Project a Réunionnais genome into the 25-dimensional Global25 space and the result defies any single continental assignment: a coordinate pulled simultaneously toward the Bantu corridor of southeastern Africa, the Dravidian coast of southern India, the forests of Normandy, and the Austronesian maritime world of Madagascar and the Indonesian archipelago. This is not genetic noise. It is the compressed record of an island that had no human population until 1642, then became one of history's most thoroughly documented mixing experiments — French colonial settlers, enslaved Africans from Mozambique and the Congo Basin, Malagasy captives carrying their own Austronesian legacy from Borneo, Tamil and Malabar indentured workers shipped across the Indian Ocean after abolition, and smaller contributions from Gujarat, Fujian, and the Comorian archipelago. La Réunion has no pre-colonial genetic baseline. Every component of its modern population arrived at a documented date, from a documented place, under a documented legal regime — making it one of the rare territories where colonial economic history can be read with unusual directness from DNA.
Key Findings at a Glance
- Réunionnais Creoles typically carry approximately 30–35% Bantu East/Southeast African ancestry (principally the Mozambique corridor), ~25–30% European (predominantly northwestern French), ~20–25% South Asian (predominantly Tamil/Malabar Dravidian), and ~8–12% Malagasy/Austronesian — a direct reflection of four successive waves of forced and semi-coerced migration.
- The African component is predominantly East African, not West African: Mozambican and Tanzanian Bantu ancestry dominates over the Guinea Coast signal, distinguishing Indian Ocean Creole populations from Atlantic (Caribbean/Brazilian) ones, where West African ancestry prevails. This geographic specificity is clearly legible in G25 PCA space.
- The Malagasy contribution is substantial and genetically dual: Malagasy ancestry carries ~30–40% Austronesian (Borneo-derived) ancestry alongside ~60–70% Bantu East African. A Réunionnais individual with Malagasy ancestry will show both "East African" and "Southeast Asian/Indonesian" signals from a single ancestral source — explaining apparent Southeast Asian components in calculator outputs.
- The South Asian component is overwhelmingly Dravidian/Tamil: the "engagisme" indentured system (1848–1882) brought over 45,000 workers from the Madras Presidency — predominantly from Tamil Nadu and the Malabar coast (modern Kerala). "North_Indian" signals in G25 results partly reflect reference-panel resolution, not genuine Gangetic plain ancestry.
- The maternal haplogroup M32c documented in a Réunionnais individual is a Dravidian South Indian marker, confirming Tamil/Malabar female lineage from the indentured period. The paternal haplogroup T-L446 points to South Asian or Middle Eastern paternal ancestry common in Indian Ocean trading populations.
- Commercial ancestry tests give broadly consistent results: approximately 33% European / 33% Sub-Saharan African / 17% South Asian / 10% East Asian and Southeast Asian for a broadly-admixed Réunionnais Creole individual — proportions that map closely onto the island's documented demographic history.
I. The Uninhabited Island — Before 1642
Réunion's genetic history begins with an absence. Unlike virtually every other inhabited island on Earth, La Réunion has no pre-colonial human genetic layer — no indigenous population whose ancestry could serve as a founding substrate for subsequent admixture. Arab sailors almost certainly knew of the island by the 9th century CE; Portuguese navigators charted it in the early 16th century under the name Santa Apolónia. But neither Arab nor Portuguese colonists established permanent settlements. When French forces took formal possession on behalf of the Compagnie des Indes Orientales in 1642, they found an island of extraordinary volcanic geology, unique biodiversity, and no human inhabitants whatsoever.
This absence is genetically decisive. In a European island like Sicily, each new wave of migration arrived on top of a population already shaped by 14,000 years of predecessors — Neolithic farmers modifying hunter-gatherer substrate, Bronze Age steppe migrations modifying Neolithic substrate, and so on. In Réunion, every genetic component can in principle be traced to a specific historical transaction: the purchase of enslaved people on the Mozambique coast, the recruitment of indentured workers in Madurai or Pondicherry, the settlement of a Norman family from the Cotentin peninsula. No pre-colonial filter distorts the record. The population genetics of Réunion is, more than almost anywhere else, documentary history read from chromosomes.
