Hero image: AI generated illustration of a Late Bronze Age plank boat approaching a rocky Atlantic headland at dawn. Not a reconstruction of any specific site, vessel or person.

Between roughly 1300 and 700 BC, the coasts from Portugal to Ireland and Scotland shared swords, cauldrons, feasting gear and gold ornaments so similar that archaeologists speak of a single Atlantic Bronze Age. Shared objects do not prove shared people. Ancient DNA now lets us ask the harder question: along this seaboard, who actually moved, how many of them, and did they move as whole communities or as individual men and women marrying into neighbouring communities? To answer, we follow the coast from the Bell Beaker baseline around 2400 BC to the early Middle Ages. The answer differs sharply between the two ends of the facade, and the Iberian end holds a Y-chromosome puzzle that autosomal data alone cannot solve.

Two ends of one coast

By about 2000 BC, both Britain and Iberia had already absorbed the great steppe-related expansion of the Bell Beaker period, but on very different terms. In Britain the newcomers replaced most of the earlier Neolithic gene pool. In Iberia, the genome-wide transect published by Olalde and colleagues in 2019 found that people with steppe ancestry replaced about 40 percent of the local ancestry but nearly 100 percent of the Y chromosomes over a few centuries, with R1b-M269 lineages taking over. The same study found that during the Iron Age, steppe ancestry kept spreading, reaching even populations that did not speak Indo-European languages, and that present-day Basques look genetically like a typical Iron Age Iberian population that escaped most later admixture.

So the starting point of the Atlantic Bronze Age was a gradient, not a uniform population: steppe-rich Britain and Ireland in the north, farmer-rich Iberia in the south, France in between. The question for the Late Bronze Age and Iron Age is whether that gradient moved.

What moved: the two ends converged

A simple three-way Global25 model (Anatolian Neolithic farmer, Samara Yamnaya, and the Luxembourg hunter-gatherer Loschbour; condition number 2.4, poles separated by 338 to 495 units of Global25 distance, that is distance times one thousand) is coarse, with fit distances of 34 to 58, but it is good enough to track direction over time. It shows the two ends of the coast moving toward each other.

Early farmer share at both ends of the Atlantic facadeThree-way Global25 NNLS model (Anatolian farmer, Yamnaya steppe, western hunter-gatherer). Early farmer share in percent.Southern BritainScotlandBasque, Navarre and Vasconic groupsNortheast IberiaBritain: axis 25 to 45 percent2530354045EBAMBALBAEIALate IAE. Med.TodaySouthern Britain, EBA: 27.8 percentSouthern Britain, MBA: 33.7 percentSouthern Britain, LBA: 37.3 percentSouthern Britain, EIA: 39.6 percentSouthern Britain, Late IA: 38.7 percentSouthern Britain, Today: 39.2 percentScotland, MBA: 33.6 percentScotland, LBA: 34.9 percentScotland, Late IA: 33.8 percentScotland, Today: 37.6 percentNorthern Iberia: axis 50 to 70 percent5055606570EBAMBALBAEIALate IAE. Med.TodayBasque, Navarre and Vasconic groups, EBA: 61.6 percentBasque, Navarre and Vasconic groups, MBA: 64.5 percentBasque, Navarre and Vasconic groups, LBA: 59.3 percentBasque, Navarre and Vasconic groups, EIA: 58.2 percentBasque, Navarre and Vasconic groups, Late IA: 54.7 percentBasque, Navarre and Vasconic groups, E. Med.: 54.9 percentBasque, Navarre and Vasconic groups, Today: 56.8 percentNortheast Iberia, EBA: 62.1 percentNortheast Iberia, LBA: 61.1 percentNortheast Iberia, Late IA: 59.4 percentBoth axes span 20 points, so slopes are comparable. Britain gains farmer ancestry after the Middle Bronze Age,while northern Iberia loses it as steppe-rich ancestry keeps arriving. The two ends of the facade converge.

Early farmer share in the three-way model, by period, at the northern and southern ends of the Atlantic facade. Late IA for Britain is the Middle Iron Age group (England MIA); for the Basque area it is the Iron Age group labelled Celtiberian, and E. Med. is the early medieval Vasconic-profile group from Burgos.

In southern Britain the early farmer share climbs from 33.7 percent in the Middle Bronze Age to 37.3 percent in the Late Bronze Age and 39.6 percent in the Early Iron Age, then holds. Scotland barely moves: 33.6 percent in the Middle Bronze Age, 33.8 percent in the Late Iron Age. This is exactly the pattern reported by Patterson and colleagues in 2022 from 793 newly reported genomes from Britain and continental Europe: a rise in early European farmer ancestry in England and Wales between about 1000 and 875 BC, absent from Scotland, carried by migrants genetically closest to ancient people from France, who contributed roughly half the ancestry of Iron Age England and Wales.

