Reading 12,346 genomes…
12,346 ancient genomes · 45,000 years
Europe’s three ancestral peoples: who they were, when they came, how much of each remains
Hunter-gatherers who were here after the Ice Age. Farmers who arrived from Anatolia about 8,500 years ago. Herders who came off the steppe about 5,000 years ago. Most people with European ancestors descend from all three. Each has its own colour on this map, shown where and when a skeleton’s DNA has been read: 12,346 people so far.
Look up your own lineScroll to start at the beginning, 45,000 years agoMore than 45,000 years ago
The earliest Europeans whose DNA has been read are not among the three#
The oldest people on this map lived more than 45,000 years ago, at Ranis in Germany, Zlatý kůň in Czechia and Bacho Kiro in Bulgaria. A man from Peștera cu Oase in Romania, about 40,000 years old, had a Neanderthal ancestor four to six generations back.
Later Europeans carry little or nothing of them. The people of Ranis and Zlatý kůň left no ancestry that can be detected in the hunter-gatherers who followed, and those of Bacho Kiro are more closely related to ancient and living people of East Asia and the Americas than to later Europeans.
The map draws the 62 Europeans older than 14,500 years in the hunter-gatherer blue because that is how they lived. The three ancestries of this story begin after the Ice Age, whose refuges have a map of their own.
14,000 to 8,700 years ago
After the ice, Europe belongs to hunter-gatherers#
From about 14,000 years ago hunter-gatherers of one shared ancestry, first seen in a skeleton from Villabruna in Italy, are found across western and central Europe. Further east, from the Baltic to the Urals, lived a related people. Both are blue here.
The best known of them is Cheddar Man, who died in Somerset about 10,000 years ago. His genome predicts dark skin, dark hair and blue or green eyes: a prediction from genes whose effects were measured in people living today.
Of the 81 people sampled in Europe between 14,500 and 8,700 years ago, hunter-gatherer ancestry averages 90%. Nearly all the rest belongs to the foragers of the Iron Gates on the Danube: this page’s model reads about an eighth of their ancestry as Anatolian. They were not farmers. A 2019 study found that, of all Europe’s hunter-gatherers, they were the closest kin of the hunter-gatherers of Anatolia.
8,700 to 7,000 years ago
Farmers arrive from Anatolia: the people moved, not only the idea#
Farming began in the Near East. Whether the idea travelled to Europe or the people did was argued for decades. The map settles it: yellow enters at the Aegean and moves along two roads, up the Danube into central Europe and along the Mediterranean coast to Iberia.
The earliest people with mostly farmer ancestry in this archive date to about 8,200 years ago in Greece, 8,100 in Serbia, 7,700 in Hungary, 7,200 in Germany, 7,200 in Spain and 5,800 in Britain and Ireland.
At first the newcomers and the locals mixed little: the first farmers of central Europe are 95% Anatolian by descent. Between 7,500 and 7,000 years ago the 379 people sampled across Europe average 78% farmer ancestry.
DNA shows that people moved and mixed, not why. Across Europe farming spread at an estimated average of about a kilometre a year, in jumps and pauses that the average hides.
7,000 to 5,300 years ago
For two thousand years the two peoples live side by side#
The hunter-gatherers did not vanish. Around the Baltic and in the north they went on fishing and hunting while farmers held the south, and the blue stays on the map there. At the Blätterhöhle cave in Germany, people who lived by foraging kept that way of life for some 2,000 years after farmers arrived.
Then the two merged. The hunter-gatherer share of the people sampled in Europe is 16% between 7,000 and 6,000 years ago and 23% between 6,000 and 5,300: the yellow turns green. Geneticists call it the resurgence of hunter-gatherer ancestry.
7,000 to 5,300 years ago, further east
On the steppe, a third people is forming#
On the grasslands north of the Black Sea and the Caspian, eastern hunter-gatherers were mixing with people whose ancestry came from the Caucasus. Their descendants kept herds and buried their dead under mounds. They are known by the name of one of their cultures: the Yamnaya.
