmies and Andaman Islanders), Australopithecus, and Paranthropus. Based on these comparisons, we conclude that it is unlikely that LB1 is a microcephalic human, and it cannot be attributed to any known species. Its attribution to a new species, Homo floresiensis, is supported.”

Today: The Anatomical Record publishes a 23-page report from Martin, Dobyns, and company (Anatomical Record (DOI: 10.1002/ar.a.20394). This is not a new paper, so much as the paper Martin et al probably wanted to publish in the first place, rather than the clipped “Technical Comment” that is all Science will allow for such matters. So it’s a mix of points raised before, more evidence marshalled in support of those points, and some new information as well. It’s a very detailed attack on Homo floresiensis, and I don’t know of anything coming down the pike that will be more substantial. So if the original discoverers decide to write a detailed rebuttal of all the recent papers that have come out, this is one I’d imagine they’d pay a lot of attention to.

One subject that is new is the matter of the tools. Martin et al argue that the tools from 18,000 years ago are not like any simple tool linked to Homo erectus. They are more sophisticated, and have only been associated before with Homo sapiens and Neanderthals. The preservation of the tools in the Liang Bua cave, where the fossils were found, suggests to the scientists modern humans coming back again and again to the cave after they arrived on Flores.

The paper also presents new information on microcephalics. The researchers scanned Jakob Moegele’s skull (the original, not the cast that Falk used) to generate an endocast of his brain. They then carried out some of the same statistical studies on the shape of the LB1 brain. They compared LB1 to Jakob, as well as to two microcephalics who lived to adulthood, and a range of hominids and apes. They conclude several important things from this study. One is that microcephalics cover a huge range of shapes. Jakob’s brain was very different from that of the adult microcephalics. But those adults had brains that in some ways resembled Homo floresiensis. “LB1 is not clearly distinct from all modern human microcephalics,” the authors write. (Martin et al cast doubt on the recent Journal of Human Evolution paper, pointing out that the two microcephalic skulls studied there were over two thousand years old. One probably died before adulthood.)

Martin et al also argue (like Richards) that the other odd traits of Homo floresiensis are not as odd as the original discoverers claimed. For example, the discoverers pointed out the massive teeth in LB1’s jaws. But teeth are like brains: they don’t follow quite the same path as a shrinking body. If you compare primate species, a primate with a smaller body has teeth that are not as proportionately small. And one of the microcephalics Martin et al studied turned out to have similarly big teeth as well.

As we come up on the second anniversary of the initial announcement of Homo floresiensis, we’re in a strange spot. Microcephaly turns out to be a very peculiar condition that makes it very hard to distinguish humans from a possible species of very small hominids. Many different genes can give rise to the same conditions, producing different shapes to the brain, as well as different changes to other parts of the body. Scientists actually have a lot to learn about microcephaly–for one thing, many studies rely on remains in museum collections, which almost never included anything below the skull. At this point it’s not even clear if discovering more tiny hominids on Flores would make the case for a separate species. Under some conditions, it might be possible that a small population of islanders had a high proportion of microcephaly-triggering genes floating about. But that may be moot if nobody’s actually digging in the Liang Bua cave.

Now, if Homo floresiensis is the result of evolutionary dwarfing, then perhaps the debate might be advanced a bit if someone could find hominid fossils on other islands around southeast Asia that have also followed the Homo floresiensis path. On the other hand, if Homo floresiensis descended from ancient small hominids, those hominids would have to have come to Flores from Africa, where the oldest hominids are found. That’s a long path, with plenty of opportunities for fossils to be formed along the way. Whether anyone finds them is another question. Finally, there have been rumors of DNA from Liang Bua, but no published reports. So there’s another avenue of hope. I have no idea when I will be writing the next Homo floresiensis post, but I can only hope it continues to be interesting.

Update: Here’s the new paper’s link.