At the Iberian end the movement runs the other way. In the Basque Country and Navarre, the farmer share sits at 61.6 percent in the Early Bronze Age and 64.5 percent in the Middle Bronze Age, then falls to 59.3 percent in the Late Bronze Age, 58.2 percent in the Early Iron Age of Navarre and 54.7 percent in the Iron Age group, with steppe-related ancestry rising from 16 to 18 percent to about 28 percent. The Middle and Late Bronze Age Basque points are single genomes, so the timing of the turn is loose; the direction, confirmed by the two-genome Navarre and three-genome Iron Age groups, is not. Northeast Iberia shows a weaker version of the same drift. Britain was receiving people from a more southern, farmer-rich continent; northern Iberia was receiving people from a more northern, steppe-rich one.

One cline from Ireland to Portugal

A local PCA run on 23 ancient groups from the facade makes the geometry plain. One axis carries 82 percent of all the variance, and the groups fall on it in broadly geographic order: Ireland and Scotland at one end, Portugal at the other, France in the middle. Within each region, time matters as much as place: Iron Age England has slid toward France, and the Iron Age Basque group, the early medieval Vasconic group from Burgos, and the late antique Vasconic individual from Saint-Lupien in France all sit in the same stretch of the cline as Iron Age Navarre, the Cantabri and Iron Age Occitanie, far from Britain.

One cline from Ireland to PortugalPosition of each ancient group on the first axis of a local PCA run on these 23 groups. This single axis carries 82 percent of the variance.Britain and IrelandFranceIberia-60-40-200204060Ireland EBAIreland EBA: 65.7Scotland LIAScotland LIA: 56.4England MBAEngland MBA: 56.3Scotland MBAScotland MBA: 55.7England LBAEngland LBA: 45.5Arras (E Yorks)Arras (E Yorks): 43.4England MIAEngland MIA: 40.1Hauts-de-France IA1Hauts-de-France IA1: 29.2Brittany EBABrittany EBA: 13.8Burgundy HallstattBurgundy Hallstatt: 1.9Occitanie La TeneOccitanie La Tene: -12.8Occitanie LBAOccitanie LBA: -16.5Basque Country IABasque Country IA: -16.9Vasconic EM (Burgos)Vasconic EM (Burgos): -19.4Saint-Lupien VasconicSaint-Lupien Vasconic: -27.7Navarre EIANavarre EIA: -30.8Cantabri LIACantabri LIA: -31.8Asturias LBAAsturias LBA: -33.3Catalonia IACatalonia IA: -33.5Basque Country LBABasque Country LBA: -37.3Aragon Urnfield LBAAragon Urnfield LBA: -42.2Basque Country EBABasque Country EBA: -45.5Portugal MBAPortugal MBA: -60.3Local PC1 times one thousand. Left: more early farmer ancestry. Right: more steppe and hunter-gatherer ancestry.

Local PC1 for ancient Atlantic groups, computed on Global25 scaled coordinates. Sample sizes range from one individual to several dozen; single-individual groups are noisy.

Distances confirm it. England MIA sits closest to other British Iron Age groups (6.7 from the Arras culture, 10.3 from England LBA), but its next nearest neighbour is the Hallstatt-period population of Hauts-de-France at 15.7, closer than England's own Middle Bronze Age at 18.7. The early medieval Vasconic group from Burgos is closest to Iron Age Occitanie (17.6), then to Iron Age Basques (21.8) and the Cantabri (21.9). No Iberian group in this panel comes within 60 units of any English group in it.

How big was the influx into Britain? A lesson in proxies

Modelling Iron Age England as a mixture of its own Middle Bronze Age population and a French source should recover the size of the Late Bronze Age migration. It does, but only with the closest proxy. With the Hallstatt-period population of Hauts-de-France as the source, NNLS assigns 55.6 percent to the newcomers, close to the published figure of about half. With more distant, farmer-rich French sources, the estimate drops to between 16.5 and 28 percent.

How big was the migration? It depends on the proxyEngland MIA modelled as England MBA plus one French source. Bar: share assigned to the French source, percent.010203040506070Patterson et al. 2022: about halfHallstatt, Hauts-de-France (IA1)Hallstatt, Hauts-de-France (IA1): 55.6 percent55.6 sep 30 cond 13.7Early Bronze Age BrittanyEarly Bronze Age Brittany: 28.0 percent28.0 sep 51 cond 8.2Late Bronze Age OccitanieLate Bronze Age Occitanie: 21.5 percent21.5 sep 77 cond 5.5Late Bronze Age Grand EstLate Bronze Age Grand Est: 16.5 percent16.5 sep 88 cond 4.6Sep: Global25 distance between the two sources times one thousand. Cond: condition number. A source closer to the real migrants absorbs a larger share.The more distant, farmer-rich sources dilute the estimate: a smaller share of something more extreme produces the same shift.

