Original Synthesis — 4 Subspecies · World Bee Day

Subspecies Refugia

Four genetically distinct Apis mellifera subspecies — ruttneri, macedonica, jemenitica, mellifera — survive intact in four very different countries. Three did so by accident. One — Malta — did so by design. The contrast is the conservation lesson.

A synthesis drawn from our 135-country honey atlas. Each case has a full country guide — this page extracts the genetic-conservation pattern that only becomes visible when you read all four together.

4
Subspecies, 4 countries
316 km²
Smallest endemic range (Malta)
60+
Years — Eritrea isolation
1
Deliberate model (Malta)

Why subspecies preservation is its own story

Most discussion of bee conservation focuses on colony losses — the annual percentage of hives that fail to overwinter, the headline number from BeeInformed and COLOSS surveys. That framing matters, but it misses a quieter form of loss: the slow homogenisation of native honey bee genetics by a century of commercial queen imports. When a beekeeper in Albania or Iceland stocks an apiary with imported A.m. ligustica or Buckfast queens, the local subspecies does not collapse — it is gradually replaced through drone drift and queen mating, often within two or three generations.

This kind of loss is invisible at the apiary level: the colonies are healthy, the honey flows, the beekeeper is satisfied. But the local subspecies — adapted over millennia to local floral sources, climate, and pathogen pressure — is being diluted at population scale. Across most of Europe and the Middle East, this dilution is now nearly complete. Native subspecies survive primarily where some external factor blocked imports for decades.

Four cases stand out. Three are accidents of history. One is the result of deliberate policy. Read together they suggest that the genetic-erosion process is reversible — but only when something stops the imports.

Four subspecies, four mechanisms

CountrySubspeciesMechanismDurationRange
🇲🇹MaltaA.m. ruttneriDeliberate policyActive program since 1990s; subspecies formally described 1997Endemic to Malta — 316 km² total range, smallest of any A. mellifera subspecies
🇦🇱AlbaniaA.m. macedonicaPolitical isolation47 years of total isolation; ongoing post-1991 dilution pressureAlbania, North Macedonia, northern Greece, Kosovo, parts of Serbia and Bulgaria
🇪🇷EritreaA.m. jemeniticaGeopolitical isolation60+ years (1961–present)Yemen, Saudi Arabia, Oman, Eritrea, Sudan, Ethiopia (lowlands), parts of Somalia
🇮🇸IcelandA.m. melliferaGeographic barrierCenturies of selection; Varroa-free until 2022Historically: Atlantic Europe from Ireland to Scandinavia. Now restricted; Iceland is one of the few unmixed populations
The replicable model

Malta — deliberate policy

Maltese beekeepers chose to preserve A.m. ruttneri. The Maltese Beekeepers Association coordinates closed mating; national policy restricts queen imports under the Bee Disease Free framework. The 250 km Sicilian Channel reinforces the policy but did not create the outcome alone. Other countries have approximated this model at regional scale (Slovenia for A.m. carnica, Læsø for A.m. mellifera) — Malta is unique in that the deliberate program covers an entire endemic range.

The non-replicable models

Albania, Eritrea, Iceland — accidents

Communist closure, liberation war, and 800 km of Atlantic — these are not policy choices. They are historical events that happened to produce the same outcome as Malta’s deliberate program by stopping queen imports for decades. Their lesson is structural, not procedural: import restriction works. Their continuity cannot be engineered without the catastrophe that produced it.

Case studies

🇲🇹
Deliberate policy

Malta — A.m. ruttneri

Maltese Beekeepers Association closed-mating program

Morphometric / lineage marker

Distinctive cubital index and tergite pigmentation; intermediate between A.m. sicula (Sicily) and A.m. intermissa (North Africa)

Subspecies range

Endemic to Malta — 316 km² total range, smallest of any A. mellifera subspecies

Malta is the only entry in this cluster where preservation is the result of deliberate human policy, not catastrophe. The Maltese Beekeepers Association operates a closed-mating queen-rearing program; the Maltese Islands enforce a strict Bee Disease Free protocol that restricts queen imports. The 250 km Sicilian Channel limits drift from A.m. ligustica colonies in Sicily. Together these create a refugium-by-design that other countries cannot easily replicate without an island geography — but the policy framework itself is exportable. Sheppard et al. (1997) formally described A.m. ruttneri based on morphometric divergence; subsequent mitochondrial DNA work has confirmed the lineage is distinct from neighbouring populations. Malta's 316 km² range makes ruttneri the most narrowly endemic Apis mellifera subspecies on Earth.

