ADVERTISEMENT

Last updated: · 14 min read

In brief
Ireland’s poorest families grew a single potato in 1845, the Lumper, and every plant was in effect the same plant. When blight arrived, it had nothing to sort through. Land tenure and British policy turned that crop failure into mass death: at least one million people died. The Cavendish banana is a clone too, and Tropical Race 4, a fungus with no commercial treatment, reached Latin America in 2019. Genebanks hold the diversity; breeding a successor is the slow part. And the endowment meant to keep genebanks running is still less than half its target.

The artificial banana flavor in children’s medicine, candy, and budget ice cream tastes more like a variety that stopped being commercially viable in the 1960s than like the banana you can buy today. The compound is called isoamyl acetate, and the variety it more closely resembles is the Gros Michel, which dominated global export markets until Race 1 of Fusarium wilt moved through Central American plantations across the 1950s and 60s and effectively ended its commercial life. But the resemblance is a coincidence. Artificial banana flavor was already being sold in “fruit essences” in the 1850s, and recipes built on isoamyl acetate were circulating by the 1860s, decades before the Gros Michel became the world’s leading export banana. The candy flavor is essentially that one ester; the Gros Michel happens to be rich in the same compound. The flavor was never derived from the Gros Michel.

The Gros Michel is not extinct, despite what gets repeated online. It is still grown on smallholder farms across Latin America and the Caribbean, and specialty growers still sell it, though Race 1 drove it out of the export trade.

The same pathogen family is now moving against the Cavendish. Tropical Race 4, a distinct Fusarium strain against which the Cavendish has no inherited resistance, has been spreading through Cavendish plantations for decades, reached Latin America in 2019, and shows no sign of being contained. What stands between that banana and the Gros Michel’s fate is the world’s genetic backup system: seed vaults, genebanks, and the breeders who draw on them.

The mathematics of disappearance

Humans have cultivated some 6,000 plant species for food and agriculture. Only about 150 are grown on any meaningful scale today. Fewer than a dozen, among them wheat, rice, maize, soybean and potato, account for the majority of the calories humanity eats.

The biggest driver was the Green Revolution. Between the 1950s and 1970s, agricultural scientists replaced thousands of landraces, the locally adapted varieties that farmers had refined over generations of selection in specific soils and microclimates, with centrally bred, high-yield cultivars engineered for uniformity and productivity. The gains in yield were enormous, and the people driving them believed the arithmetic of mass hunger justified narrowing what farmers grew. Breeders later rebuilt some diversity inside their modern variety pools; much of what had grown in farmers’ fields did not come back.

Across Asia’s major rice-growing countries, decades of standardization pushed thousands of locally adapted landraces out of active farming systems, and the modern varieties that replaced them are far more alike than the patchwork they displaced. The same pattern holds for maize, wheat and soybean. Variety after variety moved to the margins or out of cultivation entirely, pushed there by the deliberate pursuit of yield at scale.

One figure appears everywhere in writing on this subject: that 75 percent of crop genetic diversity has been lost since 1900. The basis for that specific percentage is unclear, and the FAO avoids assigning a single global figure. Nobody knows exactly how much has been lost, and the most widely cited number cannot be robustly measured. Its constant repetition is part of how this crisis gets discussed, and once the number’s weakness is noticed, critics can use it to dismiss the underlying reality. That reality, documented crop by crop and region by region, is not in serious dispute: in the largest review of the evidence, four out of five studies found real losses of crop diversity, landraces have vanished from whole regions, the narrowing was largely intentional, and its consequences are maturing.

The Green Revolution's arithmetic

Norman Borlaug's wheat varieties are credited with averting famines that could have killed hundreds of millions. He won the Nobel Peace Prize in 1970, and the varieties that narrowed global crop diversity did feed populations that existing agricultural systems could not sustain. The trade-off was accepted as the price of buying time, made in conditions where the alternative was famine. The genetic cost was never paid down.

The mechanism — Ireland, 1845

Ireland in 1845 was not growing a diverse potato crop. Its poorest families, above all in the west and south, were growing one potato: the Lumper, the country’s main variety. It was propagated almost entirely through vegetative reproduction, with tubers cut from existing plants and replanted rather than grown from seed. Vegetative propagation produces virtually no genetic variation. Every Lumper in every field was, in effect, the same plant.

