In October 2004 Australian and Indonesian announced they had discovered a three-foot tall species of hominid, Homo floresiensis, that was still alive no earlier than18,000 years ago. As I’ve detailed in previous posts, this claim has inspired a lot of debate, much which revolves around whether the fossils, found on the Indonesian island of Flores do in fact represent a new species, or whether they were human pygmies. This week a new study was published in the journal Biology Letters (link to come) that puts this debate in the proper evolutionary frame. The paper is not about hominids, however, but about bats.

Before I get to the bats, let me dwell a little longer on these Pleistocene hobbits. A great deal of the controversy has focused on the one Homo floresiensis skull found so far, which held a brain less than a third the size of a human’s and about the size of a chimpanzee’s. If Homo floresiensis really does represent a separate species, then its ancestors may have undergone a drastic evolution, which not only shrank their bodies but also their brains. One hypothesis for the origin of Homo floresiensis holds that it off from another species of hominid, Homo erectus, which arrived in southeast Asia 1.8 million years ago and may have been present there as recently as 30,000 years ago. Homo erectus was already about as tall as our own species is today, and had brains that were about three-quarters the size of ours.

Skeptics find this possibility implausible, arguing that it’s more likely this individual was just a pygmy human with some genetic defect. As far as I can tell, this skepticism about shrinking hominid brains flows from two sources.

One is the fact that digs on Flores have yielded some sophisticated stone tools and other clues that the hominids of Flores—human or otherwise—were able to hunt. Some people wonder whether it would be possible for a hominid with a chimp-size brain to use such tools, since the rise of tool use in hominids roughly coincides with the rise in brain size. It’s a fair question, since chimpanzees today can’t make the sorts of stone tools found on Flores. But it’s not any sort of slam-dunk refutation of the claim that Homo floresiensis were a separate species. First, consider the fact that the first signs of hominid tool use, 2.6 million years ago, came at a time when hominids still had brains barely bigger than a chimp’s. Second, size isn’t everything. Tool use may also depend on how a brain is wired, not just how much data-processing power it has. It doesn’t seem absurd to argue that as Homo floresiensis evolved a smaller brain, it retained the circuitry that made tool use possible. At least it’s a hypothesis worth testing.

The other source of skepticism, which I mentioned in my last post, is a vague sense that when it comes to hominid brains, evolution cannot run in reverse. It’s certainly true that if you draw a graph of hominid brain size over time, it has climbed to spectacular heights. Scientists prefer to chart brain evolution not simply by its raw increase in weight, but in how large the brain becomes in proportion to the rest of the body. For a mammal our size, we humans have a brain about seven times you’d predict. A great deal of research has gone into charting how brains get bigger over the course of evolution—not just in our immediate hominid ancestors, but over the past 200 million years of mammal evolution. The ability of our species to thrive so spectacularly seems to mainly depend on our extraordinary brains. Given that their size is one thing that makes them so extraordinary—and given that they’ve been increasing for so long—the notion of a shrinking hominid brain can seem absurd.

The discoverers of Homo floresiensis have pointed out reversals do happen. They point to how many species become dwarfs when they arrive on isolated islands. Elephants, deer, buffalo, and other species have shrunk over the course of just a few thousand years. It’s not entirely clear why this happens, but scientists suspect that being small is an advantage on an island with limited resources, and when animals arrive on an island without a lot of predators, there’s no longer a defensive advantage to being big. In some cases, these island dwarfs have evolved a simpler nervous system. So, the argument goes, Homo floresiensis is simply a hominid that happened to get washed up on a remote island and proceeded to evolve according to the rule of islands.

This argument may give you the impression that the evolution of smaller brains is just a digression from the main story of progress. Sure, a few hominids wind up on desert islands and evolve small brains, but back on the mainland, the hominid brain marches on towards our own spectacular size. In fact, it now appears that shrinking brains are a much more general feature of mammal evolution. And this is where we get to the bats.

Bats evolved about 50 million years ago. The first bats could fly and listen to the echoes of their shrieks to find prey, a radar-like technique called echolocation. These two adaptations allowed them to become efficient nocturnal airborne predators, taking advantage of a niche that may have been empty at that time. (Owls seem to have diversified at around the same point in history.) The result was a staggering evolutionary success, with bats now making up 20% of all mammal species on Earth. Bats obviously depend on their brains. They need to be able to process the complex information that they get from echolocation, and they need to be able to control their membranous wings. So you might think that bat evolution has been dominated by a steady expansion of their brains.

But as much as we may value the brain, it is just another organ. If the brain becomes bigger, an animal has to dedicate more energy to it and has less to supply to other parts of the body. This evolutionary trade-off has produced a lot of the diversity of life we see today—including even the size of beetle horns, a subject I blogged on a few days ago. And brains are particularly costly, requiring twelve times more calories ounce for ounce than muscle. It’s not easy to gauge the effect of this trade-off in our own lineage, because only 20 or so hominid species are known from the past six million years. But with so many bat species alive today, it is possible to see major trends in brain evolution by comparing them.

