Sign in

Braided Streams: Evolutionary Dynamics Among Pleistocene Hominins in East Asia

By inPaleobiology |
Score: 2.45 / 4.90

Where Did Humans Evolve?

You’ve probably heard it said many times: Africa is the cradle of humankind. For the most part, this is true. Our immediate hominin ancestors evolved in Africa, and so too did our own genus, Homo. But what about our species, Homo sapiens?

In the early 1900s, most scientists actually thought that Asia was the original homeland of Homo sapiens. There are several historical reasons why the focus slowly shifted to Africa but for the past 30 years, the origin of "modern" human biology and behavior has solidly been seen as originating on this continent  . Most paleoanthropologists now believe that our species only started occupying Eurasia following dispersals out of Africa in the Late Pleistocene (128,000-12,000 years before present/b.p.). The archaic populations living there already, such as Homo erectus, are believed to have gone extinct when Homo sapiens spread throughout the globe (replacement scenario). And in this scenario, eastern Asia is viewed as more or less an "evolutionary backwater" in terms of human evolution, while Africa and Europe are at center stage.

The fossil records from China and Indonesia are very rich, and have been known since the late 1800s. They tell a different story than the one proposed by the above so-called "Out of Africa" model. Most students of anthropology hear about "Multiregional Evolution" as an alternative to Out of Africa, e.g.  . It’s often misunderstood as proposing parallel evolution in different regions for local Homo sapiens populations through time. For this reason, it’s also often dismissed. Because how likely is it that "Asians" evolved separately but at the same rate and in the same direction as "Europeans" and "Africans" to lead to the unified human species today? But this is not what the idea of multiregionalism is proposing  . It is actually founded on the idea of gene flow among these populations – something that prevents two groups from splitting off and becoming separate species. So in reality the idea of multiregional evolution is a complex scenario involving some contribution of each regional group of pre-existing Homo populations to the evolution of local Homo sapiens in that region today, but it also very much expects that the evolutionary processes of natural selection, genetic drift, and especially gene flow shaped these populations. When understood that way, it makes sense that there was continuity between the earlier and later, immigrating populations within each region. That’s why the model is also known as Regional Continuity model. No region evolved in isolation, but also none faced complete replacement by dispersing Africans.

Looking Beyond Africa

Homo sapiens evolution took place during a time period known as the Pleistocene, or Ice Age. It spans from 2.58 million years b.p. to 12,000 years b.p.. The Middle and Late Pleistocene fossil record from China is vast – perhaps as vast as the African record for this time period. The dry, cold climate that China periodically experienced during the Pleistocene, along with its numerous limestone caves, seems to have provided the perfect combination of suitable habitation sites for our hominin ancestors, and excellent preservation for today’s fossil hunting paleoanthropologists.

The best-known hominin fossil locality in China is the home of "Peking Man". For generations, the inhabitants of the village of Zhoukoudian (about 41 km southwest of Beijing) knew about the fossil bones in nearby caves. Beginning in the early 1920s, European paleontologists searching for evidence of human evolution honed in on these caves, and from 1921 until the mid 1940s the site was meticulously excavated by a collaborative Chinese-European team. The originals were unfortunately lost during WWII when the Japanese invaded China; only high quality first-generation casts remain  . Recent advanced dating techniques confirm that the oldest levels at Zhoukoudian are approximately 700,000 years old; the youngest fossils from the site date to around 35,000 years ago. The oldest fossil hominins there are Homo erectus and the youngest are Homo sapiens; no other hominin populations or species were found. Across this nearly 670,000 year span we see important changes in this population but also some continuity that allows us to infer what evolutionary processes were shaping populations in China throughout the Pleistocene.

In addition to this decades-old data, the discovery of new sites in China over the past ten years, plus new analytical tools to study fossils, have given us a much more complex picture for human evolution in the region than the simple "Out of Africa vs. Multiregional Evolution" models provided. Data from China between 100,000 years b.p. and as recent as 11,000 years b.p. reveals how evolutionary change through time took place in China, from the earliest Homo sapiens to the populations that lived at the very end of the last glacial period just before the development of agriculture. This allows us to better understand which evolutionary forces acted on our species and how they shaped us beyond the simple focus on gene flow vs. replacement. More importantly, these sites give us a much better globally inclusive picture of what constitutes "modern humans".