When Phytophthora infestans arrived, that sameness gave it nothing to sort through. Better varieties such as the Apple and the Cup also grew in Ireland, and they fell to the blight too. No potato grown in Europe in 1845 had meaningful resistance. The strongest defenses would later come from wild potatoes in the Americas, Mexico’s Solanum demissum above all, and breeders did not begin moving them into cultivated varieties until the twentieth century.

Ireland’s poor did not come to depend on the Lumper by choice; the colonial land tenure system pushed them onto it. Absentee landlords and rack-rents squeezed tenants, and holdings were subdivided through generations until only the crop with the most calories per acre could sustain a family. On one-acre plots, under a system built to extract maximum rent from minimum land, subsistence farmers had no margin to grow anything else.

ADVERTISEMENT

The blight struck across Europe in 1845 and 1846, and Germany, Belgium and the Netherlands all lost crops, but only Ireland starved at this scale. Food exports from Ireland, including grain, livestock and dairy, continued through the famine years to meet landlord obligations and British market demand. Whig government relief was constrained by laissez-faire ideology; Charles Trevelyan, the Treasury official who administered much of the response, believed government intervention would undermine the natural correction the market was performing. At least one million people died, and some two million emigrated within a decade. The 1851 census counted 1.6 million fewer people than the census of 1841, a fifth of the country gone.

The causal chain has three links. Biologically, the crop had no defense against a new pathogen, and uniformity left the poorest families with nothing else in the ground. Colonial land tenure made a single high-yielding potato the only viable option for subsistence farmers. British policy turned a crop failure into mass death. A more varied Irish crop of the same European potatoes would not have survived the blight. The resistance Europe lacked existed only in the potato’s wild relatives across the Americas.

What the International Potato Center in Lima holds today is the range Europe did not have in 1845. The CIP genebank conserves some 4,670 traditional landrace cultivars and more than 2,500 accessions representing 140 wild potato species, genetic breadth assembled from across the Andes and beyond: the diversity that vegetative monoculture forecloses.

The Andean collection

The CIP assembled its potato collection through decades of fieldwork across the Andean highlands, where potato originated and where wild relatives continue to evolve alongside pathogens, altitude variation, and seasonal extremes. The collection has supplied breeders and researchers around the world for decades. It can answer the biological link in the three-part chain; the land system and the relief policy were political failures no genebank could have prevented.

The archive

Norway bored a tunnel into a sandstone mountain on the Arctic archipelago of Svalbard and in 2008 opened the Global Seed Vault. The facility sits 130 meters above sea level on Spitsbergen, the main island, refrigerated to minus 18 degrees Celsius and designed so that, should power fail, the surrounding permafrost keeps the seeds frozen. The operating logic is that even if the institution running it collapses, the seeds survive.

By mid-2026, the vault held more than 1.4 million seed samples from 134 depositors, still less than a third of its 4.5 million sample capacity. The remaining space is room for material that could still be collected, if it still exists.

Norway owns the facility. NordGen, the Nordic Genetic Resource Center, manages operations. The Crop Trust provides financial support to depositors and co-funds running costs. Seeds deposited remain the legal property of the depositing institution. NordGen staff can move boxes within the facility but cannot open them without the depositor’s authorization. Researchers and plant breeders cannot request seeds from the vault directly; they must go to the original depositing genebank. The vault works like a backup drive whose files belong to whoever uploaded them.

The working collections are elsewhere, above all in the eleven CGIAR genebanks, which together hold more than 700,000 accessions of some 3,000 plant species. These are the collections that researchers draw from and breeders request. Svalbard is the backup of the backup, the copy of last resort if something goes wrong with an active collection.

The vault’s first real test came from Syria. The civil war had made it impossible to operate the ICARDA genebank in Aleppo, which held some 141,000 accessions, one of the world’s great collections of landraces and wild relatives of wheat, barley, lentil, chickpea, faba bean and forages. ICARDA became the first institution in history to withdraw seeds from the Svalbard vault. The first withdrawal, in September 2015, brought back some 38,000 samples, split across two destinations: forages, faba beans, grasspea, and wild relatives of cereals and pulses went to Lebanon, and cultivated wheat, barley, lentil, and chickpea went to Morocco. Two more withdrawals followed, in 2017 and 2019. The seeds were sown, multiplied and rebuilt into working collections, and from 2017 onward ICARDA sent fresh seed back to the vault, more than 42,000 accessions by 2019.