Sheppard, Arias, Grech, Meixner (1997), "Apis mellifera ruttneri, a new honey bee subspecies from Malta," Apidologie 28(5):287–293

Full country guide
🇦🇱
Accidental — political isolation

Albania — A.m. macedonica

Hoxha-era Communist closure (1944–1991): no foreign queen imports, no commercial inputs

Morphometric / lineage marker

Lineage C (Carniolan-Balkan); cubital index 2.4–2.8; characteristic forewing venation distinct from A.m. carnica and A.m. cecropia

Subspecies range

Albania, North Macedonia, northern Greece, Kosovo, parts of Serbia and Bulgaria

Under Enver Hoxha's regime, Albania became the world's most economically closed state. The same isolation that imposed scarcity on Albanians also blocked the queen-bee imports that homogenised most of European apiculture across the 20th century. A.m. macedonica — adapted to Mediterranean scrubland and continental mountain meadows — survived in commercial apiaries that other Balkanic populations no longer have. Hatjina et al. (2014) documented the surviving morphometric integrity of Albanian populations relative to commercial populations elsewhere in southeastern Europe, where queen imports from Italian (A.m. ligustica) and Carniolan (A.m. carnica) stocks have substantially diluted local lineages.

Hatjina, Costa, Büchler, Uzunov, Drazic et al. (2014), "Population dynamics of European honey bee genotypes," Journal of Apicultural Research 53(2):233–247

Full country guide
🇪🇷
Accidental — geopolitical isolation

Eritrea — A.m. jemenitica

30-year liberation war + post-independence closure: zero documented queen imports

Morphometric / lineage marker

O-lineage (oriental); smallest of all A. mellifera subspecies; longer proboscis (6.0–6.4 mm); pale yellow tergite pigmentation; exceptional heat and arid-landscape tolerance

Subspecies range

Yemen, Saudi Arabia, Oman, Eritrea, Sudan, Ethiopia (lowlands), parts of Somalia

Eritrea's independence war (1961–1991) and subsequent political isolation under President Isaias Afwerki produced what may be the world's most genetically unmanipulated A.m. jemenitica population. This is the same subspecies that produces authenticated Yemeni Sidr honey — the small, heat-tolerant, long-tongued bee adapted to nectar-sparse arid landscapes. Across the rest of the O-lineage range (Yemen, Saudi Arabia, Oman) commercial Egyptian-pattern beekeeping and queen imports have introduced significant genetic drift. Eritrea's closure stopped that. Meixner et al. (2013) catalogued the genetic distinctiveness of jemenitica populations across the Horn of Africa; Eritrean populations sit at the conservative end of that gradient.

Meixner, Pinto, Bouga, Kryger, Ivanova, Fuchs (2013), "Standard methods for characterising subspecies and ecotypes of Apis mellifera," Journal of Apicultural Research 52(4):1–28

Full country guide
🇮🇸
Accidental — geographic isolation

Iceland — A.m. mellifera

800+ km Atlantic separation; strict bee-import biosecurity; cold-climate selection

Morphometric / lineage marker

M-lineage (West-European Dark Bee); high cubital index variance; cold-hardy with extended winter cluster behaviour; flight onset at 8°C (vs. 12°C for ligustica)

Subspecies range

Historically: Atlantic Europe from Ireland to Scandinavia. Now restricted; Iceland is one of the few unmixed populations

Iceland's 800 km Atlantic moat blocked Varroa destructor — the mite that has reshaped global beekeeping since the 1970s — until 2022, the latest arrival in any European country. The same geographic barrier limited commercial queen imports for most of the 20th century. The result was a cold-adapted A.m. mellifera population that other European countries had largely lost to A.m. ligustica and Buckfast hybrid imports. The population's 6–8 week flight season selected hard for cold-hardy traits: extended winter cluster, conservative brood-rearing, lower swarm tendency. When Varroa was confirmed in 2022 the conservation calculus shifted, but the genetic baseline shaped by a millennium of island selection remains.

Pinto, Henriques, Chávez-Galarza et al. (2014) on European A.m. mellifera genetic structure; Varroa first confirmed in Iceland reported in Morgunbladid (2022)

Full country guide

What this means for honey buyers

Subspecies labelling is rare but valuable

Most honey labels list country and floral source — not subspecies. When a producer does specify (Slovenian Kranjska sivka, Maltese ruttneri honey, Icelandic mellifera honey) it usually indicates an active conservation program and a meaningful flavor distinction worth paying for.

Geographic isolation correlates with authenticity

Honey from islands, mountain valleys, and politically closed economies is more likely to come from native subspecies than honey from countries with active queen-import markets. This is a soft heuristic, not a guarantee — but it is more reliable than most marketing claims.

Native subspecies produce distinctive flavors

A.m. jemenitica's long proboscis allows access to deep-corolla flowers (Ziziphus, Acacia) that European subspecies cannot reach efficiently. A.m. mellifera's extended winter cluster produces honey from late-season heather flows that warmer-climate subspecies miss. The bee is part of the terroir.

These populations are not permanent

Post-1991 Albania, post-2022 Iceland (Varroa-arrival), and any future opening of Eritrea would all introduce new genetic dilution pressure. Malta's policy framework is more durable but depends on continued enforcement. The window for accessing these honeys in their most distinctive form is finite.