Kamran Safi, a biologist at Zurich University, and his colleagues compared 104 species of bats, noting their brain size, the shape of their body, and the ways in which they hunted. (Some bats specialize in hunting in open spaces, for example, while others can weave their way through forest foliage.) They then extrapolated back along the bat family tree to calculate how big the brain of the common ancestor of living bats was. And from their, they then moved forward through evolution, seeing whether there was a directional trend towards bigger brains.

They didn’t. It turns out that the first bats probably had brains that would be considered average for a living bat. Some bats have bigger brains, and some have smaller ones. Safi and his colleagues looked for other factors that had changed along with brain size in different lineages. They found that bats that had specialized for hunting in tight spaces evolved broad, large wings that provided them with agile maneuverability but also use up a lot of energy. They also tended to evolve bigger brains. By contrast, the bats that adapted to open spaces evolved narrow, small wings that didn’t demand much energy but also didn’t provide much maneuverability. These bats evolved smaller brains. This trend was especially strong in bats that hunt insects, as opposed to ones that have shifted to eating fruit or flowers. When bats evolved in ecological niches that demand a lot of brain power to control their wings, they evolved bigger brains. But when they could afford to slim their brains, they did—thus saving themselves the cost of fueling this hungry organ. These bats with shrunken brainsn were not defective, nor were they even rare flukes sequestered on some tiny island. They could still fly and hunt with perfectly respectable skill. They simply adapted to their surroundings.

Safi and his colleagues conclude that mammal brains may shrink thanks to many evolutionary forces, including a species’s diet, social system, or the length of its pregnancy. “A reduction in brain size should be a general property of evolution,” they write, adding that “The assumption that larger brains are derived [a new development in a lineage] is probably associated with the quest to explain why humans have large brains.”

The question of whether Homo floresiensis really did evolve a shrunken brain remains an open one. But if it does prove to be the case, we shouldn’t consider it a bizarre fluke. The bats are beginning to fly here in Connecticut, and when I see them flit across a twilight sky this summer, I’ll think of them as flying hobbits.

So let’s recap: It’s been almost eight months now since scientists announced the discovery of Homo floresiensis, the diminutive people that some claim belong to a new branch of hominid evolution and skeptics claim were just small humans. We seem to have entered a lull in the flow of new scientific information about Homo floresiensis. The last thing we heard from its discoverers came in March, when they published scans of the Homo floresiensis braincase, which bolstered their case that the skull they found didn’t happen to belong to someone with a birth defect. The skeptics have made various noises about evidence that the fossils are indeed pathological, and thus can’t be the basis for recognizing a new species. They have told reporters about their visits to pygmies who live near the fossil site on the Indonesian island of Flores. But they have yet to publish any of this in a scientific journal, where their claims could be put to some serious scrutiny. For example, you can’t refute the claim that the fossils are a separate hominid species by showing that living pygmies on Flores are very short. You also have to deal with the odd body proportions of Homo floresiensis, such as its long arms. Perhaps these are pathological too, but no one has gone on the scientific record yet.

For now Homo floresiensis junkies like myself have to content ourselves with scraps: the various details of the nasty battles between the discoverers of the fossils and their foe, Teuku Jacob, grand old man of Indonesian anthrolopology and lead skeptic. An article in today’s Los Angeles Times, offers the latest overview of the squabbles. If you are new these misadventures, it’s pretty good way to catch up. For those who keep up on this stuff, I see a couple interesting new tidbits.

1. A lot of Teuku Jacob's arguments against this being a new species seem wacky to me, at least as they've been presented in the press. In the LA Times, he "argues that evolution cannot 'go backward' and produce a human with a smaller brain." Perhaps Jacob will eventually make this case at length in a scientific paper, but for now I'd just say that there's no Law of the Perpetually Increasing Brain that I'm familiar with. In fact, the mammal brain is surprisingly malleable over the course of evolution. This afternoon I will try to write up a post on a new study that makes this clear. UPDATE 6/16: Read it here.

2. Teuku Jacob took possession over the bones for a few months, and when he returned them, their discoverers claimed the delicate fossils were damaged. The damages included what appeared to be an attempt to reconstruct the jaw.

In the LA Times piece, one of the co-authors of the original Homo floresiensis report accused Jacob of trying to make the skull look more like a member of our own species (the other hominid species that lived in Indonesia, Homo erectus, had a weaker jaw).

For the first time that I’m aware of, Jacob admits that he was trying to “improve” the skull. "We tried to improve some of the things," he acknowledged. "We didn't damage any bones. Actually, we improved some." Improve, or match your preconceptions?

3. As if this wasn’t bad enough, the controversy has now resulted in a complete halt to digging in the cave where the original fossils were found. Apparently the team that discovered the fossils didn’t get the proper permits from the Indonesian Institute of Science, although they believed they had. Now the Institute has decided that digging should stop, so that the dispute won’t get worse. While I can only judge this decision from a brief summary in a news article, the logic behind it baffles me. I doubt that forcing scientists to cool their heels while clues to what could be one of the most important discoveries in human evolution wait to be found on Flores will put them in a more pleasant mood. What’s more, digging in the cave could yield evidence that can settle this dispute once and for all—such as DNA, the odds of finding may have gone up thanks to the invention of new methods for culling it from the environment.