Map of Late Pleistocene hominin fossil sites mentioned in the text (Image credit: Google Maps)

The Earliest Homo sapiens in China: Xuchang

The open-air site of Xuchang in Henan province, central China is just south of the Yangtze River. Two crania were found here and dated to around 100,000 years old. Anatomically, the fossils look like a combination of Homo erectus, Neanderthals, and early Homo sapiens. The most complete specimen, Xuchang 1 has a cranial capacity (i.e. brain size) of 1,800 ml that far exceeds the average value of 1,400 ml seen in humans today. But despite this "modern" brain size, other aspects of the cranial vault such as its low, sloping profile make it look much more like Homo erectus, which lived in China between 1.6 million to around 400,000 years ago. Even more intriguing is the back of Xuchang 1’s skull, which has a modest torus with a depression above it, known as the suprainiac fossa. This trait is not known in Homo erectus or Homo sapiens, but is very common among Neanderthals from West Asia and Western Europe.

The Xuchang fossils have anatomical features that don’t look like they are derived from African Pleistocene populations, nor do they always look like modern-day Asians. They are part of a morphological pattern for China that, along with the sites I review below, require us to reconsider how we approach studying our species’ evolution in Asia.

(Top) Virtual reassembly of the Xuchang 1 cranium. (A) Anterior, (B) right lateral, (C) posterior, (D) superior, (E) left lateral, and (F) inferior views. Gray, filled-in absent portions and mirror-imaged right frontal squamous portion. (Bottom) Virtual reassembly of the Xuchang 2 cranium. (A) Posterior, (B) left lateral, (C) right lateral, (D) posteroinferior, (E) superior, and (F) inferior views (From Zhan-Yang Li et al. (2017) Science 355: 969-972) .

Tianyuandong and early Homo sapiens diet and DNA

The Tianyuandong specimen (known as Tianyuan 1) from northern China (in the town of Zhoukoudian, just six kilometers from the famous Homo erectus site) is the earliest directly dated Homo sapiens from this site  . Bone samples taken from this skeleton preserved enough collagen to date it to between 42-39,000 years b.p., which is roughly the age of the youngest specimens at nearby Zhoukoudian cave. Researchers used a form of radiocarbon dating known as accelerated mass spectrometry or AMS to obtain this date. The good preservation status of the remains also allowed researchers to extract ancient DNA from Tianyuandong, and even to study its diet.

Anatomically, Tianyuan 1 is more similar to Homo sapiens than any other Pleistocene population – it is not particularly robust, aside from having relatively large front teeth (which is common among non-agricultural populations because they usually have tougher diets). The diet of Tianyuan 1 was reconstructed through studying the chemical components present in its teeth. Anthropologists look at the isotope values of bone collagen because these are related to the isotope values present in the food that a person consumes during their lifetime. For example, a higher proportion of heavier vs. lighter isotopes of carbon (13C/12C) indicates that the main source of dietary protein was marine vs. terrestrial plants and animals. In the case of Tianyuan 1, the carbon (13C/12C), nitrogen (12N/14N), and sulfur (34S/32S) isotope ratios all point to freshwater fish as a main component in the individual’s protein diet. Not only does this tell us about the diet of early humans, it also tells us that as far back as 40,000 years ago, humans knew how to fish and exploit coastal resources for their survival.

The Tianyuan 1 specimen is the only Pleistocene Chinese fossil to have its ancient DNA analyzed thus far. Its genomic sequences were compared to 11 individuals from today, which were used to represent populations from Africa, Europe, Asia, and South America  . This sample size may sound small, but when dealing with genomes it’s reasonable to assume that one individual possesses the common alleles (i.e. the genetic variants) for that gene pool. The Tianyuan genome was more similar to the two Asian individuals from the present-day than it was to the European, African, or South American. So even though Tianyuan 1 is from a 40,000 year old population, it isn’t genetically distinct from peopling living in China today. Put another way, there has been regional continuity between the genomes of ancient and present-day eastern Chinese.