This page reads the Yamnaya of the Samara region as 47% eastern hunter-gatherer, 44% Caucasus and 9% Anatolian-farmer ancestry; published estimates put the first two at about half each. A 2025 study places their formation about 6,000 years ago, among descendants of the Serednii Stih people north of the Black Sea, and their expansion from about 5,300 years ago.
On the map, the blue and grey of the steppe turn red as that mixture forms. From here on it is the third colour.
5,300 to 4,000 years ago
The steppe arrives, and within a few centuries the map of Europe changes colour#
The red moves west fast. The earliest people with at least 40% steppe ancestry date to about 5,200 years ago in Moldova, 4,900 in Bulgaria, 4,900 in Hungary, 4,800 in Romania, 4,500 in Germany, 4,300 in Britain and 4,100 in Spain.
The people of the Corded Ware culture in Germany, named after their pottery, appear at that time. In this page’s calculation they are 74% steppe; a 2015 study put it at about three quarters.
Between 4,400 and 4,000 years ago the 404 people sampled in Europe average 47% steppe, 41% farmer and 12% hunter-gatherer ancestry. All three are now in place.
What made the herders mobile was not cavalry. Wheeled vehicles appear in several regions around 5,500 years ago, and where they were invented is disputed. Widespread horse-based transport developed only around 4,200 years ago, after this arrival, and chariots about 4,000 years ago. Why the move began is not settled: plague DNA documents repeated infections among Scandinavian farmers of that time, but does not show why the steppe people moved.
Britain, 4,500 years ago
Britain changes hands twice#
Britain shows the sequence most sharply. Farmers reached it about 6,000 years ago: the 139 people sampled in Britain and Ireland between 6,100 and 4,500 years ago average 77% farmer ancestry, 22% hunter-gatherer, and none from the steppe.
About 4,400 years ago people who used Beaker pottery crossed from the continent. The 41 people sampled from the following five centuries average 54% steppe ancestry. The study that found this concluded that about 90% of Britain’s gene pool was replaced within a few hundred years.
In Iberia the same arrival replaced about 40% of the ancestry and nearly all the male lines by 4,000 years ago.
Britain since
After the Beakers, Britain’s three shares barely move#
Romans, Anglo-Saxons and Vikings all came later. In these three colours they are almost invisible. Iron Age Britain averages 50% steppe ancestry (299 people), early medieval Britain 51% (190 people), and the English sampled today 48%.
The newcomers from across the North Sea carried nearly the same mixture as the people they joined. Finer methods, which compare recent populations directly, can tell them apart. A map of three ancient ancestries cannot, and that is its limit.
4,000 to 1,000 years ago
Later migrations move the proportions, not the palette#
After the Bronze Age no migration brings a fourth ancestry to the whole continent, though several change whole regions. Between 3,000 and 2,000 years ago the 1,047 people sampled in Europe average 49% farmer, 39% steppe and 10% hunter-gatherer ancestry, close to today’s.
Where the map turns grey, people have arrived whose ancestry lies outside the three. In Italy the people sampled from the imperial centuries carry 24% of it, against 2% in the Iron Age: Rome drew people from the eastern Mediterranean. In Hungary the grey of 1,400 years ago lies over graves of the Avar period: of the 550 people sampled there, 35% of the ancestry comes from outside the three.
The Slavic migrations after that changed more: a 2025 study estimates that over 80% of the ancestry in eastern Germany and in Poland was replaced. The migrants carried the same three ancestries in northern proportions, so here they show only as a change of hue.
The great break of this stretch hardly shows in three colours. Around 3,200 years ago many palace centres in Greece, Anatolia and the Levant were destroyed or abandoned, and Egyptian inscriptions describe attacks by groups now called “Sea Peoples”. DNA from the city of Ashkelon shows a southern-European-related ancestry that arrives about then and is no longer detectable within two centuries. Palace collapse alone does not show that one people replaced another, and three colours cannot name the Sea Peoples.
