Anthropology.net

Lee Berger’s got a big problem. Rex Dalton was on his case earlier this year about Berger’s political and cultural approach to his Palaun study. And now Scott Fitzpatrick, one of the most vocal critics of Berger’s dwarves from Palau, has a new paper out in the open access journal PLoS One, where he sinks his teeth into the science behind Berger’s Palaun dwarves. His reputation as a thorough, thoughtful and careful scientist is challenged. Other authors in this paper include Greg Nelson as well as Geoffrey Clark, who specializes in island archaeology and physical anthropology — especially that of Palau.

The paper, “Small Scattered Fragments Do Not a Dwarf Make: Biological and Archaeological Data Indicate that Prehistoric Inhabitants of Palau Were Normal Sized,” argues that Berger et al. studied a fragmentary, small sample size, that other remains found from other sites in Palau fall within normal human variation, and lastly the sizes reported by Berger et al. are very similar to samples of regular humans from Chelechol ra Orrak… the latter claim was mentioned in the Elizabeth Culotta news piece, by John Hawks and Berger himself in our comment thread.

I’m not at all surprised by the points raised in this current paper. I’ve read the paper and will be reviewing it in this post. For any burgeoning physical anthropologists and paleoanthropologists out there, this is an important paper to read. The take home message of ‘don’t rush to publish and know the previous literature and samples by heart when you tread on someone else’s turf.’

In the introduction, Fitzpatrick et al.’s burst out guns a’ blazing. It is pretty much an academic slam of Berger et al.’s work. Of particular interest is Fitzpatrick et al.’s discussion about the problems with approaching comparative anatomy with a bias… Because it really gets at the heart of Berger’s issue. The authors write,

“Researchers familiar with Oceanic prehistory should work from a null hypothesis that human skeletal material found in Palau represent modern humans of normal stature and body mass… Unfortunately, it appears that Berger et al. did not do this and instead operated from a null hypothesis, based off their initial impression from a few fragments that exhibited small or primitive dimensions (one of which-apparent brow ridges—turned out to be carbonate precipitate that eventually flaked off), that their sample represented a population of small-bodied humans.”

There’s no denying that Lee Berger approached the remains with preconceived notions that he was looking at the remains of Homo floresiensis before completely assessing them. Hell, he sought out an emergency grant to study the caves thinking that he was introduced to a cash crop of Hobbits. But, that aside, there flaws can bee seen in the hard science.

Berger et al. estimated stature off of two femoral heads. John Hawks mentioned that… saying,

“Berger and colleagues have no femora sufficiently preserved to estimate length, but their two femoral heads have diameters of 38.8 and 36.1 mm.”

Hawks is right, it is hard to estimate length of a femur and ultimately height when you don’t have the whole bone. And that’s exactly what Fitzpatrick et al. address. They compare the mean values of femoral heads from Chelechol ra Orrak, a site just a few kilometers from Berger’s site. They report that while Berger et al.’s sample is smaller than the average modern human, they are only slightly less than the average diameters from Orrak. When removing the larger individual from the Orrak sample, the mean falls to ~37.16mm — meaning the maximum diameter for two individuals from Orrak is below that reported by Berger et al.

Map of Chelechol ra Orrak (blue C), Ulong (red U), Ucheliungs (green U) and Omedokel (yellow O) Cave Sites in Palau

The samples from Orrak were found in 2000 by Fitzpatrick. Orrak is just as old as Berger’s sites, around 3, 000 years before the present. The site yielded 25 individuals spanning from prenates to neonates to adolescents and adults of both sexes. There are several complete skeletons present, but only one has been excavated and the rest are a made up of a hodgepodge of elements. Here’s the catch: when looking at other measurements of the bones from Orrak, we see that they are of people of average size and stature. The Orrak females averaged about 5-foot, 1-inch in heigh, within the average accepted values of height.

Okay, what about the frontal bone? Remember the reporting of the ‘moderate bossing’ of the frontal? Well both sites reported by Berger, the Ucheliungs and Omedokel caves, are limestone rockshelters. When mixed with water, limestone as you may know deposits a hard matrix which may have caused the lumpy, primitive brow ridges on an otherwise modern human’s frontal bone. But since the frontal bone wasn’t available to Fitzpatrick et al., they take their analysis to other, quantitative measurements such the breadth of the bone.

Frontal breadth measurements taken from 14 male specimens of early Palaun populations have a range of 90 – 96 mm are most certainly small. A specimen from Orrak has a frontal measurement of 90.5mm. But the other measurements, such as the cranial length, breadth, and basion-bregma height indicate that this individual had long cranium with a normal brain size — he or she just a small face. The point to take home, as with the femoral head one, is that one measurement alone need not make a dwarf.

Berger et al. said the teeth they studied appeared megadont and therefore primitive. Megadont just means large toothed. While the teeth are undoubtedly large, they are not very primtive. Fitzpatrick et al. list four different publication which say that the ‘megadont’ measurements fall within the range of early Palaun populations of modern humans. Early Palaun populations were hunter gatherers, who did not experience a reduction in tooth morphology that came with the onset of agriculture. I’d venture to say that almost every physical anthropologist knows this. And for Berger et al. to not consider this and not review the previous literature, is just damn sloppy.