In that test the system worked as designed: the vault protected a threatened genebank from political instability and allowed material to be moved, regenerated, and returned to use. It has not been tested as a recovery mechanism for an active crop failure at agricultural scale, for a blight that has already swept through farms in multiple countries while a breeding pipeline races to catch up. The Cavendish is beginning to pose that second test.

ADVERTISEMENT

The replay

The Cavendish banana is a clone. Every commercial Cavendish on every supermarket shelf is a near-identical copy of every other; the few named Cavendish selections are clonal variants of the same plant. The variety propagates through vegetative offshoots from existing plants rather than from seed, so there is no sexual recombination and almost no variation. That uniformity, the same property that left the Lumper exposed, is why it accounts for almost all of the global banana export trade: it gives consistent size, ripening and shelf life.

Bananas are the world’s most exported fruit, a trade worth about $10 billion a year, and Latin America ships the majority of them, around 60 percent of the volume exported in 2022. More than 400 million people depend on bananas for food security and income, many of them in the world’s lowest-income producing regions.

Tropical Race 4 is a strain of Fusarium oxysporum f. sp. cubense, the same fungal species as Race 1 but genetically distinct. The Cavendish’s resistance to Race 1, the property that made it the replacement for the Gros Michel, offers nothing against TR4. There is no commercial treatment. Once TR4 establishes in soil, it persists for decades; it cannot be rotated out or fumigated away, so whatever ground it reaches stays contaminated.

TR4 first hit Cavendish plantations in Taiwan, tore through new plantations in Indonesia and Malaysia in the early 1990s, reached northern Australia before the decade was out, and arrived in Africa in 2013 and Latin America in 2019. By early 2024, Peru’s national plant protection authority had eradicated more than 400 separate outbreaks. The confirmed infected area covered roughly 330 hectares of Peru’s approximately 171,000 banana-growing hectares, a foothold that flooding after Cyclone Yaku in 2023 helped spread.

The export industry’s response to TR4 has so far tracked its response to Race 1: contain the disease and keep planting the Cavendish. Race 1 killed the Gros Michel’s commercial dominance through the 1950s and 60s, and the industry replaced it with the Cavendish. It then had sixty years to diversify away from dependence on a single export clone. It did not, because the Cavendish was profitable and exactly what consumers recognized. The lesson it took from the Gros Michel was to switch when a pathogen forces a switch. This time the search for a successor has at least started before the collapse: Australian growers’ levies pay for research meant to speed resistance breeding, and in Peru four varieties bred by the French research institute CIRAD to resist TR4 went into infested fields in 2025, in trials that run through 2027.

The raw material for a different outcome exists, though not in Svalbard. The Cavendish is seedless and most bananas are grown as clones, so banana diversity is kept as living plantlets and frozen tissue instead. The International Musa Germplasm Transit Centre in Leuven, Belgium, operated by the Alliance of Bioversity International and CIAT, holds more than 1,700 accessions of edible and wild banana species, including germplasm showing resistance to Fusarium wilt. They are kept in vitro and in liquid nitrogen, with a safety copy in Montpellier, France.

Wild bananas are where the useful genes are turning up. In early 2026, scientists in Queensland traced resistance to Subtropical Race 4 in a wild banana subspecies to a region on chromosome 5. STR4 is a milder, geographically restricted relative of TR4. TR4 itself causes disease under tropical and subtropical conditions and is the race moving through Latin American plantations now. Resistance proven against STR4 is not proof of resistance to TR4: the chromosome-5 region still has to be tested against the race that matters.

Partial answers exist. A tolerant Cavendish selection, ‘Formosana’, has replaced standard Cavendish on infested farms in the Philippines and Mozambique, and in 2024 Australia approved QCAV-4, a genetically modified Cavendish carrying a resistance gene from a wild banana, though no one yet plans to grow it at scale. A highly resistant banana that growers and consumers will accept as the Cavendish’s successor still does not exist.

