Brazil: a melting-pot genetic present and an uncharted deep past

Glimpsing humanity’s genetic future in its 7th largest nation

In 1494, the expansionary Iberian powers negotiated the Treaty of Tordesillas to officially divvy up the planet’s freshly discovered regions beyond Europe. Over 370 leagues to Cape Verde’s west, a line was drawn down the globe. To the left lay lands discovered by Spain since Columbus’ explorations, and to the right African and Asian territories the Portuguese crown had scooped up in late medieval expansion, begun by Henry the Navigator two generations prior. Enacted ‌as a compromise to assuage Portuguese anxiety about its monopoly of an African route to the east, the most momentous long-term consequence of Tordesillas bisecting the globe along a line of longitude would actually prove to be an unforeseen one: the genesis of Brazil, a singularly multiracial and multicultural melting pot that within some five centuries would become one of the globe’s most consequential nations. At South America’s very easternmost edge, barely known to 16th-century cartographers, Brazil chanced to fall east of Tordesillas’ arbitrary meridian. This technicality allowed the lesser Iberian power to claim the new continent’s eastern edge, which it would later join to the vast Amazonian hinterlands in the interior, creating a singular nation-state of imperial scope. While Portugal counts just 10 million citizens today, Brazilians number over 200 million. The Treaty of Tordesillas, a largely forgotten artifact of early Renaissance Europe and its internecine conflicts, momentously altered the human geography of the planet.

Brazil resembles its hemispheric neighbors in its historical arc. One of the New World’s most oft recycled plot arcs is colonial and settler nations leaving the metropole in the dust, both in terms of prominence and population. Mexico has nearly three times Spain’s population, and is by far the most populous Spanish-speaking nation-state. The US counts five times the UK’s citizens, and succeeded the motherland as the world’s preeminent global economic and military power by the first half of the 20th century. But these two instances pale in comparison to how Brazil’s heft outweighs Portugal’s. Founded in 1822 by a cadet branch of the House of Braganza, the Empire of Brazil was a massive territory fully ninety times larger than Portugal and already more populous at independence.

Mushrooming from a population of under 20 million in the late 1800’s to over 200 million in 2024, Brazil’s massive demographic expansion has consolidated its role as a regional power, as it noses into global prominence, with its presidential politics regularly covered even in the notoriously provincial US press. But perennially buffeted by boom-bust economic cycles driven by an overreliance on commodities like sugar, coffee and rubber, not to mention enduring political instability, Brazil has been aptly described as “the country of the future [that] always will be.” But even setting aside the dismissive qualifier, the first part of the label is genuinely meaningful: South America’s largest nation is a microcosm of the world, with immigrants from across the Old World. It has the largest African-descended population outside Africa, the biggest Japanese population outside Japan, and the largest Lebanese-descended population in the world (dwarfing Lebanon itself). 

But just as Brazil remains a nation of the future, where disparate streams of human migration continue to recombine, it also has a deep past. The bedrock of Brazil’s demographic past is its  indigenous citizens, whose roots in the region go back tens of thousands of years, and whose blood still courses through the veins of most Brazilians. And deep in the rainforests of the Amazon, two tribes carry genes bearing undiluted testament to astonishingly ancient human migrations even prior, tipping us off to a demographic wave until recently wholly unimagined.

The First Ones 

In 2015, Swedish geneticist Pontus Skoglund, was first author on a publication that produced one of the most enigmatic and still mysterious results this fervidly productive century in human population genetics has yet seen, Genetic evidence for two founding populations of the Americas. Skoglund trained cutting edge genomic methods on samples from the Human Genome Diversity Project (the HGDP, a collection comprising 1,043 individuals from 52 populations that researchers had been studying for decades) and found two of the American populations alone, the Suruí and Karitiana, whose DNA contradicts the standard narrative about the peopling of the Americas. Over recent decades, archaeologists had established that 20,000 years ago ancient Siberian mammoth hunters migrated into the lost territory of Beringia between Asia and Alaska, eventually pushing eastward into the New World by 15,000 years ago. What Skoglund and his collaborators found is that though this model adequately explains the North American indigenous populations and the other South America one from Colombia, those two isolated western Amazonian populations alone also carry some genetic ancestry which shares closer roots with the indigenous peoples of the Andaman Islands, New Guinea and Australia (groups we bracket under Australo-Melanesians), than with other eastern Eurasians like Han Chinese or Siberians. Such a result was so inexplicable given what we knew then, and remains so even today, that the researchers recall checking and double-checking, anxious to make sure they weren’t misinterpreting some statistical artifact.