The same target, four different sources, four different answers. Fit distances: 8.2 (Hauts-de-France), 11.9 (Brittany EBA), 8.6 (Occitanie LBA), 11.7 (Grand Est LBA).

Neither result is wrong in its own terms. A source that is more extreme than the real migrants needs only a small share to produce the same shift, so the share shrinks as the proxy moves away. The price of the right answer is collinearity: the Hauts-de-France model has a condition number of 13.7 and only 30 units between its two sources. There is also a circularity to keep in mind: Fischer and colleagues report gene flow between Iron Age northern France and England, so the best-fitting proxy is partly shaped by the very exchange being measured. That is why the published qpAdm figure of about half is the headline here, and our NNLS numbers serve as directional support.

Northern Iberia: British or continental newcomers?

France itself offers a clue. In 2022, Fischer and colleagues found no genetic break between Bronze Age and Iron Age groups in France, and strong homogeneity across regional cultures, maintained by gene flow with neighbouring groups in England and Spain. France was less a source of one migration than a crossroads of constant exchange.

The steppe-rich ancestry arriving in northern Iberia in the Late Bronze Age and Iron Age could in principle come from Britain, from northern France, or from central France and the Hallstatt world. We put all three in competition, on top of a local Bronze Age base.

TargetLocal baseEngland MIAHauts-de-France IA1Burgundy HallstattFit
Navarre EIA (n=2)59.00.00.041.025.0
Basque Country IA (n=3)31.40.00.068.622.9
Cantabri LIA (n=2)51.30.00.048.728.0
Catalonia IA (n=7)68.90.00.031.116.7
Vasconic EM, Burgos (n=10)46.210.40.043.321.4
Asturias LBA (n=1)80.619.40.00.026.4

Local base: Basque Country EBA for the Basque and Navarre targets, Castile and Leon MBA for Cantabri and Asturias, Catalonia EBA for Catalonia. Condition numbers 58 to 59. England MIA and Hauts-de-France IA1 are only 15.7 units apart and cannot be told apart by this method.

In four of six targets the northern component goes entirely to the Burgundy Hallstatt population, the source closest to the Urnfield and Hallstatt world of central France. This fits the archaeological picture of Urnfield influence spreading across the Pyrenees, and a Late Bronze Age Urnfield group from Aragon (n=14) is in fact the nearest ancient neighbour of Iron Age Navarre (22.6). The two exceptions call for caution. The early medieval Vasconic group gets 10.4 percent England MIA, but across 500 noise-perturbed replicates that share runs 10.3 plus or minus 5.6 percent, and models with other source sets move it anywhere between 4.8 and 28.5 percent. The lone Asturias LBA individual gets 19.4 percent, yet in 11 percent of replicates the English share falls to zero. A single genome cannot carry a migration story.

These shares are not comparable to the steppe percentages in the three-way model above. The Burgundy Hallstatt source is itself largely early farmer, so a 40 to 60 percent share of it delivers only about 10 points of extra steppe ancestry, which is what the three-way model records.

The honest conclusion: Global25 sees a consistent but variable northern, Hallstatt-like input into Late Bronze Age and Iron Age northern Iberia, between 19 and 69 percent across targets, and it cannot see a specifically British one. That is not evidence that no Britons arrived. England MIA and Iron Age northern France are nearly identical in this space, so a British contribution would simply be booked to whichever of the two fits the noise. A spike-in test makes the blindness explicit: when we add 30 percent England MIA to the Iron Age Basque group, the competition model still returns only 17 percent as English on average, and zero in more than one replicate in twenty.

Women stayed, men moved

Iron Age Britain adds a social dimension that changes how migration leaves its trace. In 2025, Cassidy and colleagues analysed 57 genomes from Durotrigian burials, 55 of them from the cemetery at Winterborne Kingston in Dorset, and found that most of them descended through the maternal line from a single woman: women stayed in their natal community, while husbands came in from elsewhere. They found the same pattern of dominant matrilines across Iron Age Britain; Dan Bradley, one of the authors, pointed to a Yorkshire matriline already established before 400 BC.

In August 2026 a much larger study from the Arras culture of East Yorkshire, by Inigo Olalde, Ian Armit, Lindsey Buster, Malcolm Lillie and colleagues, reported genome-wide data for 534 individuals from Wetwang Slack, Pocklington and Melton. At Wetwang Slack, two mitochondrial lineages, T2e1a1b and H1ao, account for 51 percent of individuals, and a pedigree of 195 people spans 13 generations. The authors describe communities organised along matrilineal lines, with recurrent unions between dominant matrilines, while the most common paternal lineage is R1b-DF13, the branch of R1b-L21 that dominates Britain and Ireland today.