It's frustrating to know that we could be enjoying a scientific feast, when all that's on the menu for the foreseeable future are scraps like these.

The feud over Homo floresiensis, the little people of Indonesia, centers on whether they were an extinct diminutive species that evolved from some ancient hominid, such as Homo erectus, or whether they were just pygmy humans, perhaps suffering from some disease. The leading skeptic, paleoanthropologist Teuku Jacob, has claimed that there are pygmies living not far from where the fossils were found, on the island of Flores. I came across a short item at Japan Today about a scientific expedition to study the pygmies, which was based on an article in Kompas, an Indonesian publication. The original article is here, and my intrepid brother Ben, expert on Indonesian anthropology (cultural, not paleo-), did an on-the-fly translation for me, which I'll run below. The team got back from Flores on April 25. While there, they went to a village called Rampasasa, made up of 77 families. About 80% of the people were pygmies. They measured 10 people who were a bit taller, with a height of 155 cm and 2 measuring 160 cm. Homo floresiensis was 130 cm. The researchers claim that these tall villagers got some extra height from having married non-pygmies from surrounding villages.

I imagine that we'll be hearing something more official about the grandly-named Rampasasa Pygmy Somatology Expedition in a couple months. I wonder if they'll have something more than height measurements to offer--just because living pygmies are close to H. floresiensis doesn't seem terribly compelling, since it's my impression that height changes can evolve relatively quickly in humans. (I can't find a paper to back up this recollection at the moment, I confess.)

Update at 3:50 PM: Apologies for the various typos, dead links, and missing facts in the first version of this post. I blame it on my Mac upgrade to Tiger today.

So here's the article...

The "Pygmy" Community of Flores

The existence of a community of pygmy people in the Manggarai Regency of  Flores, East Nusatenggara, is quite interesting but also quite mysterious.  In the context of the archaeological discovery of the prehistoric human skeleton from Liang Bua in Flores -- which has been published widely as belonging to the species named Homo floresiensis -- the existence of the pygmy community in the village of Rampasasa, Waemulu region, Waeriri subdistrict, could possibly shatter all previous arguments.

"The existence of the pygmy community there is quite interesting and also quite surprising. For many years, experts from various corners of the world have only had the chance to see their footprints, but it turns out we can now find them living in a society.  This means that for hundreds or maybe even thousands of years, this pygmy community has remained settled in that place without ever moving around," said Prof Teuku Jacob, emeritus professor at Gadjah Mada University.

Jacob, who also leads the Bio- and Paleo-Anthropology Laboratory at Gadjah Mada, further explained, "Pygmy people have indeed been reported as existing in the Andaman Islands and New Guinea, but only a few remain and it is difficult to find them because they live in dispersed conditions. Now we can find them living together in one village."

Since the 1920s, the East Nusatenggara region has been an object of interest for anthropologists, especially those from Holland, after seeing evidence that the residents there have rather short body height.  The results of the 1929 Biljmer study indicate that more than 50 percent of the residents of the region have body height of about 155 to 163 cm. Besides that, in Flores there have long circulated folk tales about short people with darkly colored skin (Negritos) who live in the hills, hiding in caves.

Dr. Theodore Verhoeven, pastor at the Ledalero Maumere Seminary, conjectured in 1958 that these short people were a Proto-Negrito community. This term refers to the Schebesta study in the Andamans, remote areas of Borneo (Kalimantan), and also the southern Philippines.

According to Teuku Jacob, if the height of the Negritos is roughly between 155 and 163 cm, they would be called pygmoid.  But if the Rampasasa people are true pygmies their height would have to be less than 145 cm for adult males and 135 cm for adult females.  The maximum weight would be 40 kg for males and 30 kg for females.

Pygmies are indeed different from dwarfs.  This is because the term dwarf indicates a small body with proportions that are out of order.  Pygmies, meanwhile, have small bodies that are proportional.

Since last year, the team working under the leadership of Prof. RP Soejono and Dr. MJ Morwood conducting an excavation in Liang Bua, Flores, has found human skeletons with an approximate height of 130 cm and with brains about a third of the size of modern humans. This discovery was later claimed to be a new species of humans called Homo floresiensis (Flores Man).

Worwood, an expert in cave paintings from Australia, in fact called the results of the discovery "hobbits" in a popular fashion, a group of pygmy people like those in the film Lord of the Rings.  The picture of miniature Flores Man then appeared as a major report in the April 2005 edition of National Geographic.

The above claim about the discovery of a new species was rejected by a number of experts.  Etty Indriati, a PhD from Gadjah Mada, called it a baseless tale.  How could there be a new species from the discovery of just one skeleton which in fact was misidentified?  They said it was a female while from the dental structure it was clear that it was a male, and also a modern one.