II. The Slave Trade — Africa Comes to Réunion
The dominant ancestry component in most Réunionnais Creoles is Bantu African, and its geographic character distinguishes Indian Ocean Creole genetics from Atlantic ones. The Caribbean and Brazil received most of their enslaved population from the Guinea Coast — western Africa, from Senegal to Angola — which is why West African ancestry (Yoruba, Ewe, Fon, Mandinka) dominates Caribbean African-descended populations. Réunion's slave trade operated on a different axis. The island's primary supplier was the Mozambique coast and the broader southeastern African corridor — the territory of Makua, Yao, and Swahili-speaking Bantu populations — because Portuguese East Africa lay directly in the path of French Indian Ocean trade routes, and because the Compagnie des Indes Orientales had established commercial relationships with Mozambican slave traders before Guinea routes became competitive for Indian Ocean-bound ships.
This geographic specificity is still legible in modern genomes. When G25 calculators analyse a Réunionnais individual, the dominant African signal resolves toward "Southeast_African" (Mozambique corridor) and "West_Kenyan" populations rather than toward "Southwest_Nigerian" or "Ghanaian" proxies — which dominate for Caribbean populations. The West African component is present — reflecting the "Cafres de Guinée" trade route from Benin and Nigeria — but it is secondary, typically representing 7–10% of total ancestry versus 30–35% for the East African component.
Madagascar: Where Bantu Africa Meets Austronesian Southeast Asia
Madagascar was the second major source of enslaved people for Réunion, particularly in the early colonial period (1650–1750). But Malagasy genetics are themselves deeply admixed: approximately 60–70% Bantu East African and 30–40% Austronesian, reflecting the remarkable historical migration of Borneo-origin Austronesian speakers who settled Madagascar around 500 CE and mixed with incoming Bantu farmers. When a Réunionnais individual's G25 results show components labeled "Malagasy," "North_Indonesian," "Lao," or "Vietnamese" — as visible in the World Regions K248 calculator — these often represent a single historical event: an enslaved Malagasy ancestor. The Austronesian fraction of that ancestor's genome (5.2% Malagasy + 7.2% North_Indonesian + 1.6% Lao + 1.4% Vietnamese in the sample analysed) is the preserved signal of Borneo-origin migrations that reached Madagascar over a millennium before those descendants were enslaved and transported to Bourbon Island.
African Origins of Réunion's Enslaved Population — Historical Trade Routes
Primary — "Cafres de Mozambique" + Madagascar
Early dominant — Bantu + Austronesian + Comoros Islands
"Cafres des Comores" + Guinea / Benin / Nigeria
Secondary — "Cafres de Guinée" + Congo Basin
Minor — Kwango/Kongo peoples → Bourbon Island
~60,000 enslaved by 1810
III. Indian Indentured Labor — The Tamil and Malabar Contribution
The abolition of slavery in 1848 triggered an immediate economic crisis: the plantation economy depended on coerced labor, and the newly freed "nouveaux libres" largely abandoned the cane fields. The colonial administration's solution was "engagisme" — multi-year indentured labor contracts with workers from British India. Between 1848 and 1882, when the British government restricted emigration from India to non-British territories, over 45,000 Indian workers arrived in Réunion. The critical geographic detail — one that directly shapes the genetic profile visible in G25 calculators — is where these workers originated.
The majority were recruited from the Madras Presidency: predominantly Tamil-speaking workers from Tamil Nadu and Malabar (Malayalam-speaking) workers from what is now Kerala. This is why modern Réunionnais carry a strongly Dravidian South Indian ancestry signal rather than the North Indian (Indo-Aryan) signal that characterizes descendants of Bihari or Uttar Pradesh indentured workers sent to Trinidad or Guyana. The "Malbars" — as the Tamil and Malabar community is known in Réunion — are the island's largest Indian-descended group, and their Dravidian genetics dominate the South Asian component. A secondary "Zarabes" community — Gujarati Muslim traders who came as free merchants, not indentured workers — adds a minor northwestern Indian and Middle Eastern genetic thread.