But what even are the origins of two distinct Amazonian ancestry streams, the Beringian and now the Australo-Melanesian-inflected one, and how do they relate to each other? Paleoanthropology and paleogenetics tell us modern humans arrived in eastern Eurasia between 45,000 and 50,000 years ago. The peoples of East Asia, Southeast Asia, Australasia and the Indian subcontinent’s earliest inhabitants all descend from these very first modern settlers of eastern Eurasia. Therefore, upon removing the confounds of both Denisovan admixture in Papuans and Australians and recent gene flow from western Eurasia into East Asia (like Iranian into Monglia), we see that Australo-Melanesians and Han Chinese are closer to each other genetically than either is to Europeans or to the people of the Middle East. Both still bear shared ancestry from that first great human push eastward. But this ancient connection dates to over 45,000 years ago, when early modern humans approaching Asia’s Pacific shore cleaved off into two streams, one moving northward and one pushing south along the shores of the Indian Ocean, onward to Australia. 

The Beringians’ Paleo-Siberian ancestors descended primarily from the northward migrating branch; totally expected, given Siberia’s position along the far northeastern edge of Eurasia. Very early genetic and fossil analyses of the first indigenous Americans make clear their overwhelming connection to Siberian populations, which led paleoanthropologists to posit the existence of ancient Beringians. What the 2015 paper showed is that over hundreds of thousands of genetic markers, patterns also reveal a subtle but unmistakable signal of affinity to the far southern edge of that migratory wave, an improbable and far-flung connection to the peoples of Australia, New Guinea and Southeast Asia’s indigenous populations. 

In the decade since, other researchers using broader datasets have confirmed this strange signal from what Skoglund and collaborators termed “population Y,” for Ypykuera, the Tupi word for “ancestor. Population Y’s signal turns out to be distributed more broadly throughout South America, from Brazil’s Atlantic seaboard out to the Pacific coast beyond the Andes. Though the vast majority of ancient remains from the New World lack a connection to population Y, a 10,400-year-old specimen from Lagoa Santa in Brazil’s southwestern state of Minas Gerais does bear the signature. This prehistoric finding rules out the subset of South Americans with this heritage just having absorbed some admixture from Australian or Papuan seafarers in the last few thousand years. Population Y’s origins stretch back deep into New World history, dating back to at least the Ice Age, over 11,700 years ago.

With only statistical patterns in DNA to go on, theories about population Y necessarily remain speculative and tentative. Two primary models currently compete to explain the persistence of this unique genetic signature in a New World dominated by Beringian ancestry. The first model posits “ancient population structure” within the Beringians themselves, while the second model is that the New World saw multiple waves of migration.

What would ancient population structure have entailed? It would simply mean some Beringian tribes carried genes from population Y that others lacked. Because all ancient American remains found thus far (save the single individual from Lagoa Santa) lack signals from population Y, under this scenario, population Y would already have accounted for a very small share of the overall gene pool in North and South America 15,000 years ago in order for ancient population structure to explain the patterns we see. Thus the variation in population Y’s presence across the New World today would just reflect its original variation across tribes in Beringia. A strength of this model is that it is entirely compatible with the standard theory of Beringian diversification 15,000 years ago after a multi-thousand-year sojourn in the north Pacific’s lost territories. But, it invites us to reconsider who the Beringians were; instead of a single well-mixed population, were they distinct tribes who varied genetically between themselves? Perhaps they were culturally diverse, reflecting varied origins in the Old World. But granting this presence of the exotic population Y among Beringians, how would an Australo-Melanesian-related population get to Siberia in the first place?