Today
Almost every group has all three, and the proportions follow latitude#
Each disc is a present-day group sampled by geneticists, drawn where the samples were taken and divided into its shares. Steppe ancestry is highest in the north and east: 55% in Lithuanians, 51% in Norwegians, 48% in the English. Farmer ancestry is highest in the south: 72% in Sardinians, who have only 8% from the steppe.
Hunter-gatherer ancestry, the oldest, is the smallest share everywhere: 19% in Lithuanians, 13% in Basques, 5% in Romanians.
The samples are small: 10 Romanians, 10 English, 10 Lithuanians. The archive has no present-day sample of Germans, Poles or Swedes, among others, so those countries have no disc.
Genes, languages, borders
The map of genes does not follow the map of languages#
A ring marks a group whose language is not Indo-European. To compare two groups, this page adds up how far apart their four shares are: 0 means the same mixture, 100 means nothing in common.
Hungarian is a Uralic language; its closest relatives are spoken east of the Urals, by the Mansi and the Khanty. In ancestry, Hungarians are 3 points from Czechs and 54 from the Mansi. The Gagauz of Moldova speak a Turkic language and are 4 points from their Moldovan neighbours, 29 from Turks.
Basques speak a language related to no other in Europe, yet stand 1 point from other northern Spaniards and have 33% steppe ancestry. Romanians are 4 points from Serbs, but 30 from Sardinians, with whom they share Latin.
A language can spread without replacing people, and people can move and keep the language they find. The steppe migration is the leading explanation for how Indo-European languages reached Europe; the 2015 study of that migration said its results support this for “at least some” of them. DNA cannot show what anyone spoke.
What three colours leave out
Where the three-source model stops working#
The grey share. Southern and south-eastern Europe carry ancestry that arrived later from the eastern Mediterranean and beyond: 16% in Greeks, 22% in Sicilians, 15% in Bulgarians by this count. Finns have 10% Siberian-related ancestry, which reached northern Europe at least 3,500 years ago.
Blank paper is missing evidence, not empty land. Colour appears only near skeletons whose DNA has been read. Laboratories have read 1,298 from Hungary and 195 from Romania.
Steppe ancestry is itself a mixture, nearly half of it eastern hunter-gatherer. Counted that way, northern Europeans have more hunter-gatherer ancestry than their blue share shows.
The coast is today’s. When the hunter-gatherers lived, part of the North Sea was land: Doggerland.
The whole map, any moment, and your own line on it
Move the clock, drag the map, press any dot to meet the person. Or type a father line or mother line from a DNA test.
Drag to move, pinch or use the buttons to zoom.
Finer branches, as a test reports themStep by step, from the oldest branch to this one
The people, oldest first
A line is one ancestor per generation out of thousands: the father’s father’s father, or the mother’s mother’s mother. It says where that one thread passed, not where your ancestors lived. How many ancestors you have is a different sum.
What a line is. The father line is the Y chromosome, which a father passes to his sons almost unchanged. The mother line is the DNA of the mitochondria, which a mother passes to all her children. Small copying errors add up over the generations and each gets a name: the longer the name, the finer the branch. A DNA test reports the finest branch it can read.
Why they are worth a look. Lines changed more sharply than ancestry as a whole, so the three arrivals show in them clearly. The shares below are of the sequenced people of Europe whose line could be read, in four periods: before farming (more than 8,700 years ago), the farming millennia (8,700 to 5,300), the next 2,300 years, and the last 3,000.
The big father lines
The big mother lines
How the shares were computed
Every number and every colour on this page comes from one public file: release v66.p1 of the Allen Ancient DNA Resource, in which David Reich’s laboratory at Harvard gathers the genotypes published by ancient-DNA laboratories around the world. The page uses 584,131 positions of the genome typed in both ancient and living people, one record per person, and only records the archive marks as passing its quality checks.