A system in which men leave home at marriage and women anchor the lineage moves Y chromosomes across the map far faster than it moves autosomal ancestry. A man who marries into a neighbouring community, generation after generation, spreads his paternal line without shifting the genome-wide average of the population he joins.

The Iberian L21 puzzle

That brings us to the puzzle raised in a recent GenArchivist forum thread on Iberian R-L21. Participant alanarchae2 recalled that L21 reaches notable frequencies among Basques, that much of it was later found to belong to an Iberia-specific cluster, and that L21 has, to his knowledge, not been reported in Iberian Beaker, Bronze Age or Iron Age ancient DNA. Participant miquirumba, an administrator of the FTDNA Catalonia and Eastern Iberia project who has studied Iberian L21 lineages for eight years, replied with a detailed map. According to his reading of FTDNA and yfull data, the main Basque L21 clade is R-Z2189, with two branches: R-BY279, most common in Navarre and parts of Euskadi but also found in native Britons and Scots, and R-DF73, a large Iberian clade present almost everywhere in Iberia except Catalonia. He reports further old Basque clades such as R-S3530, British-linked subclades at lower frequency, northwest Iberian clades that share ancestors with the British R-DF21 and R-DF49 groups less than 3000 years ago, and a Portuguese clade, R-Y134022, that he attributes to a Late Bronze Age arrival. He also points to an anonymised early medieval Basque in the Akbari et al. dataset, I38909, reported as R-S3530 with maternal haplogroup H2a5a1.

These phylogenetic claims come from citizen science trees rather than peer-reviewed ancient DNA, and we have not verified the I38909 haplogroups independently. But the model they imply is testable in principle: a few L21 men from southern Britain or northern France marrying into patrilocal Basque and Aquitanian communities during the Late Bronze Age and Iron Age. Two features of the situation at either end of the facade make this plausible rather than paradoxical. Living Basque men are overwhelmingly R1b-P312, at 75 to 87 percent by province, with R-DF27 the dominant branch, and the Iberia-wide ancient record places the replacement that produced this profile in the Bronze Age; incoming L21 men therefore joined a population of close paternal cousins. And Iron Age Britain was a society in which young men routinely left home to marry. The documented moves are between neighbouring communities, a few kilometres apart, so a crossing of the Bay of Biscay remains speculation; but a habit of male mobility makes occasional longer journeys easier to imagine, and over two and a half millennia a handful of founding lineages can drift to notable frequencies.

Can Global25 see such men? Not in the window that matters. For the Bronze and Iron Age, the competition models above are blind to a British input even at 30 percent, because British and northern French Iron Age profiles are interchangeable. For later periods we tested whether present-day Basques need any England MIA ancestry on top of the early medieval Vasconic group. They do not: the share is 0.0 percent. But that test only covers input after the early Middle Ages, and even there the spike-in test shows how little the zero means: adding known amounts of England MIA to the Basque profile with sampling noise, this particular comparison only starts recovering the addition at around 9 to 10 percent. A handful of L21 lineages entering through marriage over several centuries would sit far below that floor in either window. Autosomal averages are the wrong instrument for this question; dated Y-chromosome trees and more Iberian Iron Age genomes are the right ones.

What this method cannot say

Five limits, stated plainly. Many of the Iberian groups here are one to three individuals, and a single genome can shift an average by more than the effects being measured. Global25 NNLS is a projection method, not qpAdm; condition numbers of 58 in the competition models mean the split between neighbouring northern sources is unstable, even when the total northern share is stable (54 percent, 90 percent interval 47 to 61, for the Vasconic group). Southern English and northern French Iron Age populations are almost identical in this space, so the method cannot distinguish a British from a Gaulish origin for Iberian newcomers, and the detection floors quoted here are specific to each comparison: well-separated models return a floor near 10 percent, poorly separated ones return either large false shares or none at all. The three-way farmer, steppe and hunter-gatherer model fits every group at distances between 34 and 58, useful for direction and not for precise percentages. And none of our analyses touches Y chromosomes directly: the L21 phylogeography discussed above rests on citizen science data that published ancient DNA has yet to confirm.