"What is more unreasonable, it is not possible that a brain that has already developed as Homo sapiens could then become small and develop into a new species, left behind as prehistoric remains," she explained.

Indeed, for mammals trapped in remote islands for hundreds of years -- and with insufficient food to eat -- bodies will become smaller as an adaptation to the environment. "But, for humans, their menu is not just one type of food.  Despite being isolated, they will try to find other types of food, so their bodies do not become small," she added.

Teuku Jacob explained, "The pygmy people of Flores are not a prehistoric race. Our team has successfully found a community of pygmies living in the modern world."  What is even more ironic, the community of pygmies mentioned by Jacob is only about 1 kilometer from Liang Bua, the dwelling place of the species given the name Homo floresiensis by Worwood.

Koeshardjono, an expert in biology who was the first to announce the existence of a pygmy community in Flores, stated, "This expedition was named the Rampasasa Pygmy Somatology Expedition.  This is because the pygmy community of about 77 families resides entirely in the village of Rampasasa, Waemulu region, Waeriri subdistrict, Manggarai regency, south Flores."

The results of the team of physical anthropologists led by Teuku Jacob recorded that 80 percent of the residents of Rampasasa are classified as pygmy.  The provisional findings indicate that there are 10 people with a height of 155 cm and two people with a height of 160 cm.  It turns out that their body size is relatively tall because of marriage with residents outside of the village.  The team of researchers from Gadjah Mada has been in Rampasasa since April 18 and returns to Yogya Sunday night (April 25).

I've been catching up on my online reading, and a couple days ago John Hawks offered this tantalizing hint that Homo floresiensis a k a the Hobbit may be a pathological specimen. Such claims have been made before based on the small skull of the hominid, but they've been pretty powerfully rebutted. But Hawks is claiming that the rest of the skeleton is sickly. He seems to be basing this contention on having seen the bones, and on research by others that will be coming out soon. Now, normally I wouldn't put much stock in this sort of off-hand remark, but Hawks has been so good on his blog that I have to say I'm intrigued. Adding to my interest is the fact that he now retracts his suggestion that the Hobbit represented a very early migration out of Africa by australopithecines. Stay tuned.

At 1 p.m. today I listened by phone to a press conference in Washington where scientists presented the first good look inside a Hobbit's head. The view is fascinating. While it may help clear up some mysteries, it seems to throw others wide open.

Last October, a team of Australian scientists declared that they had found a new species of hominid that lived as recently as 12,000 years ago. It was short--maybe three and a half feet tall--and had a brain they estimated to be about the size of a chimp's. Its bones were found along with stone tools, suggesting that it made good use of its scant grey matter. The fossils of this remarkable hominid were discovered in a cave on the island of Flores, which gave the hominid its name: Homo floresiensis.

As soon as the news broke of the discovery, some researchers expressed grave doubts. They suggested that H. floresiensis was actually just a group of human pygmies. The fossils discovered on Flores include only a single skull, and these skeptics suggested that its small size might be the result of a genetic defect known as microcephaly. If H. floresiensis's discoverers had found another skull instead, it would likely have displayed the sort of shape you'd find on a living pygmy human.

This scientific debate quickly got eclipsed by an ugly tussle over the bones. A skeptical grand old man of Indonesian paleoanthropology, Teuku Jacob, wound up with the fossils for four months, during which time he arranged for other scientists to examine it and for some bone to be sent to Germany for DNA testing. Now he's surrendered the fossils, and it looks as if we may be able to enjoy some actual scientific discoveries about these bones, rather than a yelling match. (Those interested in a more detailed chronicle and a fair number of links to more information may want to check out my previous Hobbit posts.)

Before the Hobbit bones wound up in Jacob's safe, its discoverers had a chance to look inside its head. They enlisted the help of Dean Falk, a Florida State University paleoanthropologist who has spent years studying the interiors of hominid skulls to find clues to what their brains were like. While brains rot quickly, they leave behind marks where some of their folds and blood vessels were. And since the skull forms such a tight seal around the brain, it ends up with roughly the same shape. To get a good look at these details, Falk has helped pioneer the use of CT-scans to visualize the insides of hominid skulls.

The Hobbit skull was scanned in Jakarta at a 1-mm resolution, and the data was then processed at Washington University's medical school. Falk and her colleagues then analyzed the interior of the skull to calculate the size and shape of the brain, and then produced a three-dimension model of it.

Falk and her colleagues made a careful study of the size and shape of the Hobbit brain, and then they created three-dimensional models of the brains of other hominids. They compared it to the brains of average female humans, a female pygmy, and a microcephalic girl. (They chose females because the Hobbit skull is believed to belong to a female.) The scientists also looked at endocasts of fossil hominids. As I mentioned in my earlier posts, the discoverers of H. floresiensis suggested that it might have evolved from Homo erectus, a tall, large-brained hominid that is believed to have left Africa about 2 million years ago and spread across Asia. Falk's team looked not only at five Homo erectus skulls, but skulls of earlier hominids from Africa, such as Australopithecus africanus and Paranthropus aethiopicus.