Fitzpatrick and team use the measurements from femora, crania, and teeth from Orrak to show that people from 3,000 years before the present in Palau were just simply gracile. The logical consideration is that people from same period found in the Ucheliungs and Omedokel sites are the same, not dwarves. But what about the discussion Berger et al. gave to their small-bodied ‘pygmies’ transitioning into larger-sized people around 1,000 years ago? Fitzpatrick and team go the extra mile and integrate archaeological and linguistic evidence and conclude that there aren’t any signs of external influence on Palaun language and material culture that could account for the change in larger body size.

I chuckled a bit inside when I read the conclusion in Fitzpatrick et al.’s paper. They write that they used a “sledgehammer to crack a nut,” with the wealth of evidence they showed. But you shouldn’t pity Berger. He didn’t take the time to understand the area in which he was working. That was evident with the upset he caused within Palau and the story Rex Dalton was able to extract. Greg Nelson provided the press with this scolding quote:

“Any time you work anywhere, you have to understand this history. You just can’t walk in and cowboy it, pull some stuff out and draw conclusions in the absence of understanding the bigger picture.”

Fitzpatrick, S.M., Nelson, G.C., Clark, G., DeSalle, R. (2008). Small Scattered Fragments Do Not a Dwarf Make: Biological and Archaeological Data Indicate that Prehistoric Inhabitants of Palau Were Normal Sized. PLoS ONE, 3(8), e3015. DOI: 10.1371/journal.pone.0003015

Razib points me to this press release announcing a study estimating Neandertal intelligence by way of their stone tool set. The press is running wild with this news. The Independent put out a piece on it. So has the Guardian. Even the BBC has got something to say about it. And the story has made it to front pages of Slashdot, Digg, and Wired. Unfortunately, the research paper has not yet been published, but it will be appearing in the Journal of Human Evolution under this title, “Are Upper Paleolithic blade cores more productive than Middle Paleolithic discoidal cores? A replication experiment.”

In lieu of the primary source, I have extracted some information from the news I’ve read. The lead author of the paper is Metin Eren. He and the archaeologists on his team did some experimental archaeology. In other words, they recreated the Neandertal tool set as well as the more modern human tool set. The summary that Brandon Keim, of Wired, provided is rather misleading. Keim says that they analyzed tools used by Neandertals — not really. From what I can tell, Eren and crew made some wide flakes (from discoidal cores) that resembled Neandertal and human tools from the Middle Paleolithic tools and compared them to more specialized narrow blades made by modern humans, from the Upper Paleolithic, who came from a more recent expansion out of Africa.

Flakes were made by archaic Homo somewhere around 250,000 years ago. It involved taking rock like flint and subjecting it to percussion flaking. This created fragments where one side resembles a bi-convex, shell-like shape. Another heavy percussion blow to the bottom of the piece resulted in a convex lens-like shape. This methodology, often called the Levallois technique, was perfected by Neandertals into what is now known as the Mousterian culture.

Aside from being narrow, blades are more or less parallel flakes of brittle rock, like flint, chert and obsidian. They are most often twice as long as wide and the cross section of a blade is triangular or trapezoidal. Blades functioned in many different tools from knives to scrapers, spear tips, drills, awls, bruins, etc.

The authors next measured circumference of these stone tools using a method developed by Adobe and Think Computer corporations. With this, they were able to calculate how much cutting-edge was created and estimate the production efficiency as well as the life time of the tool. Their results indicate that there was no technical advantage to blades from the Upper Paleolithic. And, they conclude that Homo sapiens were not more advanced than Neandertals. Eren comments, saying,

“It’s not a better technology, it’s just a different technology.”

This is not a very surprising result. And I agree with Eren that we need to stop thinking Neandertals as clumbering cavemen. Razib has already outlined some of the basic facts, i.e. Neandertals had big brains and other conquest during human history were not won by ‘great technological imbalances.’ In 1997, people recovered mammalian DNA from the surfaces of Neandertal stone tools, which showed they were able to take down large game like rhinos and mammoths. Clearly, a sign of an intelligent being.

All this ‘let’s rethink Neandertals as intelligent beings’ reminds me of February’s isotopic study on a Neandertal tooth. There was so much press buzzing around, stating that, “Ohhh new fancy research shows Neandertals were mobile.” When in fact, any logical person would have never questioned Neandertal mobility.

One last point. This study challenges the notion that modern Homo sapiens technology gave them an evolutionary upper hand — a better tool set of narrow blades helped modern humans outcompete Neandertals in hunting of big game, and thus survived more effectively. Though Neandertals had different tools, this analysis showed that their tools didn’t have much of a difference in cutting effectiveness and were just as costly as Upper Paleolithic blades. While I haven’t had a chance to read the original paper — it isn’t online yet — I wonder if the authors discuss the differences in the applications of blades versus flakes? Both may have been just as effective in cutting surface but blades functioned as more diverse compound tools, i.e. they could be interchanged between harpoons and spears, knives and scrapers. A compound tool’s advantage over less versatile Mousterian tools, is that they can be repaired — costing the toolmaker and culture less resources spent in fashioning new tools.