Breeding a banana that resists TR4 and also yields, ships and tastes like a Cavendish is slow work, and the closest candidates are still in field trials or waiting for growers to plant them. TR4 spreads through soil that stays contaminated for decades, and it keeps spreading while that work goes on.

The Gros Michel

The Gros Michel is richer and sweeter than the Cavendish, with more of the isoamyl acetate that gives artificial banana flavor its smell. The variety lost the export trade but was never eliminated: specialty growers still sell it, and smallholder farmers across Latin America and the Caribbean still grow it for local markets, even as Race 1 keeps spreading among their farms. Race 1 moved through Central America's large commercial plantations in the 1950s and 60s and destroyed the export monoculture.

The fine print

The Crop Trust’s plan rests on a specific financial logic: an endowment of $850 million, invested to earn 4 percent a year above inflation, would yield about $34 million annually, enough to cover the essential operations of key genebanks, the Svalbard vault among them, without dependence on annual government appropriations. The endowment model was designed for permanence, removing the core of the system’s survival from the budget cycles of governments with other priorities.

By the end of 2025, the endowment stood at $398 million, less than half of the target. It is growing: 2024 brought a record $37.7 million in new contributions, and investment gains added more in 2025. But the remaining gap is a structural deficit that institutional momentum will not close on its own. The Crop Trust is, in practice, still dependent on the goodwill of a small group of governments and donors whose attention to crop diversity has been episodic at best.

In 2023, as Peru’s plant protection authority raced to stamp out a wave of TR4 outbreaks that would pass 400 by early 2024, the endowment built to keep the world’s genebanks running received less than $6 million in new capital.

Access to the germplasm itself is a smaller problem than it looks. The ITC collection in Leuven, which holds the germplasm from which a TR4-resistant banana variety might eventually be bred, is held in trust for the world under the auspices of the FAO, and breeders anywhere can request its accessions free of charge under the Plant Treaty’s standard terms. Access to the raw material is not the bottleneck; what comes after it is. Turning a wild resistance gene into a commercial variety takes years of breeding, and the varieties that emerge can carry patents and plant breeders’ rights that decide who grows them and at what cost.

Genebanks have a basic limit: they preserve a genetic snapshot frozen at the date of collection. They do not preserve the living, co-evolving diversity of crops grown by farmers in their original landscapes across centuries of ongoing selection. An Andean farmer selecting potato tubers annually across altitude gradients and seasonal variation produces a different kind of genetic resource than a genebank accession collected thirty years ago and kept ever since in a freezer or a test tube. In situ conservation, which supports the farmers whose daily selection maintains that living diversity, has received a fraction of the institutional attention that the concrete tunnel in the Arctic attracts.

The commercial seed market is the final layer. Bayer, Corteva, and Syngenta dominate the global market for high-yield commercial seeds. When genebank material is used in breeding programs, as it regularly is, the resulting commercial varieties frequently carry intellectual property restrictions: patents, plant breeders’ rights, licensing conditions that affect who can grow them and at what cost. Seeds released from vaults in a crisis may end up in varieties owned by companies that sell them back to the farmers in the most vulnerable countries. The seed banks cannot prevent that: once material leaves a vault for breeding, it can pass into the same market.

The CGIAR network is also restructuring. The One CGIAR reform has drawn its research centers under shared governance and common leadership, with efficiency and coordination as the stated aims, though each center remains its own legal entity. Part of what made the system resilient was eleven genebanks with separate management chains, hard to bring down together. The reform changes who oversees them, though, not where the seeds are: the genebanks still sit in different countries with their own collections.

The Plant Treaty's fine print

The International Treaty on Plant Genetic Resources for Food and Agriculture, in force since 2004, governs access to genebank material for research, breeding, and training. Roughly 64 crops and forages are covered; material transfers operate under a standard agreement, and payment into the treaty's fund becomes mandatory only when a commercial product built on that material is closed to further research and breeding; otherwise it is voluntary. The treaty's Benefit-sharing Fund, designed to channel some of those commercial returns back to developing country farmers and conservation programs, has received substantially less than was envisioned at negotiation. Twelve years of talks on reforming the system, raising those payments among them, collapsed without agreement at the treaty's Governing Body in November 2025. The legal framework exists, but it is structurally underfunded and its benefit-sharing provisions remain contested in practice.