The results are being published today on Science Express, the online arm of Science magazine (a printed version will come out in a future issue). The most straightforward results are the ones that address the skeptical suggestions about a small-brained human. The Hobbit brain doesn't look anything like the brain of a microcephalic. Microcephalics have smooth brains, for example; the Hobbit has a normal convoluted surface. Microcephalic brains have a pointed top and a sloping forehead; the Hobbit brain is rounded on top and unsloped in front.

Nor does the Hobbit brain seem to belong to Homo sapiens. It is small (417 cc, which is less than a third the size of an average human brain), and lacks the distinctive shape of human brain. It is wider from ear to ear than it measures from the front to the back of the head, for example. The brains of human pygmies are indistinguishable from those of taller humans, both in size and shape.

Of all the brains that Falk and co. compared to Homo floresiensis, Homo erectus came the closest--in particular, Homo erectus skulls from Java and China. They are also unusually wide, for example. But the Hobbit brain also has some strange features that set it off from Homo erectus. In some ways it is relatively primitive. For example, at the back of the Hobbit brain there is a relatively small occipital lobe. On the other hand, Falk and her colleagues noticed some traits in Hobbit brain that are more human-like. It has more convolutions at the front, for example, than Homo erectus. The temporal lobe, where hearing, memory, and emotions are handled, is enlarged, as well as the parietal association cortex, where some sensory information is handled. Perhaps most intriguingly, a region of the frontal lobes known as Brodmann's area 10 seems to be very large in the Hobbit. It is also large in living humans, and is known to be important in planning and other complex kinds of thought.

Now, a study of single skull cannot be the last word about an entire population of hominids. But it strengthen some possible explanations for Homo floresiensis, and weaken others.

1. A few ordinary pygmies and a microcephalic: It's hard to imagine how its advocates will be able to continue promoting it. I can imagine a skeptic saying, "Well, this person suffered from an unusual form of microcephaly that the scientists didn't look at." But that would be a desperate reach.

2. An extraordinary group of Homo sapiens. Imagine that humans settled on Flores and then underwent a dramatic evolution that shrank their bodies and altered their brain structure. This explanation might account for the human-like features of the Hobbit brain.

But the time range of Hobbit fossils pretty much rules this one out. Modern populations of tall, big-brained humans are believed to have arrived in Southeast Asia about 50,000 years ago, and the oldest Hobbit bones are 95,000 years old. What's more, tools on the island suggest that hominids have been on Flores for 800,000 years.

3. Descendants of Indonesian Homo erectus. It is less of a stretch to envision a population of Homo erectus evolving into the Hobbits. After all, its brain has the strongest overall resemblance to that of Homo floresiensis. And Homo erectus have been in southeast Asia for at least 1.8 million years. Perhaps a few Homo erectus individuals were swept onto Flores hundreds of thousands of years ago and gradually evolved smaller brains.

But this scenario is odd in its own way. It would require certain parts of the Homo floresiensis brain to have enlarged (relatively speaking), even as its overall brain size was shrinking drastically. And these enlarged regions allow us humans to do some of our most abstract thinking.

4. Something completely different. Before I explain what this might be, I have to explain a major problem with explanation 3.

Scientists have found a strong relationship between a person's body weight and the percentage of their total weight made up by their brain. For a grown man of normal weight, the brain may make up just two percent of his weight. But a pygmy woman's brain may be 3% of her body weight. These different percentages are the result of how the human brain and body grow. The brain grows rapidly during childhood and reaches nearly adult size around age ten. The body, on the other hand, grows more slowly and for many more years. Pygmies are smaller than average humans because their bodies stop growing earlier, but not before their brains have reached adult size. This relationship between brain and body follows a steady curve. It is so steady that you can predict what happens to the ratio of brain to body weight as humans evolve to smaller or bigger sizes.

Scientists have also drawn a similar curve for chimpanzees and other apes, but it's noticeably different. That's because their brain growth slows down dramatically after the first couple years, while their bodies can continue to grow to large size. For any given weight, a human's brain is a higher percentage of his or her body weight than an ape's brain.

It's safe to assume that Homo erectus had its own curve. Drawing the curve isn't easy, because there are precious few skeletons of Homo erectus that include both a brain case and enough of a skeleton to estimate their body mass. Actually, there's just one, a 1.5 million year old skeleton from Kenya. Its brain and body size suggest that Homo erectus had a curve midway between apes and Homo sapiens. That's a nicely unsurprising result, because adult Homo erectus stood about as tall as living humans but only had a brain about 900 cc. (Ours are 1350.)

Now come the Hobbits. Because the scientists have both the skull and some parts of the skeleton, they can estimate both its body weight--about 50 pounds--and the percentage of its body weight made up of brain: 1.7% They plotted this value on their graph and found that it did not fall on the human curve, nor on the Homo erectus curve. Instead, it fell on the ape curve.