Myth and measurement

The claimWhat the evidence shows
The Atlantic Bronze Age was a trade network of goods, not people.In Britain it was both. Migrants from the continent contributed about half the ancestry of Iron Age England and Wales between roughly 1000 and 875 BC, while Scotland was barely touched.
The Celtic Iron Age of northern Iberia came from Britain or Ireland by sea.The steppe-rich input into Late Bronze Age and Iron Age northern Iberia is best matched by a Hallstatt-like central French source in Global25. A British contribution cannot be excluded, because British and northern French Iron Age profiles are almost indistinguishable here.
Basques remained genetically sealed off since the Neolithic.Basque Y chromosomes carry the signature of a Bronze Age replacement, inferred from the Iberia-wide ancient record and the composition of living Basque lineages, and their autosomal steppe share kept rising into the Iron Age. Since then the profile has barely changed.
If L21 men entered the Basque Country, it should show in Basque autosomal DNA.Not necessarily. A few men marrying in over generations would add far less than the detection floor of autosomal models, near 10 percent in the best case and far worse when British and French sources compete. Y-chromosome dating is the appropriate test.
Iron Age British society was built around male lineages.Dorset and East Yorkshire cemeteries show dominant maternal lines and incoming husbands, a pattern that spreads paternal lineages faster than autosomal ancestry.

References

  1. Patterson 2022 Patterson N, Isakov M, Booth T, et al. Large-scale migration into Britain during the Middle to Late Bronze Age. Nature 601, 588 to 594 (2022). Read the paper
  2. Olalde 2019 Olalde I, Mallick S, Patterson N, et al. The genomic history of the Iberian Peninsula over the past 8000 years. Science 363, 1230 to 1234 (2019). Read the paper
  3. Fischer 2022 Fischer CE, et al. Origin and mobility of Iron Age Gaulish groups in present-day France revealed through archaeogenomics. iScience 25, 104094 (2022). Read the paper
  4. Cassidy 2025 Cassidy LM, et al. Continental influx and pervasive matrilocality in Iron Age Britain. Nature 637, 1136 to 1142 (2025). Read the paper
  5. Olalde 2026 Olalde I, Armit I, Buster L, Lillie M, et al. Ancient DNA reveals matrilineal organisation and recurrent unions between dominant matrilines in Iron Age Britain. bioRxiv (2026). Read the preprint
  6. Rodriguez Luis 2021 Rodriguez Luis JA, et al. The Y chromosome of autochthonous Basque populations and the Bronze Age replacement. Scientific Reports 11, 5607 (2021). Read the paper
  7. Garcia-Fernandez 2022 Garcia-Fernandez C, et al. Y-chromosome target enrichment reveals rapid expansion of haplogroup R1b-DF27 in Iberia during the Bronze Age transition. Scientific Reports 12, 20708 (2022). Read the paper
  8. Archaeology 2025 DNA study identifies Iron Age matrilocal society. Archaeology Magazine, January 17, 2025 (quoting Dan Bradley on the Yorkshire matriline). Read the article
  9. Akbari 2026 Akbari A, et al. Ancient DNA reveals pervasive directional selection across West Eurasia. Nature (2026). Read the paper
  10. Note Y-chromosome subclade claims (R-Z2189, R-BY279, R-DF73, R-S3530, R-Y134022) and the I38909 haplogroups are taken from a September 2026 GenArchivist forum discussion between alanarchae2 and miquirumba and have not been independently verified. Global25 coordinates by Davidski; regional ancient averages from the Moriopoulos 2026 collection. All NNLS models, distances and PCA were computed independently for this article.

The coordinates

Global25 scaled coordinates for every population used in the models and figures above, for anyone who wants to reproduce these models in Vahaduo.