Why "North_Indian" Appears in Réunion Calculator Results
G25 World Regions calculators frequently assign a component labeled "North_Indian" to Réunionnais individuals despite the historical evidence that South Indian workers dominated the indentured wave. This reflects a known limitation of PCA-based ancestry calculators: South Indian Dravidian populations (Tamil, Malabar/Kerala) carry a mix of AASI (Ancient Ancestral South Indian) and Iranian-related ancestry that, in Global25 space, does not resolve perfectly into any single "South Indian" reference cluster. Some "North_Indian" assignment in Réunionnais results captures the Iranian-ancestry component of South Indian populations, not genuine Gangetic plain ancestry. Similarly, the "Negev Bedouin" component (3.6%) visible in the Your Roots DNA regional breakdown for Asia is almost certainly a reference-panel proxy: Bedouin populations share certain deep Eurasian affinities with South Asian groups, and are used by the algorithm to absorb variance that cannot be captured by the Kerala and Andhra Pradesh references alone. Genuine Arab Middle Eastern ancestry from Comorian or Indian Ocean trading connections cannot be excluded, but this signal is more parsimoniously interpreted as a proxy artifact. The maternal haplogroup M32c — a Dravidian marker distributed across Tamil Nadu, Kerala, and Sri Lanka — confirms South Indian female lineage in the individual analysed.
IV. The European Foundation — French Colonial Settlers
The European component in Réunionnais Creoles is predominantly northwestern French, consistent with the documented origins of the island's settler population. The "grands blancs" — wealthy plantation owners — and the "petits blancs" — small farmers who occupied the island's interior highland cirques (Mafate, Cilaos, Salazie) — both originated primarily from Normandy, Brittany, Poitou-Charentes, and Aquitaine. The Your Roots DNA regional breakdown for a Réunionnais individual shows Ardennes Department (6.5%) and Aquitaine Landes (3.2%) as the dominant French regional signals alongside Scotland Lanarkshire (6.8%) — the latter a counterintuitive result requiring explanation.
The Scottish signal in some Réunionnais European ancestry profiles has multiple possible origins. British administration from 1810 to 1815 introduced British military and administrative personnel. Scottish merchants played a disproportionate role in British Indian Ocean commercial networks throughout the 19th century, and some Scottish trading families settled in the island. Additionally, commercial ancestry reference panels may partially overlap British and northwestern French reference populations — both share substantial Bell Beaker period ancestry — inflating "Scottish" scores when the genuine ancestry is Norman or Breton French. The two hypotheses are not mutually exclusive, and genealogical records in specific family lines would be required to distinguish them.
The "Petits Blancs des Hauts" — An Isolated European Enclave
Among the most genetically distinctive subgroups in Réunion are the "petits blancs des hauts" — poor white farming families who retreated to the island's mountainous interior cirques as the sugar economy consolidated in the 18th century. Isolated in steep-sided volcanic basins with limited contact with coastal populations for generations, some of these families maintain genealogical records tracing ancestry to specific French provinces from the 17th century — making them one of the few populations in the Indian Ocean whose European ancestry can be geographically assigned with genealogical rather than merely statistical precision. Their genetic profile shows substantially higher European ancestry fractions than the coastal Creole average, and correspondingly lower African and Indian components.
V. Haplogroup Evidence — Reading Individual Lineages
VI. Modern Réunionnais in G25 Space
In Global25 PCA space, Réunionnais Creole individuals occupy a position unlike any reference population — pulled simultaneously toward the Bantu African, South Asian, and European clusters in a pattern that has no close continental analog. The first principal component in Global25, which primarily separates sub-Saharan African populations from Eurasian ones, places heavily African-admixed Réunionnais individuals at PC1 values around -0.25 — consistent with roughly 40–45% sub-Saharan African ancestry. This is substantially more negative than any European population and more positive (less African) than any West African reference population.
