For a people whose name became a common noun in every European language, the Vandals have left astonishingly little behind. They crossed the frozen Rhine in the last days of 406, fought their way across Gaul and Hispania, and in 429 did something no other barbarian confederation attempted: they took to ships and carried an entire nation, women and children included, across the strait into Roman Africa. Within ten years they held Carthage, the richest city of the western empire after Rome itself. They built the only barbarian navy worth the name, took Sicily, Sardinia and the Balearics, and in 455 sacked Rome at leisure over fourteen days. Their kingdom lasted a little over a century, until a Byzantine expeditionary force under Belisarius dismantled it in a single campaigning season in 533 and 534. And then they simply stop. There is no Vandal successor state, no Vandal language remnant, no Vandal minority anywhere in the modern Maghreb, and, in the entire published ancient DNA record, not one genome anybody has been able to label Vandal. This article asks the obvious question that follows: is the genetic invisibility of the Vandals a gap in the sampling, or is it the actual historical result?
A Kingdom With No Genome
It is worth being blunt about the starting position, because it shapes everything that follows. The merged Global25 dataset used throughout this site contains labelled averages for a remarkable range of migration-period peoples: Visigoths in Spain and Portugal, Ostrogoths in Hungary, Italy and Serbia, Gepids, Lombards in Austria, Czechia, Hungary and Italy, Alemanni and Thuringians in Germany, Avars, Huns. Search it for the Vandals and nothing comes back. Not one sample. This is not an oversight peculiar to this dataset; no published study has yet produced genomes from a securely identified Vandal-period burial in North Africa, and the Vandal century in Africa remains, in aDNA terms, a blank.
That blank forces an indirect approach, but not a weak one. If the Vandals cannot be sampled directly, they can still be triangulated from two ends. At the northern end, the archaeological culture conventionally associated with the Vandals before the migration, the Przeworsk culture of the upper Oder and Vistula basins, is sampled. At the southern end, the population they ruled for a century, and its living descendants, are sampled abundantly. If a nation of the size the sources describe, some tens of thousands of people, settled and reproduced in the African provinces for four generations, that should be visible as a northern European layer in the modern Maghreb. The question is not whether we can find Vandals. It is whether we can find their descendants.
Where the Vandals Came From
The Przeworsk culture ran from roughly the third century BC to the fifth century AD across what is now central and southern Poland, and its association with the Vandals, and with the wider Lugian complex, is the standard reading of the classical geographers alongside the material record. The dataset contains four Przeworsk population entries from Gaski, Boczne, Koziarnia and Zarska. Averaged into a single Przeworsk pole and ranked against everything else, they resolve exactly where the archaeology says they should.
The Przeworsk average sits closest to the Wielbark culture of the Gothic horizon at 0.029, then to modern Germans, Iron Age Denmark, Iron Age Sweden and early medieval Germany, all inside a tight band from 0.029 to 0.040. This is an ordinary northern and central European Iron Age profile, with nothing exotic about it. Two things about that band matter for what follows. The first is that it is a real, distinctive genetic position, well separated from the Mediterranean. The second, and the reason this whole test is possible at all, is the size of the gap at the other end of the chart: the Przeworsk pole sits 0.164 from Punic Carthage, 0.167 from Roman-period Algeria, and a full 0.260 from modern Tunisians. A northern European population and a North African one are about as far apart as any two West Eurasian-shifted groups get. If Vandals had settled in Africa and left descendants, the signal would not be subtle.
One honest caveat belongs here rather than at the end. The four Przeworsk entries are single or paired genomes and they scatter among themselves by 0.043 to 0.056, which is wide. Their average is a directional pole for northern European ancestry of the right period and region, not a precise Vandal ancestral population, and the argument below is built so that it does not depend on the fine detail of where that pole sits. It depends only on the pole being northern European, which every one of the four individuals is.
Searching North Africa
The direct test is a four-source model of the modern Maghreb with the Germanic pole offered as one of the options. The three other sources are the ones the region's documented history actually requires: a local North African substrate, taken from a Roman-period genome at Sitifis in Algeria, an Arabian pole from the Najd interior standing for the conquests from the seventh century onward, and a Sub-Saharan pole for the trans-Saharan gene flow of the medieval and early modern periods. The model is then free to assign as much or as little as it likes to the Vandals.