Atlantic facade Bronze and Iron Age, Global25 scaled
Turkey_N,0.1179017,0.1800873,0.0034255,-0.1010589,0.0512402,-0.0479692,-0.0037992,-0.006846,0.0361667,0.0806776,0.0082614,0.0113088,-0.0241636,0.0005791,-0.0427122,-0.0103696,0.0225564,0.0013883,0.0136487,-0.0104478,-0.0142612,0.0056932,-0.0049042,-0.0037505,-0.0044357
Russia_Samara_EBA_Yamnaya,0.1258378,0.0892539,0.0429079,0.1154556,-0.0278684,0.0446846,0.0044911,-0.0029487,-0.0548579,-0.0729957,0.0018583,0.0003497,-0.0016518,-0.0236099,0.0372629,0.015734,1e-07,-0.001478,-0.001704,0.0125059,-0.0031197,0.001374,0.0112292,0.0184362,-0.0045237
Luxembourg_Loschbour.DG,0.130897,0.109677,0.203645,0.198,0.162492,0.059125,0.015041,0.038075,0.100217,0.016219,-0.015427,-0.017235,0.019921,-0.001239,0.061346,0.07067,0.002608,0.007348,-0.008925,0.065406,0.117543,0.010387,-0.049422,-0.173639,0.019519
England_BellBeaker,0.1277348,0.1230484,0.064341,0.0710779,0.0259021,0.0226986,0.0034598,0.0032563,-0.0043746,-0.0100229,-0.0033109,0.0038798,-0.0110009,-0.0162548,0.0270989,0.010585,-0.0057948,0.0015414,0.0024651,0.0071702,0.0079027,0.0022669,-0.0002327,0.0074641,3.32e-05
England_MBA,0.1267396,0.1288397,0.0632251,0.0576344,0.0339193,0.0200195,0.001737,0.0040434,0.0042772,-0.0021869,-0.0053306,0.005519,-0.0109557,-0.0143307,0.0234678,0.0077248,-0.009603,0.0016856,0.0001967,0.0027513,0.0052027,0.0027365,-8.57e-05,0.0041545,-0.0006352
England_LBA,0.1264311,0.1321749,0.0617608,0.0475555,0.0364565,0.0172912,0.0011389,0.004065,0.0031466,0.0026074,-0.0034975,0.0062483,-0.0147175,-0.0163558,0.020713,0.0082511,-0.0031594,0.002105,0.0013055,0.0042424,0.0082642,0.0059162,-0.0037828,0.0079437,-0.0006172
England_EIA,0.1271972,0.1345213,0.0616055,0.0452431,0.0399964,0.0159067,0.0027445,0.0044586,0.0078157,0.009717,-0.0028766,0.0054648,-0.0156678,-0.0140276,0.0205568,0.0039398,-0.0065565,0.0007329,0.0003232,0.0014336,0.0070099,0.0029324,-0.002016,0.0023798,0.0015354
England_MIA,0.1295708,0.1351132,0.0598424,0.0468578,0.0397974,0.0161462,0.0021952,0.0048324,0.0075168,0.0059838,-0.0049366,0.0066259,-0.0143694,-0.0150964,0.0211628,0.0056312,-0.0057799,0.0017841,0.0016267,0.003409,0.0060467,0.0038886,-0.0014703,0.0054663,-0.0009608
England_EastYorkshire_MIA_LIA,0.1299209,0.1326715,0.0604605,0.0492806,0.0385785,0.0159665,0.0023668,0.0040382,0.0051569,0.0055778,-0.0044192,0.0067654,-0.0149404,-0.0165934,0.0222339,0.0086751,-0.0037626,0.0015067,0.0011178,0.0025281,0.0065955,0.0043013,-0.0023857,0.0064897,0.0011162
Scotland_MBA,0.1288476,0.128566,0.0598866,0.058786,0.035822,0.0200244,0.002162,0.0046616,0.001759,-0.0013124,-0.004969,0.0061146,-0.0123388,-0.0200652,0.0238326,0.0147172,0.0012516,-0.0010642,0.0013826,0.002101,0.0058148,0.0020526,-0.0016762,0.0037114,-0.0031374
Scotland_LBA,0.1294738,0.131765,0.0641105,0.0520838,0.0338522,0.0192435,0.0052875,0.0036345,0.0015338,-0.0020048,-0.0048718,0.006819,-0.018062,-0.016033,0.0254475,0.0050718,-0.0151242,0.002122,0.002954,0.0009378,0.0058022,0.0055955,-0.0014788,0.007079,0.0046402
Scotland_LIA,0.1269644,0.133773,0.0627049,0.0575527,0.0336845,0.0230465,0.0025636,0.0039859,0.0031978,-0.0005798,-0.0052408,0.0061175,-0.0137442,-0.0162519,0.0264283,0.0032423,-0.0139629,0.0023956,0.0006972,0.0043428,0.0063411,0.0076216,-0.0037199,0.0072408,0.0011866
France_EBA_Brittany_(n=1),0.124067,0.139128,0.059585,0.041021,0.051702,0.01004,0.003995,0.004384,0.020452,0.022051,-0.00065,0.007943,-0.01115,-0.022157,0.012893,0.018828,0.01017,-0.000633,0.003645,0.010755,0.004866,0.003957,-0.008874,-0.005663,-0.000239