And if you want to see the data that Eren and team produced, you know to do your own number crunching, they’ve made it available on Think Computer corp’s website.

Whatever you call him Ötzi, Oetzi or simply the 5,300 year old Tyrolean iceman mummy found in the Alps in 1991… you can’t deny that he doesn’t have a special place in our collective curiosity. We’ve explored his fertility, his last meal, and his cause of death. Why? Well, he’s the most intact late Neolithic human we know of — preserved almost immaculately. His body and the artifacts associated with him have allowed us to peer into his everyday life and technologies.

Otzi Under The Knife

In this new paper, “Species identification of Oetzi’s clothing with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry based on peptide pattern similarities of hair digests,” published in the journal Rapid Communications in Mass Spectrometry, the authors investigated the origin of the skins Oetzi was wearing before he died. As the title of the publication and specialized journal suggests, the methodology deployed was type of mass spectrometry – specifically MALDI-TOF mass spectrometry.

Mass spectrometry is a tool used to detect the molecular mass of a sample — it is especially useful in extracting the structural information of peptides and other organic compounds. I’ve used a mass spec machine in my organic chemistry courses as an undergraduate to identify small organic molecules. It wasn’t a particularly enjoyable experience. I’ve also used them in my protein chemistry classes as a graduate student to identify larger molecules.

In general, mass spectrometry devices have three parts: an ionizer, a analyzer and a detector. The ionizer adds or removes charges to the molecules studied. The ionized molecules are then separated by the analyzer, which filters molecules according to their mass (m) -to-charge (z) ratios (m/z). The detector identifies this m/z ratio.

In the current study, Matrix Assisted Laser Desorption Ionization (MALDI) was used because it is really good at dealing with molecules of high molecular mass, like proteins and peptides. A laser bombards the sample molecules in an absorbing matrix. When the molecules become ionized, they displace the matrix. The analyzer in this study, a time-of-flight (TOF) analyzer, measures the time it takes for ions in the displaced matrix to travel through a field free region known as the flight tube. The heavier ions are slower than the lighter ones.

The authors applied MALDI-TOF mass spectrometry to polypeptides from proteins off of his moccasins, coat, and leggings. We’ve known from the get-go that Ötzi’s clothing was made of animal skin — but what type of animal has been subject to debate. Some believed his moccasins were made of bearskin, indicating he was a hunter-gatherer. But if he’s wearing domesticated animals, then that’s a different story.

The MALDI-TOF mass spectrometry analysis indicated his moccasins were made of cattle. And his coat and pants were made of sheepskin. A table documenting the diagnostic m/z peaks compared to other artiodactyls and mammals from the Alps was provided by the authors. I’ve modified the table, highlighting the similarities between the two samples and the animals they matched up with. Ha 2/91 is the sample from the moccasins, while Ha 43/91/130 and Ha 6/91 were from his coat and Ha 5/91 from his pants:

Table 2. List of diagnostic peaks from selected species from the artiodactyls and their counterparts in the archaeological samples. Values of m/z in bold are used as guidance peaks for the artiodactyls. Diagnostic peak masses underlined were used as precursor ions for species identification. The value m/z 2622 marked with (*) was only selected after MS/MS analysis

This finding indicates that at the very least Ötzi was in touch or trade with animal pastoralists. Of course, the press and authors don’t take this conservative estimate. They’re touting him as an animal herder. It is really hard to say what Ötzi’s real life story was — he’s got scraps of metal under his fingernails which indicates he was into smelting or other metallurgy. He was also found with a remarkable knife and copper-bladed axe by his side, and with that blood clot from the arrow wound and the blunt force trauma to his skull, he definately didn’t die as a peaceful shepherd or cattle herder.

If you questioned why the authors didn’t do an ancient DNA analysis of the hairs from Ötzi’s clothing, then you are not alone. Ancient DNA analysis seems to be all the rage. And for a guy that was frozen under a glacier for 5,000 years, you’d think that DNA would be well preserved. But the authors offer up a pretty convincing argument on why they didn’t do a comparative ancient DNA study. Hides that were fashioned into shoes, pants, and coats had to have been heavily processed — i.e. treated to heat and smoked, salted, soaked in urine, rubbed with animal dung, beaten, dragged over sharp sticks and/or put into herbal tanning baths for long incubation periods that would have undoubtedly damaged DNA. Protein, however, is much more resilient.

I welcome this sort of study. I’m always one for integrating new methodologies, especially from other fields… in this scenario chemistry meets archaeology. MALDI-TOF mass spectrometry, if you don’t know, has been used to check the purity of animal hair, such as Cashmere wool, and in this study it was used to figure out Ötzi’s link to a pastoralist culture.

Hollemeyer, K., Altmeyer, W., Heinzle, E., Pitra, C. (2008). Species identification of Oetzi’s clothing with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry based on peptide pattern similarities of hair digests. Rapid Communications in Mass Spectrometry, 22(18), 2751-2767. DOI: 10.1002/rcm.3679

Michelle Glantz, Sheela Athreya, and Terrence Ritzman have taken up yet another a reassessment of Teshik-Tash Neandertal child in the latest issue of the American Journal of Physical Anthropology. They’ve published the paper under the title, “Is Central Asia the Eastern Outpost of the Neandertal Range? A Reassessment of the Teshik-Tash Child.”