What this means is that if you scaled down Homo sapiens to the size of the Hobbit, its brain would be much bigger than the Hobbit's brain. And it also means that if you scaled down Homo erectus, it would also have a much larger brain than a Hobbit.

This suggests that the Hobbits actually descend from some other hominid, one with an even smaller brain than Homo erectus. How could this have happened, given that Homo erectus and Homo sapiens are the only hominids known from southeast Asia? Perhaps the Hobbits represent a major branch of hominid evolution that's been hidden from view till now.

About six million years ago, the first hominids branched off from other apes in Africa, and until about 2.5 million years ago they had brains that weren't much bigger than a chimpanzee's. One branch appears to have evolved big brain and tall bodies, and paleoanthropologists believe that this branch gave rise to Homo erectus, Homo sapiens, Neanderthals, and perhaps a few other species. Paleoanthropologists generally believed that these big-brained hominids were the only ones to leave Africa. Homo erectus moved out first, followed by successive waves of Homo, until our own species expanded out of Africa about 50,000 years ago.

But in 2002 scientists found a baffling hominid skull in the former Soviet republic of Georgia. It was 1.75 million years old, it had some resemblance to Homo erectus, but it had an amazingly small brain, measuring only 600 cc. Larger Homo erectus-like fossils have also been found at the same site, dating back to the same age, and scientists have been arguing whether they belong to the same species or to different ones. Some scientists have suggested that the tiny Georgian hominid represented a second migration out of Africa. These migrants were not Homo erectus, but perhaps belonged to an older lineage of hominids. That lineage might have included one or more of the small-brained hominids that are the oldest known species to have used stone tools (Homo habilis, or Australopithecus garhi). Michael Morwood, one of the Hobbit's discoverers who was at the press conference this afternoon, mentioned that Homo floresiensis has some Australopithecus-like traits in the lower part of its skeleton.

So here is a fascinating scenario to consider: a small-brained African hominid species expands out of Africa by 2 million years ago, bringing with it stone tools. It spreads thousands of miles across Asia, reaching Indonesia and then getting swept to Flores. It may not have undergone any significant dwarfing, since they were already small. This would change the way we think about all hominids. Being big-brained and big-bodied could no longer be considered essential requirements for spreading out of Africa. And one would have to wonder why early lineages of hominids became extinct in Africa when one branch managed to get to Flores.

So explanations 3 and 4 seem to come out strongest at the moment. Either one would mean that the Hobbit represents an amazing experiment in hominid brain evolution. They suggest that some human-like features emerged in hominids that were separated from us by two or maybe three million years of evolution. Yet their brains were mosaics, sharing features with us and with other hominids, and also had features of their own. These strange brains, Dr. Morwood argues, allowed Hobbits to do things some pretty elaborate things, such as butcher dwarf elephants or make fires. It would be wonderful to know how these strange brains were wired together, but we have to be content with their shadows. But even shadows can sometimes reveal a lot.

The Sydney Morning Herald reports today that the bones of Homo floresiensis, aka the Hobbits, have at last been returned to the team that originally discovered them.

The team, made up of Indonesian and Australian scientists, discovered the bones on the Indonesian island of Flores. Last October they declared that they had found a new species of diminutive, small-brained hominid that existed just 12,000 years ago. Then, in November, the bones wound up in the hands (or, rather, the locked safe) of the Indonesian paleoanthropologist Teuku Jacob. Jacob claims that a member of the Hobbit team asked him to look at the bones. Members of the team see things a bit differently: they accuse him of poaching. In response, Jacob has called them a bunch of would-be conquistadors. (I have some more background in my previous Hobbit posts.) Jacob promised to return the fossils by the end of 2004, but it is only now, two months later, that he's made good.

But I'll bet that we'll be hearing again about this conflict before long.

Jacob has scoffed at the notion that the Hobbits are a separate species, saying instead that they were pygmy humans. He thinks the team that discovered the bones was fooled by the skull they unearthed. Its seemingly primitive condition, he claims, is the result of a birth defect called microcephaly.

This interpretation would, I imagine, please the minority of paleoanthropologists who are unhappy with the current consensus about modern human origins. Most researchers agree that all living humans descend from a small group of Africans who lived less than 200,000 years ago. Other hominids--Neanderthals, Homo erectus (and possibly Homo floresiensis)--became extinct. They may have mated with members of our own species when Homo sapiens came out of Africa about 50,000 years ago, but they've left behind little or now DNA in living humans.

But some researchers have stuck to an older hypothesis, that living humans have multi-regional roots that include Neanderthal and Homo erectus ancestors. In fact, they see no need to split up hominids of the past one million years into separate species.

It was thus not a complete surprise to learn last week that Teuku Jacob had arranged for two multiregionalists, Alan Thorne and Maciej Henneberg to examine the Homo floresiensis bones. Publicity may have been one motivation--a 60 Minutes crew was apparently filming the proceedings--but something very significant happened along the way. Two grams of the fossil material was extracted and sent to Germany to look for DNA.