Crucially, the African pull in PC1 is directed toward the East African rather than the West African cluster — distinguishing Réunionnais coordinates from Caribbean Creole populations, where West African ancestry dominates and the PC1 displacement follows a different axis. The Malagasy/Austronesian component adds a secondary displacement along dimensions capturing Southeast Asian ancestry — a direction with no equivalent in Atlantic Creole populations. In a PCA that includes Southeast Asian reference populations alongside African and European ones, Réunionnais individuals are pulled toward the Malagasy cluster in a way that distinguishes them from Mauritian Creoles (whose Indian component is proportionally higher) and from Seychellois Creoles (whose African component is higher and more consistently East African).
VII. A Genetic Timeline — How Réunion's Population Was Built
Unlike the deep-time genetic palimpsest of a European island like Sicily, Réunion's demographic history can be reconstructed from colonial census records and slave trade documentation with unusual precision. The chart below maps estimated ancestry composition at each major demographic turning point, from the island's first permanent settlement to the modern Réunionnais Creole profile.
First settlers
Coffee economy
Peak slavery
Abolition
End of indenture
G25 NNLS est.
European proportions estimated from French colonial censuses. African breakdown (East / West African / Malagasy) estimated from slave trade records (Gerbeau 1978; Fuma 1992) combined with G25 NNLS modelling of modern Réunionnais individuals. Post-1946 "Zoreils" immigration from metropolitan France increases the European fraction in some individuals. Values are approximate; substantial intra-population variation exists by community.
VIII. Ancestry Profiles — Samuel and the Indian Ocean Creole World
The following ancestry bars present the G25 World Regions K248 NNLS breakdown for the individual "samuel_scaled," modelled using the Vahaduo calculator (distance 1.11%). The components are consolidated into five ancestry clusters for clarity and compared with estimated profiles for the broader Réunionnais Creole population and related Indian Ocean island populations. The "Malagasy/Austronesian" bar aggregates Malagasy, North_Indonesian, Lao, and Vietnamese components — which collectively reflect, in this context, predominantly Malagasy enslaved ancestry rather than direct Southeast Asian connections.
IX. Comparison Table — Réunion in the Indian Ocean Creole World
| Population | E. African ~% | W. African ~% | European ~% | S. Asian ~% | Malagasy/SE Asian ~% | Key note |
|---|---|---|---|---|---|---|
| Samuel (Réunion, individual) | ~43.8 % | ~8.0 % | ~20.4 % | ~20.4 % | ~15.4 % | Higher E. African and Malagasy than population average; K248 NNLS |
| Avg. Réunionnais Creole | ~30–35 % | ~7–10 % | ~25–30 % | ~20–25 % | ~8–12 % | 4-way admixed; South Indian dominant Indian component |
| Mauritian Creole | ~25–30 % | ~5–8 % | ~18–22 % | ~35–40 % | ~8–12 % | Higher Indian (incl. North Indian Bhojpuri); British colonial influence |
| Seychellois Creole | ~45–55 % | ~5–8 % | ~25–35 % | ~10–15 % | ~5–10 % | Higher African fraction; lower Indian component than Réunion |
| Comorian (average) | ~60–70 % | <5 % | <3 % | ~5–10 % | ~15–25 % | Predominantly Bantu + Malagasy; minimal European; Arab admixture ~5–10% |
| Malagasy (average) | ~60–70 % | <2 % | <2 % | <2 % | ~30–40 % | Bantu + Austronesian 2-way; primary source of SE Asian signal in Indian Ocean Creoles |
| Petits blancs des hauts | <5 % | <2 % | ~85–95 % | <5 % | <3 % | Isolated European enclave; highland cirques; markedly divergent from coastal Creole profile |
X. Myths and Realities
Common Misconception
"The Southeast Asian components in my Réunion ancestry result mean I have Chinese ancestry."
Genetic Reality
For most Réunionnais Creoles, Southeast Asian calculator components — "Indonesian," "Malagasy," "Lao," "Filipino" — reflect Malagasy enslaved ancestry, not Chinese immigration. Malagasy people carry ~30–40% Austronesian (Borneo-origin) ancestry, which reads as Southeast Asian in PCA-based calculators. The Chinese community in Réunion is small (~1–2% of the population). Genuinely Chinese ancestry would produce much higher East Asian (Mandarin/Cantonese cluster) signals, not Indonesian or Lao ones. (Brucato et al. 2017)
Common Misconception
"Réunion's Indian ancestry comes from North India — Gujarat, the Hindi belt."