The child, Teshik-Tash 1, was found in 1938 in Uzbekistan. There’s been a lot of back and forth since the original publication on whether or not Teshik-Tash is really a Neandertal. I won’t be reviewing all the literature since the authors do a good job covering the discussion. The authors argue that one of the main reasons why Teshik-Tash is still considered a Neandertal is because of historical precedence. They also argue that this precedence has prevented people from fully appreciating variation.

They’ve found several things in the literature that may have affected its assessment, such as a reconstruction of the cranium. They claim that distortions and missing elements may have affected the morphology of the sample. For example, missing pieces of the fossil at the base of the cranium are parts that would affect the shape and size of the foramen magnum — a trait that has been used to attribute the Teshik-Tash child as a Neandertal. I’ve taken the liberty of highling an image, provided by the authors, to better illustrate the missing pieces.

Missing Parts of the Teshik-Tash Neandertal

But because of the not measuring reconstructions, there’s a lot of missing data. To compensate, expectation-maximization algorithms, a type of maximum likelihood estimations were deployed as well as  multiple imputation, a technique for fitting models to incomplete data sets. Ultimately the data was analyzed under a principle component analysis and multinomial logistic regression. These statistical procedures are more fitting for small sample sizes with missing data.

The results suggest that Teshik-Tash share a lot of cranial and mandibular similarities to Upper Paleolithic modern humans, and fail to support the suggestion that Teshik-Tash is like its European Neandertal sub-adult comrades. But there really aren’t any other Central Asian Neandertal subadults to say for sure that this guy ain’t Neandertal. Until more specimens are found from that region, its really hard to say that the morphology swings either way.

Glantz et al. give about one sentence to the results of Krause et al.’s mtDNA analysis of Teshik-Tash from last year. A shame they play down the results. The results, which I covered, conclude that 190 base pairs from the Teshik Tash kid’s mtDNA is very similar to other Neandertals. The seqeunce can be found under this Genbank entry. I’ve decided to do a quick phylogenetic comparison of these 190bp to modern humans and other Neandertals. Here are the results, which clearly show Teshik-Tash, in these 190bp, is definitely Neandertal (the Teshik-Tash individual is the yellow, unknown item in the tree, and click thru to see a larger image):

Comparison of Teshik-Tash mtDNA to Other Humans & Neandertals

I’m not saying the DNA analysis is definitive. If more DNA could be sequenced from the mitochondrion of Teshik-Tash, that would be better. But given that ancient DNA is fickle, what we have is pretty damn convincing. Despite the argument for distortion, previous morphological analysis (studies that didn’t rely on compensating for missing data) also support the claim that Teshik-Tash is a Neandertal. With these two lines of evidence, why then are we beating this dying horse?

Glantz, M., Athreya, S., Ritzman, T. (2008). Is Central Asia the eastern outpost of the Neandertal range? A reassessment of the Teshik-Tash child. American Journal of Physical Anthropology DOI: 10.1002/ajpa.20897

Ivan Nasidze and Mark Stoneking, along with a half dozen or so other colleagues, have studied the mitochondrial and Y-chromosome diversity of the Talysh. They’ve published their analysis in the American Journal of Physical Anthropology. The paper can be found under this title, “mtDNA and Y-chromosome variation in the Talysh of Iran and Azerbaijan.”

The Talysh are an ethnic Iranian group who speak a language that most Talysh identify as Tolish. Not surprisingly, Tolish is an Iranian language, which its origins can be traced back to Medean empire. Currently, the Talysh people can be found in northern Iran — in and around Gilan and Ardabil as well as southern Azerbaijan, i.e. the Lenkoran, Astaran, Lerik, and Massalin districts. Between these two physically separated populations, there exists a big linguistic rift. There’s some loose evidence that both the northern and southern dialects are a hodge-podge of other regional languages. Some, like Donald Stilo, have suggested the two dialects should be considered their own respective languages because they are so different.

Since linguistic differences often manifest along with genetic differences (check out the Spitton’s ‘Genes and Languages: Not So Strange Bedfellows?‘ post),  Nasidze et al. investigated the mtDNA and Y-chromosome of the Talish and compared their results to other groups in Iran and the Caucasus. 377 bp of the hypervariable 1 region of the mitochondrial genome was a target, as well as 10 SNPs on the Y-Chromosome from 50 Talysh from Iran and 78 Talysh from Azerbaijan, for sequencing. The sequences have been submitted to Genbank.

Compared to Gilaki, Mazandarani, Turkmen and people from the rest of the Iran and southern Caucasus regions, the mtDNA of both northern and southern Talysh are closely related to each other and neighboring groups. In other words, there’s high amounts of variation with these groups than between these groups for the mitochondrial analysis… which doesn’t support genetic differentiation of the two groups.

But, the variations in the Y-Chromosome of northern Talysh differ from neighboring groups. Southern Talysh Y-Chromosome are very similar to other Iranian groups. The northern Talysh group most closely with Turkmen. Could there have been men from Turkmenistan migrating and influencing the genetic and linguistic composition of the northern Talysh? Sure, about 1,000 years ago, Turkic-speakers, like the Oguz hauled over to the souther Caucasus bringing over Turkic language — now seen in Azeri and Turkish.