It assigns nothing. Not a small residual, not a noisy one or two percent: zero, in all ten populations tested, from Tunis and Bizerte on the coast of the old Vandal heartland to the Berber villages of the Tunisian south and the Rif of Morocco. What the model does find is a coherent and entirely expected history: a majority local North African component running from 39 percent in Libya to 76 percent in Sfax, a substantial Arabian layer that rises as one moves east toward Libya exactly as the pattern of the conquests predicts, and a Sub-Saharan component of 9 to 18 percent. Every documented demographic event in the region's post-Roman history shows up in these numbers. The one that is supposed to have happened first does not appear at all.
The Control: Does the Probe Work?
A zero is only as good as the instrument that produced it, and a model that never finds Germanic ancestry anywhere would be worthless. So the identical Germanic pole was run against Iberia and Italy, where a northern European layer is not in dispute.
The probe works, and it works loudly. Visigothic-period Catalonia takes 67 percent from the same pole that North Africa refused entirely; the Ostrogothic and Merovingian cemetery at Bardonecchia in Piedmont takes 44 percent; the Visigothic burials at Miroico in Portugal take 32 percent, and modern Iberians take a third. The contrast could hardly be sharper: an instrument that reads two thirds in sixth-century Catalonia reads a flat zero in twenty-first-century Tunis.
The Iberian percentages themselves need a caveat, and it is the same collinearity problem this site has flagged in its Pashtun, Roma and BMAC coverage. Those figures are not estimates of Visigothic descent. The local Iberian pole used here is a late antique southern Spanish average, and any population sitting further north on Europe's continuous north-south genetic cline will load onto the northern pole regardless of whether a Germanic migration ever reached it. Modern Catalans would score high on this model if the Visigoths had never existed. The chart is doing one job only, and doing it well: demonstrating that the instrument is capable of reading a northern European signal at high sensitivity when one is present. That is what makes the North African zero interpretable.
How Small Could a Vandal Layer Be and Still Hide?
Zero in an NNLS model does not mean zero in reality; it means below the model's floor. That floor can be measured directly rather than guessed at, by taking the modern Tunisian average, mixing a known quantity of the Germanic pole into it artificially, and asking the model to find what was just put there.
The result is a genuinely useful number and it cuts against an over-confident reading of the previous two charts. Injected shares of 1, 2, 3, 5 and even 8 percent come back as exactly zero: the model cannot see them at all. Only at 10 percent does anything register, and even at 15 percent the model recovers just 6 of the 15 points injected. The honest conclusion is therefore narrower than a headline would like. This analysis rules out a Vandal contribution to the modern Maghreb above roughly 10 percent. It says nothing whatever about a contribution of 2 or 3 percent, which is precisely the range a militarily dominant but demographically thin ruling caste, absorbed over fifteen centuries, would be expected to leave. The Vandals are not proven absent. They are proven small.
That is not a disappointing result, because 10 percent is a demanding ceiling in context. The sources credit Geiseric with transporting something on the order of eighty thousand people, in a Roman Africa whose population ran into the millions. Even at full reproductive success and zero attrition, the Vandals were always going to be a minority; the Byzantine reconquest, the documented deportation of Vandal fighting men to the eastern frontier, and the collapse of the Arian church that had been the community's institutional spine then removed the mechanisms by which a small minority stays visible. What the genetics can say is that nothing survived of them at the scale a founding population normally leaves.
Vandalusia: The One Place They Left Their Name
There is a second, more famous claim to test, and it concerns the twenty-two years the Vandals spent in Hispania rather than the century they spent in Africa. Since the thirteenth-century De rebus Hispaniae, and Ibn Khaldun after it, the name al-Andalus has been derived from the Vandals, giving the modern Andalusia a folk etymology as Vandalusia, the land of the Vandals. That derivation has been under sustained attack since the 1980s, with Vallve proposing a link to Atlantis, Halm reconstructing an unattested Gothic landahlauts meaning the allotted land, and Bossong arguing for a pre-Roman substrate; philologists remain genuinely divided, and none of the alternatives is decisive either. What the genetics can do is test the demographic version of the claim, which is a different and much more tractable question: whether southern Spain carries any excess northern European ancestry relative to the rest of Iberia.
It carries the opposite. Under the identical model, Andalusia comes out at 34 percent northern European, at the bottom of mainland Spain, below Extremadura and Castile, well below Galicia and Cantabria, and 15 points below northern Aragon and Catalonia. The gradient across the peninsula runs cleanly from the Pyrenean north down to the Mediterranean south, which is the ordinary European cline plus the deeper Mediterranean and North African ancestry of the south, and the Vandal name is attached to the very corner of Iberia with the least northern ancestry in it. Whatever al-Andalus means, and it may well mean the land of the Vandals, the toponym records a political episode of two decades, not a population. This is a useful reminder in its own right: names travel far more easily than genes, and a great deal of ancestry folklore consists of reading the first as evidence of the second.