Genetic Reality
The dominant Indian ancestry in Réunion is Dravidian South Indian — Tamil Nadu and Kerala (Malabar) — reflecting the demographic reality of "engagisme." Over 45,000 Madras Presidency workers arrived between 1848 and 1882, vastly outnumbering the smaller North Indian wave. "North_Indian" components visible in G25 results partly reflect reference-panel resolution: South Indian Dravidian populations have an Iranian-ancestry component that some calculators partially absorb into "North Indian" proxies. The maternal haplogroup M32c, a Dravidian marker, confirms South Indian female lineage. (Fuma 1992)
Common Misconception
"Réunion's African ancestry is like Caribbean African ancestry — mainly West African Yoruba, Fon, Mandinka."
Genetic Reality
The African ancestry in Réunion is predominantly East and Southeast African — Mozambican Bantu (Makua, Yao, Swahili corridor) rather than West African. This is the key genetic difference between Indian Ocean and Atlantic Creoles: Caribbean populations received most enslaved people from the Guinea Coast because Atlantic shipping made that economical; Réunion received most from Mozambique because the French Indian Ocean network ran through Portuguese East Africa. This distinction is clearly visible in G25 PCA space. (Gerbeau 1978)
Common Misconception
"All Réunionnais have similar mixed ancestry — it's a Creole island so everyone is the same."
Genetic Reality
Réunion shows substantial intra-island variation by community. The "petits blancs des hauts" of the highland cirques carry predominantly European ancestry. Tamil Malbars-descended families carry predominantly South Indian ancestry. "Zoreils" (metropolitan French settlers) carry purely European profiles. The island's founding communities admixed unevenly and relatively recently: the five main ancestry streams only began mixing across community lines in the 19th century, and distinctive community-level genetic clusters remain detectable in modern individuals.
References
- Gerbeau H. (1978). Les esclaves noirs: pour une histoire du silence. Paris: André Balland. Livre — traite négrière Réunion
- Fuma S. (1992). L'esclavagisme à la Réunion, 1794–1848. Paris: L'Harmattan. Livre — esclavage et abolition
- Fuma S. (1999). De l'Inde du Sud à La Réunion. Saint-Denis: Graphica. Livre — engagisme indien
- Chaudenson R. (1979). Les créoles français. Paris: Nathan. Livre — créolisation et démographie
- Brucato N. et al. (2017). The Malagasy Ancestry Comes from an Historical Malay Trading Post in Southeast Borneo. Molecular Biology and Evolution 34(12): 2996–3010. DOI:10.1093/molbev/msx226 Génétique malgache — origine austronésienne
- Pierron D. et al. (2014). Genome-wide evidence of Austronesian–Bantu admixture and cultural reticulation in Madagascar. PNAS 111(3): 936–941. DOI:10.1073/pnas.1321860111 Admixture Madagascar — génome entier
- Hurles M.E. et al. (2005). The Dual Origin of the Malagasy in Island Southeast Asia and Sub-Saharan Africa: Evidence from Maternal and Paternal Lineages. American Journal of Human Genetics 76(5): 894–901. DOI:10.1086/430051 Double origine malgache — mtDNA + Y
- G25 World Regions K248 Calculator — Vahaduo (Davidski). Modern population averages: Moriopoulos 2025 Modern Population Collection. ExploreYourDNA Calculator 173.
- G25 World Countries Calculator K105. ExploreYourDNA Calculator 148. Distance 1.40% (samuel_scaled).
- Your Roots DNA Ancestry Report YRDBR6NRPX, Standard Mode, 29.12.2024. YourRootsDNA.com.
- 23andMe Ancestry Report — r/23andme user "yayaskin" (2019). European 33.7% / Sub-Saharan African 33.3% / Central and South Asian 17.1% / East Asian and Native American 10.5%. Maternal haplogroup M32c; paternal haplogroup T-L446. Reddit thread.