These two sites are important for understanding how the traits we see in Asia came to be established. Between the 100,000 year old Xuchang and the 40,000 year old Tianyuan fossils, a great deal of population change was happening throughout Eurasia, Africa and Australia. By 200,000 years ago, Homo erectus is no longer found anywhere in the world, and Homo sapiens are established in China (Zhirendong) as well as West Asia (Israel) and in Northeastern (Ethiopia), Northwestern (Morocco), and Southern Africa (South Africa). By 50,000 years b.p. our species is in Oceania. But in Western Europe, Neanderthals dominate the landscape throughout this time period, until around 35,000 years b.p.. And indeed, we see morphology from all of these populations in Pleistocene East Asia at various times.

To understand the processes that shaped these populations beyond a simple binary of replacement vs. continuity my colleague Wu Xinzhi at the IVPP in Beijing developed a model that seems to have a lot of explanatory power for China known as "Continuity with Hybridization"  . Wu was trying to make sense of the mosaic of traits he observed in the Middle Pleistocene fossils in China, many of which he excavated and was the first to analyze. Based on comparative anatomy, he saw that in some respects, they had features that were also common in European but not African fossils. He also saw that there was a lot of variation within the Chinese fossils, but that they were still unified by a handful of features.

This prompted him to propose a model of gene flow through time that he described as a braided stream. Populations in what is present-day China would experience periods of "isolation" that would lead them to be shaped by very specific local forces, much the way streams are shaped by local landscape and climate. Over time, these lineages would come back into contact with other populations, exchanging genetic material for a period, which would create commonalities among them. This process, Wu argues, continued throughout the Pleistocene.

Maludong and Longlin: Southwest China at the End of the last glacial period

Two sites where we can test the Continuity with Hybridization model are the Maludong and Longlin caves in Southwest China which have been known for the past 20-30 years. A partial cranium was found in Longlin in 1979 by a miner; and a number of different skeletal elements were excavated from Maludong in the 1980s and again in the past five years. Like the Xuchang crania, the Longlin and Maludong fossils have a suite of traits that are not seen in other fossil populations although individually each trait is present in Neanderthals, Homo erectus, or early African Homo sapiens. These two sites are much younger than Xuchang – they date to around 11,000-14,500 years b.p.. This period is known as the Terminal Pleistocene because climatically, it spans the end of the last glacial period and the onset of the warm period we live in now, known as the Holocene. In China, the climate in the region went through a cool and dry spell, and communities were just on the cusp of developing a growing dependence on plant foods that they would begin to domesticate by around 10,000 years before present. This would mark the beginning of agriculture in the region.

In this context, the partial human skull from Longlin Cave and the fragmentary skull, teeth and jaw bone from Maludong shouldn’t necessarily be expected to look like other fossil populations that lived tens if not hundreds of thousands of years before. What makes them remarkable is that they don’t align with any living East Asian populations, or in fact with any Homo sapiens populations at all from the Later Pleistocene. They have brain sizes that are average for our species, but their skulls are overall fairly narrow in the front, and their cheek and teeth regions are large and flared. Even the teeth themselves are fairly robust. This may be related to diet. Maludong means "red deer", named for the large amount of red deer bones that were found along with the human remains. Presumably this population relied on game as a major food source, which is in contrast to older populations such as Tianyuandong who relied upon freshwater fish.

Longlin 1 cranium (each bar = 1 cm) (From Curnoe D et al. (2012) PLOS One 7: e31918)

What could have created the combination of traits in these human populations? To answer this question, we need to do more than simply achieve "hypothesis compatibility" (Templeton) between these data and the existing models. Having worked in this region for the past 20 years, I know that the pattern of morphology in Asia categorically does not support the simple "Out of Africa" model to explain Homo sapiens evolution there. Nor does the simple contrast between this and a model of "regional continuity" help elucidate the processes; such an either/or proposition is both facile and outdated. The new finds from Longling, Maludong, and others offer us an opportunity to update our models by giving voice to a wider and more diverse set of scholars who do not restrict their interpretations to the Western "Out of Africa vs. Regional Continuity" mindset.