The Uniparental Check: Y-Chromosomes and mtDNA
An autosomal model with a floor of 10 percent is not the last word, and the objection to everything above writes itself: autosomal ancestry from a small founding group gets diluted by half every generation, but a Y-chromosome or a mitochondrial lineage does not blend. It passes down intact, and a single male line can persist at low frequency for fifteen centuries while contributing almost nothing measurable to the genome as a whole. If any Vandal trace survives in North Africa, the uniparental markers are where it should be looked for. This is also where the popular version of the claim actually lives, in the long-running folklore about blond and blue-eyed Berbers in the Rif, the Kabyle mountains and the Jebala, routinely attributed to Vandal ancestors.
European-derived uniparental lineages in North Africa are real and reasonably well characterised. On the maternal side the region carries substantial West European mitochondrial input, above all haplogroup H1 together with H3 and V, reaching around 21 percent in some Tunisian Berber groups and an extreme 61 percent among the Libyan Tuareg of the Fezzan. On the paternal side, alongside the dominant autochthonous E-M81 that tracks the Berber-speaking area almost exactly, Algerian and wider Maghrebi samples carry a genuine European-derived R1b presence in the branches R-M412, R-S116, R-U152 and R-M529. The raw observation that Europe left paternal and maternal marks in North Africa is not in dispute at all. The question is when they arrived, and the dating is what dismantles the Vandal reading.
Ottoni and colleagues sequenced complete mitochondrial genomes from Libyan H1 carriers and dated the arrival of European H1 in North Africa to roughly 8,000 to 9,000 years ago, an early Holocene crossing from Iberia after the Younger Dryas, followed by the formation in situ of North African specific branches labelled H1v, H1w and H1x. That is some seven millennia before Geiseric was born. The European R1b branches found in the Maghreb are equally unhelpful to the Vandal case, and for a more precise reason: R-S116 and its relatives are the Western European and specifically Iberian side of R1b, the DF27-rich lineage that dominates the Iberian Peninsula and expanded there around 4,200 years ago at the Neolithic to Bronze Age transition. The researchers who documented them in Algeria read them, alongside the H1, H3 and V mitochondrial lineages, as the counterpart of direct maritime contact across the western Mediterranean, which is exactly what the geography of the Strait of Gibraltar would predict.
The specifically Germanic paternal markers tell the opposite story by their absence. The lineages that actually characterise the northern European populations the Vandals came from are haplogroup I1, the classic Nordic and Germanic line, and R1b-U106, its western Germanic counterpart. Neither registers as a meaningful presence in North African samples. This matters because it converts a vague argument into a falsifiable one, and it exposes a common error in the amateur literature, where the reasoning runs that finding haplogroup R in a Berber man demonstrates Germanic origin. It does not. R1b is not a Germanic marker; it is a Western European one with several deeply divergent branches, and the branches actually present in the Maghreb are the Iberian ones, which arrived by sea and by Neolithic and Bronze Age expansion rather than by longship in the fifth century. Distinguishing R-S116 from R-U106 is the entire question, and the distinction points away from the Vandals.
The pigmentation argument dissolves along with it. Light hair and eyes in the Rif or Kabylia need no Germanic explanation once the European-derived component in North African genomes is understood to be eight or nine thousand years old rather than fifteen hundred. A population that has carried Iberian maternal lineages since the early Holocene has had ample time to carry the associated pigmentation alleles too, and the trait clusters in mountain regions that are known refuges with strong drift and founder effects, which is precisely where unusual frequencies of any variant are expected to turn up.
A Calibration: The Morisco Expulsion
There is one further test available, and it is the most useful of all, because it takes the same question and puts a documented answer behind it. In 1609 the Spanish crown expelled the Moriscos, the Muslim converts of Iberia, and tens of thousands of them settled in Morocco and Tunisia, founding identifiable communities at Tetouan and Chefchaouen and in the Tunisian villages of Testour, Slouguia, Qalaat-el-Andalous and El Alia. Here is a European migration into North Africa that is dated, named, documented in the archives and traceable in the family traditions of living people. If the methods used in this article can see that, they are working; if they cannot, the Vandal null result is worthless.