The question asked of each person
Which mixture of source populations shares DNA variants with a set of distant reference populations in the same way this person does? This is the approach of the qpAdm program used in the papers listed below, written again for this page so that it could be run on every person in the archive. 368,049 positions had data in all the references and were used.
For people who lived more than 5,300 years ago the sources are western hunter-gatherers (40 Mesolithic people from Luxembourg, Spain, France, Switzerland, Belgium, England, Hungary and Italy), eastern hunter-gatherers (19 from Karelia and the Volga), early farmers of Anatolia (47) and hunter-gatherers of the Caucasus (3). For later people they are western hunter-gatherers, the same Anatolian farmers and the Yamnaya of Samara (30). The reference populations are Mbuti, Ust’-Ishim, Kostenki and Sunghir, Mal’ta and Afontova Gora, Han, Papuan, Onge, Karitiana, early Levantines, the Iberomaurusians of Morocco, early farmers of Iran and the Nganasan.
Three colours from those sources
Blue is western plus eastern hunter-gatherer ancestry. Yellow is Anatolian-farmer ancestry. Red is Yamnaya ancestry; before the Yamnaya existed, it is eastern-hunter-gatherer and Caucasus ancestry together, in the proportions the Yamnaya have, counted from 7,000 years ago and only when the Caucasus share is clearly above zero. Grey is everything else. People older than 14,500 years in Europe are drawn blue without a calculation.
A fourth source is allowed for later people: ancestry related to early Iran, or to the Nganasan of Siberia. For one person it is kept only at 5% or more and three standard errors from zero. Siberian-related ancestry needs the same in groups, because Sardinians and Basques, who have no known Siberian ancestors, get 2–4% in this method: living and ancient people are typed with different technologies.
How sure each dot is
A person’s own estimate is used when at least 60,000 positions were read and the standard error is at most 10 points: 9,278 of the 12,346 people on the map. For 2,713 others the colour is the estimate of their archive group taken together. 355 have too little DNA for either and are drawn as empty rings. Standard errors come from leaving out one stretch of the genome at a time (129 stretches).
The colour and the dates
How strong the colour is at a spot grows with the number of sequenced people buried within about 110 km and within a few centuries of the clock (longer for older periods); the colour itself is their average. The last frame has no wash: living groups are discs at the place where they were sampled. “The earliest people with…” in the text means the second-oldest such person in a country, so that a single misdated skeleton cannot set the date. Dates are years before 1950, as the archive gives them.
The distance between two present-day groups is half the sum of the differences between their four shares: the part of one mixture that would have to change to make it the other.
Checked against published tables
The same three-source model, run on groups for which the original studies printed their own estimates. Shares are western hunter-gatherer / Anatolian farmer / steppe, in percent. The samples and the reference populations differ somewhat from study to study, so agreement within a few points is what can be expected.
| Group | People here | This page | Published | Published in |
|---|---|---|---|---|
| Corded Ware, Germany | 20 | 6.8 / 18.8 / 74.4 | 11.1 / 18.3 / 70.6 | Olalde 2018, Table S4 |
| Bell Beaker, Britain | 19 | 12.4 / 28.5 / 59.1 | 16.2 / 26.0 / 57.8 | Olalde 2018, Table S4 |
| First farmers, Britain | 32 | 21.3 / 78.4 / 0.3 | 20.8 / 76.7 / 2.6 | Patterson 2022, Extended Data Table 1 |
| First farmers, central Europe (LBK) | 64 | 6.1 / 95.2 / 0.0 | 4.8 / 95.2 / 0.0 | Olalde 2018, Table S5 |
| Early farmers, Iberia | 16 | 13.4 / 86.5 / 0.1 | 12.6 / 88.3 / 0.0 | Olalde 2018, Table S5 |
| Globular Amphora, Poland | 33 | 23.5 / 73.2 / 3.3 | 25.0 / – / 0.0 | Mathieson 2018, main text |
What a father line or mother line can tell you
The archive records a father line for 6,331 of the men on the map and a mother line for 9,340 people. Father lines are placed on version 12.03 of the YFull tree, the one the archive used, so that “carried R-M269” means the man’s own marker is R-M269 or lies on a branch below it; 6,128 could be placed. Mother lines follow the standard naming, in which U5a2a lies inside U5a2, inside U5a, inside U5.