Despite disagreement in the scientific community about the uniqueness of these populations, we can still come to an understanding of the forces that may have shaped them. In the "Continuity with Hybridization" model, these populations are not descended from a very recent presence of humans in Asia, nor are they part of an uninterrupted continuous lineage within China over the past 1.6 million years. Rather, they are the product of population dynamics that included gene flow between eastern and western Eurasia, as well as southeast and north Asia.

Viewed in these terms, we can get a better sense of the truly interesting evolutionary questions about our species. We do not just want to know what things looked like so that we can categorize them, we want to know what forces shaped the traits that we are seeing so we can ultimately reconstruct our evolutionary history. There were many paths that different populations took to get to where they are today, and our goal is to reconstruct what those paths were, including how much gene flow, genetic drift, and natural selection shaped us. We need to move beyond shoehorning all of this information into 30 year old models of "Out of Africa" and "Multiregional Evolution". Different regions have been shaped by different forces.

This is not a new idea – paleoanthropologists in China have been saying it for decades. Questions about species and hybridization are not always relevant. It’s not that they are not interesting, but they are a distraction from understanding the processes that shaped the populations.

If we focus on whether or not these populations were separate species, what we should label them ("Neanderthals", "Denisovans", or "modern humans"), and whether or not they interbred, we are taking a typological approach to human evolution – creating categories with boundaries and then seeing how pure they are. Not only is this too much like early racial science, leading to questions about who decides what is "modern", it also does not answer meaningful questions about evolutionary history. It just reduces the variation – the really interesting thing about humans – into types.

What happend to Homo erectus in East Asia?

In the case of China, the old view about East Asian Homo erectus being an "evolutionary dead end" in Asia, replaced by incoming "modern" groups does not really stand up to the data  . China was a dynamic region of evolutionary change, with a long history of important fossil finds, from well-excavated contexts. With recent advances in dating and ancient DNA extraction and analysis, it has become clear that humans have been living there for the past 1.6 milion years and that they were biologically highly variable, as would be expected for a land mass that is 9 million km2 large. It is also clear that these populations were not isolated. These two facts together solidly refute the idea of an African replacement there of pre-existing Homo populations. There was continuity there throughout the Pleistocene shaped by gene flow, random genetic drift, and natural selection. Just how much of each was at play remains to be understood.

The fate of Homo erectus in Asia plays an important role in understanding human evolution during the Pleistocene. When the models of "modern human" origins were first developed in the 1980s one driving issue was: did Homo erectus in Asia go extinct, or evolve into Homo sapiens? The success of the Out of Africa model meant the default assumption was that Asian Homo erectus was an evolutionary dead end. At Xuchang and Tianyuandong, neither the fossil nor genomic data shows an African origin for these populations, not even when considering moderate gene flow in the region from outside. In order to understand the biology of these populations, we have to start with a fresh slate, which means setting aside the age-old idea that Homo erectus went extinct.

The morphology in Eastern Eurasia in the Middle Pleistocene suggests that while there may have been population extinctions and recolonizations, we do not see anything indicating a wholescale replacement of the archaic humans known as Homo erectus. They did not die out, or get replaced, or go extinct. They contributed to the ancestry of Homo sapiens in the region.

With each find, and the application of new technologies, we are finding out about the evolutionary history of our species beyond Africa. This is the kind of holistic picture we need to truly understand to find out who we are and where we came from. The story will be, like us, very varied.

Did you like this article?
Communicating science in an appealing manner matters to our scientists, and if you would like to show your appreciation of their commitment, please feel free to beef up our Author of the Month award.

Other Articles by This Author

Further Articles in Paleobiology


This work is distributed under a Creative Commons Attribution ShareAlike 4.0 International License (CC BY-SA 4.0). Its use, distribution and reproduction in other forums is permitted provided both the original author(s) and Capeia are accredited, and the license is unchanged (CC BY-SA 4.0). Images or other third party material in this article are covered under the article’s Creative Commons license unless otherwise indicated; if the material is not covered under a Creative Commons license, users will need to obtain permission from the license holder before it can be reproduced.