They see it without difficulty. Substituting a modern Iberian pole for the Germanic one, the Morisco community of Tetouan comes out at 48 percent Iberian, the Jebala of Chefchaouen at 22 percent, the Riffians at 13 percent, and general Tunisians at 6 percent, against the flat zero the Germanic pole returned in every one of those same populations. The gradient falls away exactly as distance from the Morisco settlement zone increases, which is what a real and geographically anchored migration looks like in this kind of model.
The comparison is worth holding still for a moment, because it is the sharpest thing in this article. A refugee movement of civilians who arrived four centuries ago, with no army and no state behind them, is one of the most conspicuous signals in the modern Moroccan genome. A conquering nation that ruled the same region for a hundred years, held its capital, commanded its fleet and sacked Rome leaves nothing that any method used here can detect. The Morisco figure does carry a caveat, since an Iberian pole and a North African pole are partly collinear and modern Iberians themselves carry some North African ancestry, so 48 percent should be read as a strong directional signal rather than a precise descent estimate. The contrast with zero is not sensitive to that caveat at all.
Why the Vandals Vanished and the Visigoths Did Not
The comparison that makes the Vandal case legible is with their closest analogue. The Visigoths and the Vandals were similar peoples doing similar things at the same time: East Germanic, Arian Christian, mobile, and installed as a warrior elite over a Roman provincial population. One is visible in the genetic record of the country it conquered and the other is not, and the difference is not mysterious.
The Visigothic kingdom lasted three centuries in Iberia rather than one in Africa, and it converted from Arianism to Catholicism under Reccared in 589, dissolving the confessional barrier that had kept Goth and Roman as separate marriage pools. The Vandal kingdom never made that move; Arianism in Vandal Africa was an instrument of rule and a mark of separation from the Catholic provincials, enforced through periodic persecutions that guaranteed the two communities stayed distinct. And where the Visigothic kingdom fell to a conquest that absorbed it, the Vandal kingdom fell to one that deliberately dismantled it. Procopius records the surviving Vandal fighting men being shipped east and enrolled in units for the Persian frontier, the settlement lands being redistributed, and the Arian church being suppressed outright. A small endogamous elite that is stripped of its land, its church and its men in the space of a few years does not get a long demographic afterlife.
There is also a point about the northern end of the journey that deserves stating. The Przeworsk chart at the top of this article shows a population thoroughly at home in northern and central Europe, and modern Poles, Germans and Czechs sit close to it. The Vandal genetic legacy, such as it is, is very likely to be sitting in central Europe among the relatives who never left, rather than in North Africa among the descendants of the people they ruled. The migrating fraction is the part of a population that historians write about and the part that genetics is least likely to find.
Limits and Caveats
Several caveats apply and are worth stating plainly. The most important is the detection floor established above: this analysis excludes a Vandal contribution to the modern Maghreb greater than about 10 percent and cannot address anything smaller, so no claim of a genuine zero is being made anywhere in this article. The Przeworsk pole is an archaeological association standing in for the Vandals, not a Vandal genome, and the four entries averaged into it are thin and internally scattered; a different northern European pole of the same period would change the exact percentages but not the qualitative result, since every plausible candidate sits in the same 0.25-plus band away from North Africa. The Roman-period Sitifis genome used as the local North African substrate is a single individual, and the Punic Carthage and Numido-Roman comparanda are equally small, so the local pole is directional. The Iberian percentages in the control chart are explicitly not estimates of Visigothic ancestry, for the collinearity reason set out in that section, and should not be quoted as such. The uniparental section rests on published haplogroup literature rather than on any analysis run here, and frequency data for I1 and R1b-U106 in North Africa is thin enough that their near-absence is better described as unreported than as formally quantified. The Morisco calibration uses an Iberian pole that is partly collinear with the North African one, so its percentages are directional. Finally, and most obviously, an argument built on the absence of Vandal-labelled genomes would be overturned tomorrow by the publication of a Vandal-period cemetery from Carthage or Hippo Regius, which is exactly the sort of thing this site would rather see than continue to reason around.