Related synthesis

Accidental sanctuaries: when wars and geography preserve ecosystems

Three of the four cases here (Albania, Eritrea, Iceland) appear in our broader inadvertent-conservation synthesis alongside Laos’s UXO-meadow ecology. That page covers the wider pattern — where catastrophe preserves wildflower habitat as well as bee genetics. This page narrows the focus to subspecies preservation specifically and adds Malta as the deliberate-policy contrast.

Read the inadvertent-conservation synthesis

Frequently asked questions

How many Apis mellifera subspecies exist worldwide?
Approximately 27–31 subspecies of Apis mellifera have been formally described, organised into four (sometimes five) evolutionary lineages: M (West European Dark Bee, including A.m. mellifera), C (Carniolan-Italian-Balkan, including A.m. carnica, A.m. ligustica, A.m. macedonica), A (African, including A.m. scutellata and A.m. intermissa), O (oriental, including A.m. jemenitica, A.m. caucasica), and the more recently distinguished Y (Ethiopian highland) lineage. The exact count depends on which morphometric and molecular criteria are applied. A.m. ruttneri (Malta) was formally added to the C-lineage in 1997 by Sheppard et al.
Why does subspecies preservation matter for honey buyers?
Subspecies differ in foraging behaviour, tongue length, climate adaptation, and disease resistance — and these differences shape honey character. A.m. jemenitica's longer proboscis allows it to access narrow-corolla flowers that A.m. mellifera cannot. A.m. macedonica's adaptation to Mediterranean scrubland produces different floral fingerprints from imported Carniolan stock on the same landscape. When local subspecies are replaced by commercial imports the floral profile of the resulting honey changes even if the apiary location is identical. Native-subspecies honey is, in this sense, irreproducible by geography alone.
What is the Maltese closed-mating program and could it be replicated elsewhere?
The Maltese Beekeepers Association maintains a coordinated network of apiaries and queen-rearing stations on the Maltese Islands, with national restrictions on queen imports under the Bee Disease Free framework. The 250 km Sicilian Channel reinforces the policy by limiting drone drift from Sicilian A.m. ligustica colonies. The policy framework itself — restricted imports, coordinated mating, association-level cooperation — is exportable; what makes Malta unusual is that the geography reinforces the policy. Similar deliberate refugia exist in Læsø (Denmark, A.m. mellifera), Norway's closed conservation zones, the Irish Native Honeybee Society's island programs, and Slovenia (A.m. carnica protected by national law as the country's only legal subspecies). Malta is distinguished by the small size of its endemic range and the formal subspecies status of its bee.
Why is A.m. ruttneri considered the most narrowly endemic Apis mellifera subspecies?
A. mellifera ruttneri's natural range is essentially identical to the Maltese Islands themselves — approximately 316 km², the surface area of Malta, Gozo, and Comino combined. Most other Apis mellifera subspecies range across hundreds of thousands or millions of square kilometres (A.m. carnica covers most of central and southeastern Europe; A.m. jemenitica covers the Arabian Peninsula and Horn of Africa). No other subspecies is restricted to a comparable footprint. This narrow endemism is also what makes ruttneri vulnerable: a single uncontrolled import event could effectively dilute the entire global population.
Are these populations still genetically distinct in 2026?
The published evidence indicates yes, but with important caveats. A.m. ruttneri (Malta) remains the most rigorously protected by active policy. A.m. mellifera in Iceland retains its M-lineage character despite the 2022 Varroa arrival, which adds a treatment-management layer but does not in itself dilute genetics. A.m. macedonica in Albania faces post-1991 commercialisation pressure as queen imports have begun; rural highland populations are less affected than commercial lowland operations. A.m. jemenitica in Eritrea faces the lowest documented dilution of any O-lineage population because the country's ongoing isolation has prevented import programs that have substantially modified Saudi and Yemeni populations. None of these populations is permanent in the sense that no further conservation action would be needed.
Can the Albanian, Eritrean, and Icelandic outcomes be turned into policy?
Partly. The structural lesson — that subspecies dilution is driven primarily by commercial queen imports, and that restricting those imports preserves native genetics — is fully replicable as policy. What is not replicable is the catastrophe-driven duration of restriction. Malta, Slovenia, Læsø (Denmark), Norway's protected zones, and Ireland's Native Honeybee Society programs all pursue the same outcome through deliberate import restriction; none has yet matched the 47-year continuity of Albania's political isolation. The Maltese model is the most replicable because it combines policy with a small, geographically bounded community of beekeepers — a feature most countries can approximate at the regional rather than national level.
RHG

Edited by Sam French · Raw Honey Guide Editorial Team

Reviewed by certified beekeepers and apiculture specialists. Our editorial team consults with professional beekeepers, food scientists, and registered dietitians to ensure accuracy. Health claims are cited against peer-reviewed literature from Cochrane, JAFC, BMJ, and Nutrients.

Expert ReviewedFact CheckedEditorial Policy ↗