As I've mentioned before, finding Hobbit DNA is the best way to test the hypothesis that these fossils belong to another species. If the Australians are right, its DNA should be only remotely similar to the DNA of all living humans. If Jacob is right, the DNA should resemble the DNA of living Southeast Asians more than other humans.

But any results that come from the DNA Jacob and company have extracted will probably be viewed with a lot of skepticism. It is very easy for fossils to become contaminated with the DNA of living humans once they are unearthed, and it very difficult to distinguish between contamination and any ancient DNA the fossils might contain. Jacob didn't help matters when he "borrowed" the bones; apparently they were simply stuffed into a leather bag and brought to him. And his new colleague, Alan Thorne, has already drawn some intense criticism for not being careful enough about DNA contamination when he claimed to have found ancient genetic material from the fossils of early Australians.

For the time being the Australian-Indonesian team may not be able to help matters much. Reports indicate the they hadn't extracted DNA from the fossils before the fossils were extracted from them. Still, the discovery has focused the world's attention on the caves of Flores and nearby islands, which may translate into many hours of digging, which may in turn translate into a lot of new fossils in years to come. Here's hoping that they aren't yanked around so pointlessly.

The Guardian has a long but disjointed report about the dispute over Homo floresiensis. Articles like these rarely give a very good picture of scientific disputes, since all parties involved only get a couple catchy quotes apiece. I've been particularly puzzled by Teuku Jacob, the elderly Indonesian paleoanthropologist who sparked the controversy by taking possession of the bones and locking them away from the Indonesian and Australian researchers who found them. So I was pleased when my brother, a linguistic anthropologist who does research in Indonesia, passed on this link to a translation of a long essay by Jacob. My brother promises me that the translation is accurate. There's a fair bit of science here, although Jacob isn't averse to calling his Australian rivals "latter-day conquistadors."

The Australian media are doing a fantastic job of keeping up with the developments with Homo floresiensis. Here's the first three-dimensional reconstruction I've seen of the little hominid, made by an Australian archaeologist. It's published on the Australian Broadcasting Corporation's web site. I'm sure that as more bones emerge, the image will improve, but this is still a wonderful first look.

Homo floresiensis update: The Economist weighs in on the "borrowing" of the fossils. They mention that when the bones were removed, they were simply stuffed in a leather bag. This is not exactly the sort of procedure you see in protocols for avoiding contamination of ancient DNA. In the Australian, the discoverers of "Florence" vow to return to the fossil site, and this time they'll put their discoveries in a really good safe. Wise move.

The tension continues to mount over the locking-up of the Homo floresiensis fossils, according to this new article in the Australian. (via Gene Expression)

Last month saw the bombshell report that a tiny species of hominid lived on an Indonesian island 18,000 years ago. Since then there has been a dribbling of follow-up news. Some American paleoanthropologists have expressed skepticism, pointing out that while bones from several small individuals have been found, only one skull has turned up. The skull was the most distinctive part of the skeleton, with a minuscule brain and other features that suggested it was not closely related to our own species. The skeptics suggest that these hominids were actually modern human pygmies, and that the skull came from an individual who suffered a genetic disorder called microcephaly.

In Friday's issue of Science, Michael Balter reports that a prominent Indonesian anthropologist, Teuku Jacob of Gadjah Mada University, thinks Homo floresiensis was a microcephalic. He has taken possession of the fossils to study them, and this has a number of researchers worried. Jacob is known to guard fossils in his vault, and so he may essentially be making it impossible for other researchers to look at them. Balter quotes one of the authors of the original report on the fossils, Peter Brown of the University of New England in Australia, saying, "I doubt that the material will ever be studied again."

This could be staggeringly tragic, because the world is waiting for the other shoe to drop: is there any DNA in the fossils?

The fossils are so young that they might well contain some genetic fragments, and this DNA could quickly resolve the debate over which species the bones belong to. If they belong to human pygmies, their DNA should be more similar to the DNA of Australian aborigines or Southeast Asians than to Europeans or Africans. But if, as Brown and his colleagues suggest, they belong to a species that branched off from an Asian population of Homo erectus, then their DNA should not be particularly close to any living human's genes. Most evidence indicates that Homo erectus in Asia shares a common ancestor with Homo sapiens that lived two million years ago. It might even be possible to compare Homo floresiensis DNA to the fragments of Neanderthal DNA that have come to light in recent years. If Brown is right, then Neanderthal DNA should be more similar to human DNA than that of Homo floresiensis, because Neanderthals and humans share a common ancestor that lived roughly 500,000 years ago--four times younger than the ancestor we share with Homo erectus.

According to an Australian newspaper, Brown and his colleagues have found hair that may belong to H. floresiensis, and which may contain DNA. But if that turns out to be a dead end, the next best hope will be the fossils. And the biggest challenge in finding fossil hominid DNA is contamination. You don't want to accidentally grab DNA from a lab assistant's thumbprint. If the Homo floresiensis goes down a bureaucratic rabbit hole, that challenge could become enormous.

Get to know that little skull. Scientists are going to be talking about it for centuries.