Conclusion
The Vandals ran a Mediterranean naval power for a century, sacked Rome, and gave a European language its word for wanton destruction, and they are the only major migration-period people whose genetic trace this dataset cannot find at all. The archaeological culture they came from resolves as an unremarkable northern and central European Iron Age population, far enough from North Africa that a real Vandal legacy would be conspicuous. It is not there: a four-source model of ten Maghrebi populations assigns them nothing, while the identical probe reads 67 percent in Visigothic Catalonia and 44 percent in Ostrogothic Piedmont, and Andalusia, the province that may still carry their name, has less northern European ancestry than anywhere else in mainland Spain. The uniparental record points the same way: the European maternal and paternal lineages that North Africa really does carry are Iberian in branch and early Holocene or Bronze Age in date, some seven millennia older than the Vandal century, while the specifically Germanic markers are not there. And the calibration is unanswerable, since a civilian refugee movement of 1609 shows up at 48 percent in Tetouan while a conquering kingdom of a hundred years shows up at nothing. The one qualification that matters is that the method's floor is around 10 percent, so a thin surviving trace of a few percent remains entirely possible and would be invisible here. Within that limit the answer is clear enough. The Vandals were a ruling caste that stayed separate by religion and by policy, was militarily dissolved rather than absorbed, and left a kingdom, a reputation and possibly a place name, but no people. Of all the barbarian nations that carved up the western empire, they are the one that genuinely disappeared.
Poland_IA_Przeworsk_Gaski_(n=1),0.127482,0.128972,0.060716,0.057494,0.036622,0.015339,0.007285,-0.003923,0.000614,0.001276,-0.004872,-0.001349,-0.015758,-0.009221,0.020358,0.005701,-0.00339,0.001647,0.003394,0.008629,0.003119,0.011252,0.012325,0.013134,-0.002754
Poland_LIA_Przeworsk_Boczne_(n=1),0.141141,0.122879,0.063356,0.050065,0.029544,0.022311,0.009635,0.00923,-0.00634,-0.003462,-0.010718,0.010041,-0.018285,-0.007294,0.002443,0.023071,0.010822,0.006081,0.010056,0.0,0.003119,0.013107,-0.004314,0.007109,-0.007305
Poland_LIA_Przeworsk_Koziarnia_(n=1),0.14797,0.129988,0.064865,0.037791,0.052625,0.014502,0.012456,0.014769,-0.014726,0.008018,-0.000162,0.009591,-0.000446,-0.000275,0.0019,0.001856,0.009518,0.002787,0.002011,-0.007629,0.003369,0.009398,0.004807,0.006748,0.00479
Poland_LIA_Przeworsk_Zarska_(n=2),0.1411405,0.1462365,0.062602,0.0576555,0.0450855,0.02775,0.010223,0.005769,-0.0053175,-0.001276,0.002842,0.0005995,-0.0118185,-0.0176155,0.0128935,0.009414,0.0182535,-0.004751,0.003645,0.0136315,0.0011855,0.003895,0.0008625,0.0206655,-0.005987
Poland_LIA_Wielbark_(Maslomecz)_Grodek_(n=6),0.13336283,0.13337283,0.0658705,0.053402667,0.0403665,0.019615167,0.004426,0.009115,0.0036471667,-0.0085346667,-0.0046821667,0.0012488333,-0.004881,-0.0014221667,0.0173495,0.005193,-0.0058456667,0.0012878333,0.0017178333,0.0027305,0.0047208333,0.004101,-0.0023621667,0.0066073333,0.0021355
Denmark_IA_Zealand_(n=145),0.12908322,0.1322849,0.070815441,0.062956041,0.043123062,0.021970807,0.0050501448,0.008986931,0.0035685793,-0.0057121241,-0.0057183241,0.0034097034,-0.0087146207,-0.0078160276,0.023392552,0.0087729034,-0.0065299931,0.0033926276,0.0056581103,0.0080055379,0.0089557793,0.0045333379,-0.00048448966,0.015007455,-0.00019408276