France_LBA_Occitanie_(n=2),0.133742,0.1436975,0.053363,0.0143735,0.052779,0.0108765,-0.005875,0.0008075,0.0357915,0.038634,-0.003329,0.008767,-0.0172445,-0.0139,0.0137075,-0.0039775,-0.002673,-0.00114,-0.0001885,0.0072535,0.0082975,0.0026585,-0.000678,-0.0043985,0.008622
France_LBA_Grand_Est_(n=1),0.126344,0.127957,0.05506,0.014212,0.048009,0.014223,-0.00282,0.006461,0.021475,0.046652,-0.014777,0.009741,-0.03107,-0.038259,-0.00285,0.00716,0.007562,-0.007095,0.014832,0.011005,0.003119,0.011376,-0.001602,-0.013737,-0.003712
France_IA1_Hallstatt_Hauts-de-France_(n=10),0.1256606,0.138315,0.0596982,0.0403104,0.0409,0.0136936,0.0013395,0.0064614,0.0098785,0.0105698,-0.0062845,0.0049906,-0.0143012,-0.0089316,0.013029,0.0035931,-0.0033509,0.0009503,0.0013952,0.0036642,0.0031696,0.0025226,-0.0009982,0.0020122,-0.000898
France_IA1_Hallstatt_Bourgogne-Franche-Comte_(n=27),0.12668081,0.14514556,0.052433667,0.021294074,0.046287778,0.0066726667,0.0013925556,0.004392963,0.017354259,0.023386963,-0.0021771852,0.0075487778,-0.015003593,-0.011595926,0.0074545185,0.0019790741,-0.0030374815,0.00040351852,0.003351963,-0.0010237037,0.0037664815,0.0041492593,-0.0051992593,-0.0017896296,0.00021722222
France_Late_Antiquity-Early_Medieval_Saint-Lupien_(Vasconic_Profile)_(n=1),0.117238,0.155376,0.050534,0.017119,0.062165,0.004183,-0.00423,0.005769,0.030679,0.043008,-0.002761,0.008093,-0.02215,-0.008945,0.019815,0.003182,0.002477,-0.009248,-0.01081,-0.006128,0.017594,0.000247,-0.010599,-0.014098,0.007784
Spain_Pais_Vasco_EBA_(n=3),0.12899967,0.15842267,0.057448,0.0037683333,0.066986667,-0.0037186667,-0.0050916667,0.0024613333,0.032246667,0.050175667,-0.0064953333,0.013588,-0.017839333,-0.014450333,0.004569,0.0036686667,0.011343333,0.0033363333,-0.0010056667,0.00033366667,0.010481333,0.005482,-0.005546,-0.023015333,0.007983
Spain_Pais_Vasco_MBA_(n=1),0.120652,0.158423,0.05506,0.000969,0.062781,-0.007251,-0.00141,0.004384,0.041927,0.057951,-0.002761,0.007793,-0.016353,-0.011147,0.001357,-0.006497,-0.01369,0.011655,-0.005405,-0.014507,0.010606,0.001484,-0.012325,-0.0241,-0.002036
Spain_Pais_Vasco_LBA_(n=1),0.120652,0.149283,0.05242,-0.00323,0.063704,0.002231,-0.00564,0.010153,0.026179,0.042643,-0.007957,-0.0003,-0.023488,-0.0139,0.010858,0.023071,0.009257,0.002787,-0.003142,0.009004,0.006738,-0.009892,-0.012941,-0.018195,-0.008502
Spain_Navarre_EIA_(n=2),0.119514,0.1477595,0.0526085,0.0066215,0.063396,0.0008365,-0.00987,0.0036925,0.026588,0.0411855,-0.0068205,0.0116895,-0.0211095,-0.0235335,0.00984,-0.006762,-0.000717,0.000317,0.0051535,-0.005315,0.013039,0.000742,-0.008997,-0.009941,0.0017365
Spain_Pais_Vasco_IA_Celtiberian_(n=3),0.12103167,0.14318967,0.063104667,0.013566,0.057139,0.0035326667,-0.00564,0.0076153333,0.028429,0.034624667,-0.00016233333,0.0078933333,-0.020565,-0.012110667,0.0075546667,-0.0046403333,-0.0074316667,0.0034203333,0.0027233333,-0.00066666667,0.007071,0.0068423333,-0.008052,-0.0047796667,-0.0046303333
Spain_Castile_and_Leon_Early_Medieval_(Vasconic_Profile)_(n=10),0.1213353,0.1468455,0.0586423,0.0154394,0.0537329,-0.0010319,-0.002397,0.0029307,0.0303513,0.0388892,-0.008915,0.0094864,-0.0200096,-0.0158678,0.0155264,0.0017236,-0.0049416,0.0016091,-0.0013699,-0.0061029,0.0095209,-0.0013478,-0.0070375,-0.0066634,0.0001197