A man placed on a branch tens of thousands of years ago belongs to the lineage that led to it; he did not necessarily carry every marker that defines it today. The ages quoted for father lines are YFull’s estimates.
A line is a single thread. Ten generations back a person has up to 1,024 ancestors, and the father line and the mother line are two of them.
Sources
- Allen Ancient DNA Resource, release v66.p1 (2026), Harvard Dataverse, CC0: the data.
- Mallick S. et al., “The Allen Ancient DNA Resource (AADR): a curated compendium of ancient human genomes”, Scientific Data 11 (2024).
- Haak W. et al., “Massive migration from the steppe was a source for Indo-European languages in Europe”, Nature 522 (2015).
- Lazaridis I. et al., “Ancient human genomes suggest three ancestral populations for present-day Europeans”, Nature 513 (2014).
- Lazaridis I. et al., “Genomic insights into the origin of farming in the ancient Near East”, Nature 536 (2016).
- Fu Q. et al., “An early modern human from Romania with a recent Neanderthal ancestor”, Nature 524 (2015).
- Fu Q. et al., “The genetic history of Ice Age Europe”, Nature 534 (2016).
- Hajdinjak M. et al., “Initial Upper Palaeolithic humans in Europe had recent Neanderthal ancestry”, Nature 592 (2021).
- Sümer A. P. et al., “Earliest modern human genomes constrain timing of Neanderthal admixture”, Nature 638 (2025).
- Brace S. et al., “Ancient genomes indicate population replacement in Early Neolithic Britain”, Nature Ecology & Evolution 3 (2019).
- Feldman M. et al., “Late Pleistocene human genome suggests a local origin for the first farmers of central Anatolia”, Nature Communications 10 (2019).
- Mathieson I. et al., “The genomic history of southeastern Europe”, Nature 555 (2018).
- Hofmanová Z. et al., “Early farmers from across Europe directly descended from Neolithic Aegeans”, PNAS 113 (2016).
- Bollongino R. et al., “2000 years of parallel societies in Stone Age Central Europe”, Science 342 (2013).
- Lipson M. et al., “Parallel palaeogenomic transects reveal complex genetic history of early European farmers”, Nature 551 (2017).
- Nikitin A. G. et al., “A genomic history of the North Pontic Region from the Neolithic to the Bronze Age”, Nature 639 (2025).
- Olalde I. et al., “The Beaker phenomenon and the genomic transformation of northwest Europe”, Nature 555 (2018).
- Olalde I. et al., “The genomic history of the Iberian Peninsula over the past 8000 years”, Science 363 (2019).
- Patterson N. et al., “Large-scale migration into Britain during the Middle to Late Bronze Age”, Nature 601 (2022).
- Antonio M. L. et al., “Ancient Rome: a genetic crossroads of Europe and the Mediterranean”, Science 366 (2019).
- Gretzinger J. et al., “Ancient DNA connects large-scale migration with the spread of Slavs”, Nature 646 (2025).
- Lamnidis T. C. et al., “Ancient Fennoscandian genomes reveal origin and spread of Siberian ancestry in Europe”, Nature Communications 9 (2018).
- YFull YTree v12.03, the tree of father lines, CC BY 4.0.
- ETOPO 2022 global relief, NOAA National Centers for Environmental Information (relief of the base map).
- Natural Earth, 1:50 million coast, lakes, rivers and borders, public domain.