But since the Talysh retained an Iranian language, this is scenario may not have been reality. Remember, people, more often than not, reproduce within their linguistic groups…  especially in the past. Furthermore, the haplotypes associated with the Y-Chromosome SNPs M172 and M173 do not support a relationship between northern Talysh and Turkmen. Rather, the shared resemblances M172 and M173, are considered to be a by product of genetic drift, either from a founder effect with the initial migrations into Azerbaijan or because of reduced population sizes.

Since Nasidze et al. could not find any genetic evidence supporting the linguistic divergence between northern and sothern Talysh, they suggest that the differences between the two dialects are due to internal linguistic changes, such as contact with other languages. They recommend an in depth analysis/comparison of dialects to figure exactly why the two are so different. Aside from this conclusion, they do add to the history of the Talysh, one where there was most likely a male-limited bottleneck in the founding of the northern Talysh.

I’ve been trying to do some historical research to see if there’s any documented genocide or war in which Talysh populations have plumeted, something that would affect their genetic diversity. According to census from the Soviety Union, between 1926 and 1989, the population of Talysh speaking peoples in Azerbaijan dropped from 77,039 to 21,914. Could this be the drop that affected the genetic diversity of the Talysh? Anyone know the history of the Talysh a bit more?

Nasidze, I., Quinque, D., Rahmani, M., Alemohamad, S.A., Asadova, P., Zhukova, O., Stoneking, M. (2008). mtDNA and Y-chromosome variation in the Talysh of Iran and Azerbaijan. American Journal of Physical Anthropology DOI: 10.1002/ajpa.20903

In late May, Paul Sereno was in town to talk at the 2008 conference titled, ‘Integrating Evolution, Development, & Genomics.’ He was invited to also give a talk titled, “Living Lakeside in the Sahara: A Chronicle of Holocene Adaptation,” to the Primate Biology Group. I eagerly attended. Paul Sereno, if you don’t know, is primarily a dinosaur paleontologist and geologist. And a really well known one at that. He’s discovered around 10 or so new dinosaur species.

But, in 2000 while on a excavation for dinosaurs and giant crocodiles in Niger, National Geographic photographer, Mike Hettwer, stumbled upon a Neolithic graveyard. Sereno shifted his search for dinos to studying these bodies and the artifacts associated with them. He discussed his finds in his talk.

It was so inspiring to be there, and I’m not alone in sharing this sentiment. I was attending the talk with a couple of friends, and they also felt the same way. The clarity and enthusiasm with which Sereno explained the site and his work was impressive — I’ve rarely seen someone so excited to explain so much material. Sereno’s definitely got a reason to be enthusiastic — he has what is now most likely the largest collection of Early to Mid-Holocene bones ever discovered at a single site in Africa.

Gobero, a Neolithic Site in Niger.

After the talk, I got a chance to have a one and one with Sereno. He let me know about his time frame and publication plans. I’m happy to see that everything has come to fruition. Published today, in the open access journal PLoS One is Sereno et al.’s analysis of the site, “Lakeside Cemeteries in the Sahara: 5000 Years of Holocene Population and Environmental Change.”

In the paper, Sereno and team describe the paleoecology of the site and climatic change. They include their interpretation of the burials and associated artifacts. I’ll do my best summarizing the piece in this blog post, but I really recommend you read this gem for yourself. It is open access and well written — you don’t have really any excuse not too.

This site has been called Gobero, after the local Tuareg name for the area. About 10,000 years ago (7700–6200 B.C.E.), Gobero was a much less arid environment than it is now. In fact, it was actually a rather humid lake side hometown of sorts for a group of hunter-fisher-gatherers who not only lived their but also buried their dead there. How do we know they were fishing? Well, remains of large nile perch and harpoons were found dating to this time period.

Of the 67 burials excavated, five of them date to an occupation span from 9,750 to 9,500 years ago. That’s 250 years or so. Looking at photos of the field site, it is hard to believe these prehistoric people got comfortable in Gobero. They began making pottery and ritually burying their dead. One of these guys, G3B8, is a 2 meter tall dude (that’s like 6 feet 6 inches!). He’s pictured below, buried rather utilitarian, with hands covering his mouth and crossed feet, just as he was found:

G3B8, a 6 foot 6 individual from Gobero, Nigeria. Photo (c) Mike Hettwer, courtesy Project Exploration.

G3B8 is not alone in his stature and robustness. There other burials, both male and female, from this time frame are of similar height. These Early Holocene hunter-gathering fishermen also have characteristic skulls — long and low, with a unique occipital bun and broad nasals. These features aren’t restricted to only adults, in fact, juveniles as young as 4 years exhibit similar traits which are not shared by the later inhabitants of Gobero. These bodies were tightly bound when buried.

Around 6200 B.C.E (8,200 years ago) Gobero began to resemble what we see today. The paleolake dried up, and these tall, robust inhabitants hauled out. The youngest early-Holoecene burial dated to 6210 B.C.E. This regional climate change persisted for about 1,000 years, correlating to climatic deterioration across the Chad Basin and linked to the chilling of the North Atlantic.