Germany_Early_Medieval_Drantum_(n=16),0.11062181,0.12973387,0.058830812,0.051074375,0.040238188,0.019626938,-7.35e-05,0.0081343125,0.0040010625,-0.0028018125,-0.005379125,0.0060976875,-0.0098023125,-0.01143125,0.023530563,0.008021625,-0.0054518125,-0.0001266875,0.004894375,0.00362675,0.0086331875,0.005664875,0.0016483125,0.015212937,0.001175
Spain_Catalonia_Early_Medieval_Visigoth_(n=6),0.12729233,0.13557333,0.048397167,0.020672,0.0400585,0.0050201667,0.0046608333,0.001923,0.0095103333,0.016613667,-0.0068473333,0.0043983333,-0.0129085,-0.00578,0.0053835,0.0051268333,0.0057803333,0.0034418333,-0.0012568333,0.0038976667,0.0069668333,0.0033798333,-0.0022595,0.0033338333,-0.0006585
Portugal_Late_Antiquity-Early_Medieval_Visigoth_Miroico_(n=4),0.1021565,0.1416665,0.03997475,-0.00524875,0.0427005,-0.004741,-0.00475875,0.0013845,0.02873575,0.0315725,-0.00633325,0.00846725,-0.00944,-0.0145535,0.00793975,-0.004077,-0.00430275,-0.00126675,-0.00857875,-0.0012505,0.00418025,0.00061825,-0.000647,-0.004609,-0.0003295
Spain_Andalusia_Late_Antiquity_(n=3),0.079296667,0.14589767,0.016970667,-0.033269,0.037750333,-0.019336333,-0.00188,0.0075383333,0.029997,0.036568667,-0.004655,0.0036466667,-0.006244,-0.015230333,0.006424,0.0057013333,0.0072146667,-0.005321,-0.011229,-0.0012086667,0.001414,-0.0038333333,0.0067786667,-0.001165,-0.0035523333
Italy_Piedmont_Early_Medieval_Ostrogoth-Merovingian_Bardonecchia_(West_Med/North_Adriatic_Profile)_(n=8),0.12065238,0.14737863,0.02734125,-0.007146375,0.034506375,0.00094125,-0.000910625,-0.00054825,0.00961275,0.02460175,-0.003694375,0.00524525,-0.015869625,-0.007552,0.00274825,0.00648025,0.008295625,-0.00038,0.001901,-0.002438625,0.002230375,0.0049305,-0.00101675,0.000768125,0.001916
Tunisia_IA_Punic_Carthage_(Punic_Profile)_(n=2),0.0614645,0.1436975,-0.0092395,-0.0473195,0.0236965,-0.0214745,-0.010105,0.002077,0.0247475,0.0303425,0.0024355,-0.0002995,0.003196,-0.015345,-0.0040715,-0.005171,0.0020865,0.003167,-0.012067,-0.005065,0.0005615,-0.0097685,0.006902,0.004097,-0.001916
Algeria_Roman_Empire_Sitifis_(North_African-Mixed_Profile)_(n=1),0.078538,0.152329,-0.019987,-0.056525,0.013233,-0.022032,0.007285,-0.001154,0.020861,0.030251,-0.002923,0.002997,-0.008474,0.000138,0.001764,-0.006762,0.005476,-0.00228,-0.001257,0.004377,-0.006738,-0.0115,0.012818,0.001928,0.00491
Algeria_NumidoRoman_Berber.SG,0.0916275,0.150298,-0.024324,-0.0521645,0.008155,-0.021335,0.0034075,-0.0003465,0.006954,0.021048,0.0022735,0.003822,-0.0039395,0.0001375,-0.002375,-0.0028505,0.005411,-0.002154,-6.3e-05,-0.005565,-0.005116,-0.005626,0.00228,0.001145,0.0044305
Tunisian_Arab_Tunis_(n=11),-0.044786273,0.13229755,-0.013569364,-0.072845182,0.019174727,-0.029378545,-0.021972364,0.0054736364,0.052262727,0.020673364,0.0052499091,-0.0072060909,0.024197636,-0.010161636,0.012055091,-0.0078126364,-0.0030554545,-0.012778909,-0.028812455,0.0065407273,-0.010167182,-0.022534091,0.017274818,-0.0037531818,0.0015925455
Tunisian_Arab_(n=8),-0.037392375,0.1355325,-0.014760125,-0.071326875,0.0164375,-0.031420625,-0.022569625,0.00505275,0.05130675,0.02235725,0.0059335,-0.006414625,0.021604375,-0.012366375,0.012531,-0.0093685,-0.001454625,-0.016936125,-0.029655875,0.008586875,-0.0103425,-0.025935625,0.015473125,-0.003195875,0.001535125
Tunisian_Arab_Sfax_(n=6),0.013469167,0.1391275,-0.011816333,-0.061693,0.017079667,-0.0229155,-0.012416667,0.002577,0.040700333,0.025391333,0.0033831667,-0.002248,0.0091428333,-0.009175,0.004275,-0.0027403333,0.0070625,-0.0074748333,-0.021976167,0.0051273333,-0.0077986667,-0.017228833,0.0076413333,-0.0030121667,0.0057078333