Well, first of all, we need to take into account that the planet’s human geography during the Pleistocene was radically different. Today the genetic and cultural boundary between Australo-Melanesians and other East Eurasian people falls across east-central Indonesia, with a transition zone on islands like Timor. But during the Pleistocene, the line lay much further north and west. Until about 4,000 years ago, the dominant archaeological tradition of Southeast Asia was the Hoabinhian culture, and the people who practiced it were genetically similar to today’s Andaman Islanders. Hoabinhian people and their close relatives were also present in southern China before 2000 BC, and the most ancient layer of modern Tibetan ancestry is derived from a Hoabinhian-related population of foragers who arrived in the highlands tens of thousands of years ago and mixed with Denisovans. It is not implausible then that as we go further back in time the neat division between “northern” and “southern” eastern Eurasians breaks down. If foragers genetically closer to Hoabinhians, and thus to Australo-Melanesians, entered the Tibetan plateau, it is not implausible that some might have continued migrating as far as Siberia itself.

And yet, though the theory of ancient population structure to explain population-Y admixture allows us to retain the Beringian model with minimal changes, one attendant implication is that later admixture should eventually more equitably distribute that unique ancestry component across most groups, as opposed to it being concentrated solely in a band across northern South America. This condition doesn’t seem to hold for population Y, which is found in only one ancient sample and totally absent from North America to this day. This discrepancy suggests another possibility: that population Y arrived in a distinct migration from the Beringian one. Though this adds complexity to the model, it makes it more compatible with population Y’s very patchy distribution across modern populations and its near absence from ancient Beringian samples. Though some fantastical scenarios have been proposed starring South Pacific voyagers on the move during the Ice Age, the most likely one is that population Y traversed the same general route as Beringians along the north Pacific since this was the only contiguous portion spanning the Old World and the New; they simply would have done so sooner. Beringians and population Y could also have arrived ‌simultaneously over the prehistoric land-bridge (or its fringe), but the absence of population-Y results from the ancient DNA data set, and its low frequency and geographic restriction to a northern band across South America today, argue against this. An elegant explanation is that though population Y arrived earlier, in most locales they were subsequently replaced by a tidal wave of Beringians starting around 13000 BC.

Just a decade ago, the possibility of a pre-Beringian population would not even have been on the table; the earliest archaeological site with human artifacts in the New World was Monte Verde in Chile, dating to 14,500 years ago. But recent credible discoveries have exploded our parameters of the possible. Imprints of human footsteps dating to more than 21,000 years ago have been found in New Mexico, clear evidence of human occupation predating the Last Glacial Maximum, when massively expanded glaciers left the path between Old and New World blocked and impassable. Further south, caves indicate human presence in Mexico more than 20,000 years ago, with some lithic artifacts and charcoal coming back with radiocarbon dating as early as 33,000 years ago, pushing us towards a radical reordering of the chronology of human settlement of the Americas. Though no definitive consensus has yet emerged, ‌the field appears to be steadily converging on the reality that the Beringians were in fact latecomers, and an earlier wave of modern humans settled the New World long prior. The earlier population seems to have left far fewer archaeological remains though, and the lack of megafauna extinction events until the last 15,000 years suggests any earlier arrivals did not engage in big game hunting like the Beringians. The near complete genetic dominance of Beringians today is understandable when we compare the widespread archaeological remains that marked their arrival with the much less tangible impact of their predecessors. Prior to the Beringians, any human imprint on the Americas was a shadow of what would follow, its faintness indicative of a landscape that was only ever thinly populated. Putative pre-Beringian sites are only coming to light with the latest archaeological techniques, whereas the tools of their successors are commonly found alongside the remains of slain mammoths. It was the Beringians who transformed the New World’s ecology, hunting to extinction all the large mammals and transforming the landscape with their schedule of intentional fires.

But the earlier human populations did not totally disappear. Their legacy persists across South America in the signatures of population Y, and this heritage continues to stamp the millions of Brazilians today who carry ancestry from the indigenous tribes and nations who ranged from Amazon to coastal Atlantic forest, and were destined to be assimilated into the Portuguese settler population during the colonial period. Modern Brazil’s deepest genetic roots reach back into a past that we only now begin to grasp.

Three races into one