The return of humid conditions came about immediately after this arid interruption. The lake refilled, and plants, animals and people moved back to Gobero. These favorable conditions persisted for much longer than the early Holocene occupation, roughly 2,700 years. The new settlers were anatomically much different from their predecessors. For starters, they are shorter. They’ve got tall, narrow skulls, with long faces. This guy, dubbed G1B11, is a mid-Holocene adult male dating to around 4,645 B.C.E. is a good example of the different morphology:

G1B11, an adult male from Gobero, Niger. Photo (c) Mike Hettwer, courtesy Project Exploration.

To better illustrate the differences between the robust, early occupants and the latter, gracile ones. Check out this comparison. On the left is a 9,500 years old skull of this mature male. The eye sockets are square, the cranium is low and check out those nasals! On the right is a 5,800 years old skull of a young adult. While not as mature as his 3,700 counter part, you can see the anatomical differences for yourself — a much taller cranium, look at the forehead.

Kiffian (9,500 year old) Skull vs Tenereian (5,800 year old) Skull from Gobero, Niger. Photo (c) Mike Hettwer, courtesy Project Exploration.

Tenerean Bracelet Girl. Photo (c) Mike Hettwer, courtesy Project Exploration.

7 of 35 burials excavated from the mid-Holocene occupation were buried with artifacts. I remember Sereno showing us a photograph of an individual buried with a turtle shell underneath him. I didn’t catch the specimen number at the time, and now found out that guy is G1B11 — pictured above. You can see in the middle image the carapace functioning as an eternal bed. The mid-Holocene occupants were much more symbolic than the early Holocene individuals. They buried their dead with more elaborate artifacts, such as this 11 year old girl (G1B2) who is wearing an upper-arm bracelet carved from the tusk of a hippo. She’s believed to have died around 4,835 years ago.

Why were these people burying their dead with beads, bracelets, and on turtle shells? Sereno et al. suggest that the latter occupants were more pastoralists and agriculturalists — because archaeological evidence for grain and remains of domesticated cattle are present in the midden from this time period. Their gracile frames support this lifestyle, as well. Additionally, curious looking fine-grained green rocks were used to make points, scrapers and adzes from this time period. This rock isn’t found in Gobero. Actually, this feldspar rock came from Alallaka — a prehistoric rock quarry about 160km north of Gobero. The relaxation from gathering and hunting for food allowed for people to develop new skills, such as jewelry making and symbolic burials — even trade their skills for green rocks from the north.

So who were these people?

Sereno et al. did a principal components analysis of craniofacial dimensions of the skulls from both periods and compared them to Late Pleistocene to mid-Holocene populations from the Maghreb and southern Sahara. The early Holocene occupants are similar to remains from Maghreb Capsian, Maghreb Iberomaurusian, Mali, Mauritania. The mid-Holocene occupants are unlike any other population tested. That doesn’t tell us much about who they were.

Kiffian Wavy Pottery. Photo (c) Mike Hettwer, courtesy Project Exploration.

So Elena Garcea, a specialist in African archaeology from this period, analyzed the material culture. She believes that the bone harpoon points and hooks, as well was the dotted wavy-line, zigzag ceramic exhibit attributes of the Kiffian people. Kevin MacDonald described the Kiffian technology, one that specialized in harpoons and microliths, in the text “Archaeology and Language.” Several other publication also support this claim. On Sereno’s website, there are more photos of the bone harpoons and pottery. I don’t have any examples of Kiffian artifacts to compare, so I’ll just trust that Garcea and Sereno got this identification.

The mid-Holocene occupants are believed to be Tenereans. Their affinity for green feldspar and the small projectile points as well as the disc knives characterize the Tenerean material culture. An example of Tenerean style projectile points from Gobero is below: IMAGE REMOVED AS PER MIKE HETTWER’S REQUEST.

Tenerean Triple Burial, from Gobero, Niger. Photo (c) Mike Hettwer, courtesy Project Exploration.

If you’re not completely floored by the wealth of archaeological and anatomical material from Gobero, let me share with you the Tenerean triple burial. The triple burial includes a female, presumably the mother, laid to rest on her right side who died somewhere around 5,300 years ago. Facing her are two children (ages 8 & 5 years old) and buried on their left side, interpreted as her children. These people were buried with their arms and legs around each other and holding hands. Abundance of pollen residues underneath them suggest they were buried on a bed of flowers. Multiple burials like this, and in this condition are rare. Furthermore, this is first triple burial ever discovered on Africa.

Sereno did not excavate these individuals like a normal archaeologist would. Instead, being a dinosaur specialist, he jacketed the remains.

Gasp!

You’re probably thinking, “Sacrilegious technique! He shoulda used brushes and dental picks, removing each bone and shipping them off for study in the lab.” But had he not done so, burials like the triple burial would have not survived excavation — the extreme heat of the Sahara has made the bones exceptionally fragile. That’s why having a multidisciplinary approach to doing this sorta fieldwork works. And I commend Sereno for taking these measures to preserve the bodies and site.