Tunisian_Arab_Sousse_(n=2),-0.0022765,0.137604,-0.0092395,-0.062016,0.0235425,-0.0237055,-0.014688,0.0059995,0.0416205,0.022597,0.0031665,-0.004646,0.0138255,-0.0052295,0.0084825,-0.0072925,-0.004824,-0.015646,-0.0243225,0.0051895,-0.0101695,-0.019475,0.0167,-0.006507,0.004251
Tunisian_Arab_Bizerte_(n=2),-0.016504,0.1320185,-0.0135765,-0.066861,0.0136945,-0.033606,-0.016216,-0.001846,0.0440745,0.0220505,0.0113675,-0.0094415,0.0202175,-0.0101155,0.008822,0.0021875,-0.0014345,-0.006968,-0.024134,0.0076285,-0.0039305,-0.0174965,0.009305,0.0028315,-0.003892
Algerian_Arab_(n=32),-0.042356352,0.13539464,-0.0087983729,-0.068888725,0.022477383,-0.029508074,-0.022103493,0.006953225,0.055356433,0.023909376,0.0061268833,-0.006492849,0.021554307,-0.01277548,0.013191184,-0.010463791,-0.0023207172,-0.01499231,-0.03196306,0.0074879984,-0.009903688,-0.028748807,0.017521279,-0.0037945203,0.0040842594
Algerian_Arab_Algiers_(n=24),-0.055820708,0.13413458,-0.0073537917,-0.068301042,0.022401458,-0.030782542,-0.022649125,0.006451625,0.054130583,0.025961,0.00546025,-0.0059634583,0.021456625,-0.013773833,0.013555042,-0.011005,-0.0043135417,-0.016955125,-0.032571583,0.0079465,-0.010777875,-0.028043458,0.018908167,-0.0010894583,0.0019259167
Libyan_Arab_(n=21),-0.030189113,0.13412185,-0.023095635,-0.083920434,0.011595913,-0.035469485,-0.021773945,0.0019832286,0.052886679,0.013887487,0.0069750624,-0.012951735,0.029802945,-0.0071135395,0.0088001629,0.00083614048,-0.0013740329,-0.010478087,-0.021803691,0.011699529,-0.0027208738,-0.014831933,0.010566527,-0.0007129581,0.0012382529
Berber_Tunisia_East_Zenati_Chenini_(n=18),-0.027949944,0.13907106,-0.0080870556,-0.076479222,0.027731556,-0.035202278,-0.031386722,0.0052817778,0.068424611,0.029795667,0.0040057222,-0.0043877222,0.019631389,-0.016124833,0.014092278,-0.016905167,0.00015211111,-0.023289556,-0.046724722,0.0078579444,-0.016831444,-0.0404275,0.028134722,-0.0044516667,0.0063666111
Berber_Morocco_Riffian_(n=29),-0.021744172,0.13919762,-0.0025488276,-0.068052759,0.030010759,-0.03184169,-0.022633931,0.0045913448,0.061914379,0.029905586,0.0050228621,-0.006222069,0.017193345,-0.014350793,0.015153828,-0.01283831,-0.0014072414,-0.018011655,-0.038623966,0.0083273103,-0.013188,-0.035215517,0.022061345,-0.0054305862,0.0048188276
Saudi_Arab_Najd_(n=16),0.057950625,0.13694631,-0.062342563,-0.10618569,-0.013198,-0.043737625,-0.013241562,-0.0087779375,0.047378562,-0.0065645625,0.013764313,-0.02849125,0.058584875,0.0037919375,0.0043161875,0.023631437,-0.024332375,0.00591,-0.0010865,0.024357313,0.0091581875,0.014434813,-0.005340625,0.0064899375,-0.006394
Esan_(n=99),-0.63108589,0.064070616,0.021332101,0.016636131,-9.6323232e-05,0.013336101,-0.041953051,0.045331455,-0.050290131,0.034427909,0.0064348384,0.00040416162,0.02316404,0.0021810909,0.010835657,-0.008499101,0.006349303,0.0013961212,0.0053249899,-0.0015095556,0.0011532323,0.0026616667,-0.001516303,7.1757576e-05,-0.00036048485
Spanish_Andalusia_(Andalusian)_(n=33),0.10795939,0.14555918,0.036386424,-0.0041598485,0.044129303,-0.0056200303,-0.0024568182,0.002517303,0.024505606,0.032012788,-0.0020470909,0.0063306364,-0.012401788,-0.01148103,0.0096484545,-0.0013219394,-0.0057527576,-0.0013168182,-0.0040375152,-0.0014779091,0.0026128182,-0.00067075758,-0.0018262727,-0.0039141818,0.0012046667
Spanish_Catalonia_(Catalan)_(n=9),0.120659,0.14658556,0.042502222,0.0044887778,0.043712556,0.00034311111,-0.00074622222,0.0030055556,0.020485667,0.030803222,-0.0023845556,0.0087437778,-0.016527444,-0.014499667,0.0085938889,0.0073778889,0.0041776667,-0.00073922222,-0.00069066667,0.0016701111,0.0057662222,-0.0037213333,-0.0010778889,0.00079277778,0.00098555556