As researchers report in tomorrow's issue of Nature, the skull--and along with other parts of a skeleton--turned up in a cave on the Indonesian island of Flores. Several different dating methods gave the same result: the fossil is about 18,000 years old. (Additional bones from the same cave date back to about 38,000 years.) If all you had was the 18,000 year figure and this picture to go on, you might assume that the skull belonged to a small human child. After all, there is plenty of evidence that Homo sapiens had already been in this part of the world for 25, 000 years. But you'd be wrong.

The skull actually belongs to a previously unknown species of hominid, whose ancestors split off from our own some 2 million years ago. Homo floresiensis, as it's known, stood three feet high as an adult and had a brain less than a third the size of our own.

To understand just how mind-blowing Homo floresiensis is, you have to consider it in the context of hominid evolution. Our closest living relatives (chimpanzees and bonobos) live in Africa, and both genetic and fossil evidence indicate that the common ancestor we share with them lived in Africa as well. The oldest known hominids--those species more closely related to us than chimps or other primates--date back 6 million years. They were short, probably could walk upright, and had brains about the size of a chimpanzee--about 350 cubic centimeters. It was only about 2.6 million years ago that hominids started using stone tools, and only about 2 million years ago that species emerged that stood as tall as we do. Its brain was also bigger--850 cc. The increase in brain size may not have been all that significant, since bigger mammals tend to have bigger brains, smart or not. But shortly after this evolutionary surge, the first hominids turned up outside Africa. Homo erectus moved as far east as China and Indonesia within just a few hundred thousand years. At the very least, their migration suggests an expanding population of meat-eaters who have to seek out much bigger ranges than their ancestors.

The Asian population of Homo erectus had little, if anything, to do with our own origins. The oldest human fossils, dating back 160,000 years ago, were found in Africa, and there's a pretty good chain of evidence showing that Homo sapiens descends from hominids who stayed home on the mother continent while Homo erectus swept across Asia. For instance, African hominids underwent a massive burst of brain expansion around 500,000 years ago to close to our own capacity. Meanwhile, Homo erectus in Asia underwent a slight increase, if any. Humans only expanded successfully out of Africa about 50,000 years ago. They may have interbred with Homo erectus, but most of our genome still points back to a recent African origin.

Paleoanthropologists were first attracted to Flores when 800,000 year old tools were found on the island in 1998. Boats seem to have been essential for getting to Flores, which speaks of a pretty impressive mental capacity for Homo erectus . (On the other hand, lizards and elephants and other land animals got to the island without a boat--perhaps by swimming being swept away on logs during storms.) Researchers poked around on Flores, and last September they turned up something none of them had expected: Homo floresiensis. Homo floresiensis was not an ape--it had the signature traits of a homind, such as a bipedal anatomy and small canine teeth. But it wasn't a pygmy human, either. Pygmy brains are in the normal range of variation for our own species. What's more, the floresiensis brain wasn't just small but had a drastically different shape than ours--a shape more like the brain of Homo erectus. This and other anatomical details have led the researchers to conclude that Homo floresiensis branched off from Homo erectus and evolved into a dwarf form.

Here is case-closed proof that today's solitary existence of Homo sapiens is a fluke in the history of hominids. Even 18,000 years ago, at least one other species walked the Earth with us. Exactly how Homo floresiensis went extinct no one knows, but close to the top of the list would have to be ourselves. Neanderthals survived only a few thousand years after humans turned up in Europe, and Homo erectus seems to have disappeared from Indonesia around 40,000 years ago, just around the time humans came on the scene. Perhaps Homo floresiensis lasted longer on Flores because it was harder for humans to reach.

A dwarf hominid on an island is fascinating for another reason--islands are famous for fostering the evolution of dwarf animals, from deer to mammoths. It's possible that the small territory of islands and the lack of competition and predators favors the small. For the first time, hominids have fallen under the same rule. Islands mammals have also been shown to sometimes evolve much smaller brains, and, incredibly, the hominid brain is subject to the same rule. Homo floresiensis's brain shrank down to the smallest size ever found in a hominid. Did Homo floresiensis lose the mental capacity to use tools along the way? The researchers found stone tools in the same site where they found Homo floresiensis, but it's not clear whether Homo floresiensis made the tools, or humans used them (perhaps to kill Homo floresiensis?).

One of the most interesting questions that comes to mind with the discovery of Homo floresiensis is how far back it goes in the fossil record. Just how long did it take for a lineage of hominids to lose half their height and two-thirds of their brain? It may have taken a million years, or a few hundred thousand, or maybe less. In a commentary in Nature, Marta Lahr and Robert Foley of Cambridge point out that it only took 12-foot high elephants on Malta only 5,000 years to shrink to the size of a dog. I've always been a bit skeptical when people forecast dramatic change for our species. But if evolution can produce Homo floresiensis, who knows what a few thousand years on Mars or another solar system could take our descendants?

Update, 11/1/04: Here's a bundle of papers, interviews, and such on H. floresiensis from Nature. Much of it is free.