This publication has shown us very eloquently how important Gobero is to our understanding of climate change and prehistoric peoples and archaeology of the Sahara. As more and more of Gobero becomes exposed and weathered, more and more of it will be lost to time if we do not support Sereno’s research. Hettwer has documented the impact 5 years has made on one such exposed skeleton:

Weathering of an exposed skeleton at Gobero, Niger. Photo (c) Mike Hettwer, courtesy Project Exploration.

I’ve mentioned before only 67 burials were excavated. There are at least 182. Gobero needs to be preserved, for its wealth of evidence and the cultural heritage it provides Niger.

Sereno, P.C., Garcea, E.A., Jousse, H., Stojanowski, C.M., Saliège, J., Maga, A., Ide, O.A., Knudson, K.J., Mercuri, A.M., Stafford, T.W., Kaye, T.G., Giraudi, C., N’siala, I.M., Cocca, E., Moots, H.M., Dutheil, D.B., Stivers, J.P., Harpending, H. (2008). Lakeside Cemeteries in the Sahara: 5000 Years of Holocene Population and Environmental Change. PLoS ONE, 3(8), e2995. DOI: 10.1371/journal.pone.0002995

Last week, I shared with you all some research on the Neolithic/agricultural revolution in Iran and Turkey, specifically on barley and cattle domestication. Since then, PNAS has published a related paper, “Domestication and early agriculture in the Mediterranean Basin: Origins, diffusion, and impact,” by Melinda Zeder. Zeder believes that the domestication of sheep and goat happened 1,000 years earlier than previously assumed.

She bases this hypothesis on the observation that the previous, younger date is based upon the ‘culling of female goats and sheep,’ who are smaller, more gracile, and manageable. But, according to her, people were herding these animals much earlier. In fact, in places like Cyprus, native species like the pygmy elephants and pygmy hippopotamuses were replaced by introduced, domesticated species such as the sheep, goat, cattle, and pig by 10,500 years ago.

Zeder has generated two useful graphs to document the the regions and dates where the four domesticated species were first pastoralized. The evidence from last week’s paper on the milk jugs in northwestern Turkey isn’t included, instead a Zeder marks a 10,000 year old mark for the domestication of cattle in northern Iraq and Syria.

A representation of the time at which these animals were domesticated

In this second graph, it is pretty obvious what the red circles stand for, the dates when farming sprung up around the Mediterranean. Green dots represent the diffusion of agriculture, whereas red and blue dots indicate regions where farmers migrated too and integrated into preexisting populations. From this you can tell that Zeder suggests both a diffusion and assimilation model of agricultural revolution.

Agricultural Expansion in Neolithic Mediterreanean

Zeder, M.A. (2008). Domestication and early agriculture in the Mediterranean Basin: Origins, diffusion, and impact. Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.0801317105

In the open access paper, “Metabolic changes in schizophrenia and human brain evolution,” Phillipp Khaitovich and team have compared the gene expression and metabolite concentrations in the healthy human brain to a schizophrenic brain. Their results indicate that the biggest differences between a normally functioning brain and an unhealthy one was in 9 genes and metabolites involved in energy metabolism and energy-intensive cognitive functions.

If you remember, this paper from earlier this year, you’d wouldn’t be entirely shocked and awed by Khaitovich et al.’s finds. That’s because in this previous gene expression analysis, mice who were feed a human cooked diet (Supersize Me) versus a ‘chimp’ diet exhibited expression and regulation differences in genes involved in metabolism and detoxification. The authors of this PLoS One paper wrote that they think that the ‘introduction of cooking, have caused a relaxation of selective constraints on diet-related genes.’

In line with this hypothesis, Khaitovich and his team write that advent of cooking reduced the energy needs of our digestive system, freeing up calories for our brains. Our relatively small digestive systems funnel up 20% of total energy to the brain, whereas in non-human primates only 13% energy is allocated.

Khaitovich, P., Lockstone, H.E., Wayland, M.T., Tsang, T.M., Jayatilaka, S.D., Guo, A.J., Zhou, J., Somel, M., Harris, L.W., Holmes, E., Pääbo, S., Bahn, S. (2008). Metabolic changes in schizophrenia and human brain evolution. Genome Biology, 9(8), R124. DOI: 10.1186/gb-2008-9-8-r124

The mitochondrial macrohaplogroup M is a descendant of the macrohaplogroup L3, a really old East African haplogroup thought to have originated around 104,000 years ago. Sun et al., explained that within haplogroup M, lie many smaller haplogroups of which the M2 lineage is thought to be the oldest mitochondrial lineage in India. You can check out this preliminary assessment in this Molecular Biology and Evolution paper, “The Dazzling Array of Basal Branches in the mtDNA Macrohaplogroup M from India as Inferred from Complete Genomes.” Ever since the Sun et al. paper, people have wondered why M2 is way more prevalent in southern India. One of the leading hypothesis is that M2 may represent the phylogenetic signature of early settlers. And by early we’re talking 50,000 years old.