Spanish_Aragon_North_(Aragonese)_(n=10),0.1273681,0.1477595,0.0490634,0.0057171,0.0498553,0.0010876,-0.000141,-0.0005538,0.0255859,0.0364654,-0.004417,0.0096965,-0.0186272,-0.0150697,0.0143594,0.004296,-0.003038,0.0026858,-0.0031172,-0.0011756,0.004517,0.0005194,-0.0056325,-0.0052896,0.0009341
Spanish_Murcia_(Murcian)_(n=8),0.10688675,0.14428975,0.0364405,-0.00739325,0.042939125,-0.004932,-0.004181875,0.000382875,0.0246175,0.03224675,-0.002883375,0.005347375,-0.013005,-0.009649625,0.0061355,-0.00352575,-0.00333675,-0.000663625,-0.002077875,-0.001193375,0.00372425,-0.003777,0.000667625,-0.004051625,0.001436625
Portuguese_(n=54),0.10437014,0.1448952,0.037173499,-0.0034089926,0.04220484,-0.0033778874,-0.003401937,0.0029970748,0.025839181,0.029040641,-0.0023557796,0.0062321719,-0.012222189,-0.011196649,0.010799551,-0.0015183393,-0.0030798074,-0.00096792444,-0.0043671878,-0.00083191259,0.001744523,-0.0023202933,0.0012529078,-0.0018361115,-0.00036340611
- Procopius History of the Wars, Books III and IV, the Vandalic War, on the Byzantine reconquest of 533 to 534 and the deportation of Vandal soldiers to the eastern frontier.
- Merrills and Miles The Vandals, Wiley-Blackwell, 2010, on the Hasding migration, the African kingdom, Arian church policy and the fate of the Vandal population after 534.
- Juras et al. Ancient DNA Reveals Matrilineal Continuity in Present-Day Poland over the Last Two Millennia, PLOS ONE, 2014, on Przeworsk and Wielbark Roman Iron Age populations. journals.plos.org/plosone, pone.0110839
- Ringbauer et al. Punic and Roman Mediterranean ancient genomes, on the Carthaginian and North African populations used here as local comparanda.
- Wikipedia contributors Etymology of Andalusia, on the Vandal derivation and the Vallve, Halm and Bossong alternatives. en.wikipedia.org/wiki/Etymology_of_Andalusia
- Wikipedia contributors Przeworsk culture, on the archaeological association with the Vandals and the Lugian complex. en.wikipedia.org/wiki/Przeworsk_culture
- Ottoni et al. Mitochondrial Haplogroup H1 in North Africa: An Early Holocene Arrival from Iberia, PLOS ONE, 2010, dating European H1 in North Africa to roughly 8,000 to 9,000 years ago. journals.plos.org/plosone, pone.0013378
- Bekada et al. Introducing the Algerian Mitochondrial DNA and Y-Chromosome Profiles into the North African Landscape, PLOS ONE, 2013, on R-M412, R-S116, R-U152 and R-M529 as evidence of western Mediterranean maritime contact. journals.plos.org/plosone, pone.0056775
- Sole-Morata et al. Analysis of the R1b-DF27 haplogroup shows that a large fraction of Iberian Y-chromosome lineages originated recently in situ, Scientific Reports, 2017, dating the Iberian R1b branch to about 4,200 years ago. ncbi.nlm.nih.gov, PMC5544771
- Arredi et al. A Predominantly Neolithic Origin for Y-Chromosomal DNA Variation in North Africa, American Journal of Human Genetics, 2004, on E-M81 as the dominant autochthonous Berber paternal lineage. sciencedirect.com, S0002929707624173
- Fadhlaoui-Zid et al. Insights into the Middle Eastern paternal genetic pool in Tunisia, on the Tunisian Andalusian villages founded after the Morisco expulsion. ncbi.nlm.nih.gov, PMC8333252
- Davidski Global25 coordinates dataset.
- Vahaduo G25 analysis tool used for NNLS modelling.
- Moriopoulos 2026 collection Aggregated Global25 population averages, including the Przeworsk and Wielbark entries, the Visigothic, Ostrogothic and Lombard comparanda, the Punic and Roman North African genomes, and the modern Maghrebi and Iberian populations used in this article.