Spain_Castile_and_Leon_MBA_(n=3),0.12254933,0.14555933,0.052797,-0.001938,0.067294667,-0.0057636667,-0.003525,0.0023076667,0.040427667,0.051694333,-0.0036266667,0.012489,-0.02557,-0.022753667,0.0090933333,0.016087667,0.020035667,-0.0012246667,-0.005908,0.0018756667,0.014100333,-0.0013186667,-0.0056693333,-0.014098333,0.0057476667
Spain_Asturias_LBA_(n=1),0.119514,0.15436,0.060716,0.007752,0.064935,0.006414,0.000235,-0.001385,0.027815,0.047746,-0.002761,0.008992,-0.031219,-0.03413,0.013572,0.009016,0.015255,0.003167,-0.008673,0.00025,0.010606,0.003462,-0.002711,-0.00723,-0.000479
Spain_Castile_and_Leon_LIA_Cantabri_(n=2),0.1257745,0.155376,0.0631675,0.005814,0.052933,-0.000697,-0.001175,0.0074995,0.032826,0.0429165,-0.0010555,0.008842,-0.0237115,-0.0187855,0.013572,0.0043755,-0.004759,0.0051945,-0.0022625,-0.0090665,0.0131645,0.000371,-0.0077645,-0.0163875,-0.000778
Portugal_MBA_(n=4),0.1186605,0.15385275,0.0553425,-0.00815575,0.06670475,-0.00725125,-0.00658025,0.0047885,0.04335925,0.06013775,0.00259825,0.01210175,-0.018917,-0.013659,0.00790575,0.00235375,0.0003585,0.0007285,0.00182275,-0.00109425,0.00424275,0.004668,-0.01213975,-0.022473,-0.00104775
Spain_Aragon_LBA_Urnfield_(n=14),0.115693,0.14826729,0.056621929,0.0042220714,0.064319571,-0.0045817857,-0.0052203571,0.0035768571,0.035353429,0.050948,-0.0034796429,0.011818071,-0.024380286,-0.014991,0.0076680714,0.0025097857,0.0013783571,-0.00054292857,0.0025230714,-0.0041267857,0.0087435,0.0048844286,-0.0080022143,-0.013960643,0.0017191429
Spain_Catalonia_EBA_(n=10),0.126116,0.1505013,0.0605279,0.0028101,0.0661969,-0.0058846,-0.0006345,-0.0003231,0.0355667,0.055017,-0.0022083,0.0083886,-0.0249453,-0.0191022,0.0014795,0.0115088,0.0118519,0.001989,0.0013702,0.0026387,0.0107309,0.0040683,-0.0119799,-0.0175688,0.0001436
Spain_Catalonia_IA_(n=7),0.12553071,0.14333457,0.056945286,0.0018918571,0.057944714,-0.0014342857,-0.00329,0.0039891429,0.029013,0.044752143,-0.0054748571,0.010469286,-0.023042429,-0.011697857,0.0062625714,0.011573143,0.014193286,-0.0010135714,0.0040584286,0.000929,0.0021034286,0.00097157143,-0.0096661429,-0.010345429,-0.0010264286
Ireland_EBA_(n=2),0.1263435,0.129988,0.0607165,0.0644385,0.0296975,0.022172,0.0019975,0.006,0.002761,-0.0058315,-0.00138,0.0086925,-0.009737,-0.0165835,0.0229365,0.0146515,-0.003716,0.006968,0.0011315,0.0049395,0.006613,0.00643,0.00228,0.0110255,-0.0029335
France_IA2_La_Tene_Occitanie_(South_Euro_Profile)_(low_res)_(n=10),0.1297585,0.146033,0.0583405,0.0162469,0.0509632,0.0050199,-0.0005171,0.0014768,0.0249519,0.0357549,-0.003329,0.0055301,-0.022329,-0.0132944,0.0107762,0.0010871,-0.0018254,0.0025971,0.0017221,-0.0023136,0.0058647,0.002411,-0.0071607,-0.0045669,-8.36e-05
English,0.131855,0.137043,0.061788,0.044013,0.03923,0.016748,0.004985,0.005745,0.005264,0.005678,-0.004769,0.005616,-0.012589,-0.010351,0.020615,0.003552,-0.010369,0.004094,0.003685,0.002962,0.005937,0.003404,-0.003269,0.013794,3.1e-05
Scottish,0.131466,0.134232,0.062265,0.047043,0.039106,0.017281,0.00355,0.005176,0.003572,0.002825,-0.005614,0.005047,-0.01168,-0.011973,0.023024,0.003125,-0.011255,0.002774,0.00255,0.001452,0.004225,0.003308,-0.000656,0.013556,-0.00077
Basque_Spain_(n=25),0.12547856,0.1489578,0.05441092,0.00854012,0.05412692,0.00059132,-0.0006298,0.00081216,0.02897684,0.0417466,-0.00528736,0.01038276,-0.0249632,-0.0186396,0.0152658,0.0019464,-0.00584644,0.00297968,-0.00213696,-0.00316148,0.00836528,0.00164704,-0.00701032,-0.00850236,0.00085736