To investigate the impact of Middle/Early Upper Paleolithic settlers on the genetic diversity of current populations in India, a dozen or so Indian academics sequenced 72 mitochondrial genomes from 16 different populations in India. Their results are published in this provisional BMC Evolutionary Biology paper, The earliest settlers’ antiquity and evolutionary history of Indian populations: evidence from M2 mtDNA lineage

Except for the Sonowal Kachari, the M2 lineage is completely absent from tribes in northeast India. Whereas in southern India, the Kuruba show a high frequency of M2 signatures at 39%. Furthermore, M2 signatures are found in high frequencies of Dravidian and in Korku, an Austro-Asiatic tribe.

Upon sequencing the mitochondrial genomes of 72 people, the authors identified a deep split between two sister clades, M2a and M2b. M2a differs from M2b in a transitional mutation at positions 7961 and 12,810. Subclades within M2a, which I won’t get into, differ from subclades in M2b — one that shows late branching patterns.

Coalescent estimates were determined, and the authors figured that the founder age for the m2 mtDNA lineage is 50,000 years old, plus or mine a couple thousand. M2a and M2b are believed to have diverged around 21,600 years ago.

The authors next investigated the effects of population decline due to the glaciation. They were able to determine a rapid population growth from 7,000 years ago to 3,000 years ago — with a 500 year decline between 1,000 years ago and 1,500 years ago — which I think could be in part because of the first wave of black plague that swept thru Asia, Europe and Africa.

Because of the persistence of the M2 lineage and its alignment to the L3 macrohaplogroup, the authors believe that the Dravidian people and the Korku represent modern variants of earliest settlers of India. That’s because these people do not exhibit such a high frequency of Eurasian specific mtDNA haplogroups, as other Indian population. Furthermore, there are not any significant M2 signatures in Tibetan nor Burmese people, suggesting that these populations have remain relatively isolated both genetically and linguistically.

Kumar, S., Padmanabham, P.b., Ravuri, R.R., Uttaravalli, K., Koneru, P., Mukherjee, P.A., Das, B., Kotal, M., Xaviour, D., Saheb, S.Y., Rao, V.r. (2008). The earliest settlers’ antiquity and evolutionary history of Indian populations: evidence from M2 mtDNA lineage. BMC Evolutionary Biology, 8(1), 230. DOI: 10.1186/1471-2148-8-230

Both Razib and Dienekes have put a posts about this new Current Biology paper, “Correlation between Genetic and Geographic Structure in Europe.” The authors of the paper compare the genetic make up of 2,514 individuals from Europe using the Affymetrix GeneChip Human Mapping 500K Array Set.

Correlation between Genetic and Geographic Structure in Europe

Always the over achiever of science blogging, Razib has dutifully labeled the populations on the graph. His modifications help better visualize the genetic similarities and differences among and between the European populations tested. And there are some interesting patterns. There’s a similarity among northern European populations as well as a similarity among southern European populations.

Fins tested are the least similar group to other European populations. Swedes and Spanish people are clearly different, while the Irish and British share a lot of admixture among the 500,000 SNPs tested. So what does that all mean? This result indicates that there is a genetic component to European ethnic groups.

Not entirely surprising, because in 2006, we saw the open access journal PLoS Genetics publish a typing of 5,000 SNPs among about 1,000 Europeans and European Americans. In that paper, the researchers were able to resolve the genetic differences between northern and southern European groups. Image below. Also, in January of this year I read and reviewed two papers that did similar tests, comparing 300,000 SNPs between approximately 4,198 European Americans. After some principal component analyses (PCA), there was a clear distinction between individuals with northern from southern European ancestry, as well as separation of Italian, Spanish, and Greek individuals from those of Ashkenazi Jewish ancestry.

European Population Substructure

PLoS Genetics has also recently published a similar paper, “Tracing Sub-Structure in the European American Population with PCA-Informative Marker,” which announces a purely computational method of identifying ancestry — one that doesn’t require a poll of the individuals’ identified ethnic background. The researchers analyzed 1,521 individuals for more than 300,000 SNPs across the entire genome.

While not as robust of a data set as the Current Biology paper, the authors were able to pluck out 200 ancestry informative SNPs that accurately predict fine structures in European American datasets, as identified by PCA. They did so by removing any redundant SNPs uncovered during the modeling process. Moreover, much of the genetic variation identified were between the northern and southern European ancestry groups.

Going back to the ‘is this surprising?’ point, in 1990, Barbujani et al. noted the delineation of northern and southern Europeans between the distribution of 63 allele frequences, in “Zones of sharp genetic change in Europe are also linguistic boundaries,” and attributed the language affiliation of European populations playing a major role in maintaining and probably causing genetic differences. Makes sense.

LAO, O., LU, T., NOTHNAGEL, M., JUNGE, O., FREITAGWOLF, S., CALIEBE, A., BALASCAKOVA, M., BERTRANPETIT, J., BINDOFF, L., COMAS, D. (2008). Correlation between Genetic and Geographic Structure in Europe. Current Biology DOI: 10.1016/j.cub.2008.07.049

Paschou, P., Drineas, P., Lewis, J., Nievergelt, C.M., Nickerson, D.A., Smith, J.D., Ridker, P.M., Chasman, D.I., Krauss, R.M., Ziv, E., Pritchard, J.K. (2008). Tracing Sub-Structure in the European American Population with PCA-Informative Markers. PLoS Genetics, 4(7), e1000114. DOI: 10.1371/journal.pgen.1000114