TR4 is already establishing in Latin American soil, and the endowment meant to fund the world’s genetic insurance is still more than half short of its operational target. In the Peruvian trial fields, the CIRAD bananas bore their first bunches in early 2026, beside Cavendish plants already sick with the disease. None of them is yet a banana the export trade has agreed to grow.

Related articles

Further reading

As an Amazon Associate, Eikleaf earns from qualifying purchases. The book links below are affiliate links: if you buy through them, we may earn a small commission at no extra cost to you.

Recommended videos

Cite this article
Johan Karlsson. “How a handful of seed banks are the last line against global food collapse.” Eikleaf, 12 May 2026, updated 2 November 2026. https://www.eikleaf.com/how-a-handful-of-seed-banks-are-the-last-line-against-global-food-collapse/

Gen AI Disclaimer

Some contents of this page were generated and/or edited with the help of a Generative AI.

Media

The Svalbard Global Seed Vault was inaugurated in February 2008 as a secure backup facility for the world’s crop diversity on the Svalbard archipelago of Norway. As of February 2025 it has received seed samples from 123 genebanks in 85 countries around the world. Photo: Michael Major for Crop Trust – Wikipedia

Key Sources and References

Svalbard Global Seed Vault, official website, seedvault.no, and news release “Svalbard Global Seed Vault Crosses Major Milestone – 1.4 Million Seed Samples Secured,” 18 June 2026. Total of 1,401,285 seed samples from 134 depositors; capacity of 4.5 million samples; storage at minus 18 °C with permafrost as backup; governance by Norway, NordGen and the Crop Trust. URLs: https://www.seedvault.no/ · https://www.seedvault.no/2026/06/18/svalbard-global-seed-vault-crosses-major-milestone-1-4-million-seed-samples-secured/ · https://www.seedvault.no/about/the-facility/ · https://www.seedvault.no/about/purpose-operations-and-organisation/

CGIAR Genebank Accelerator, “About,” genebanks.cgiar.org. The eleven CGIAR genebanks hold over 700,000 accessions of 3,000 species, freely available to researchers, breeders, communities and farmers under the terms of the International Plant Treaty. URL: https://genebanks.cgiar.org/about/

Crop Trust, Annual Reports 2023–2025, report.croptrust.org. Endowment value $305 million at end 2023 (new contributions $5.9 million), $357 million at end 2024 (record $37.7 million in contributions), and $398 million at end 2025 (with $51 million in investment gains). URLs: https://report.croptrust.org/2023/our-work/finance-and-investments/ · https://report.croptrust.org/2024/what-we-did/finance-and-investments/ · https://report.croptrust.org/2025/administration/finance-and-investment/

Crop Trust, Financing Strategy 2030. Endowment target of $850 million (2022 prices) with a return target of 4 percent a year plus US dollar inflation, supporting a 4 percent withdrawal of about $34 million a year for the essential operations of key genebanks, including the Svalbard Global Seed Vault, and system-wide support. URL: https://www.croptrust.org/fileadmin/uploads/croptrust/Documents/Policy_Documents/Crop_Trust_Financing_Strategy_2030.pdf

Munhoz, T., Vargas, J., Teixeira, L., Staver, C., and Dita, M. “Fusarium Tropical Race 4 in Latin America and the Caribbean: status and global research advances towards disease management.” Frontiers in Plant Science, Volume 15, 2024. TR4 infection data for Peru: 330 hectares affected of approximately 171,000 total banana hectares; more than 400 outbreaks eradicated by early 2024 per SENASA; Latin America’s share of banana exports; ‘Formosana’ (GCTCV-218) on infested farms; Gros Michel still grown in Latin America and the Caribbean for local markets, with Race 1 still spreading among farms; flooding after Cyclone Yaku (2023) and the rise in Peruvian outbreaks. URL: https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2024.1397617/full

Chen, A. et al. “Identification of a QTL conferring resistance to the Subtropical Race 4 of Fusarium oxysporum f. sp. cubense in Calcutta 4 (Musa acuminata ssp. burmannica).” Horticulture Research, Volume 13, Issue 4 (April 2026), article uhag001, Letter; published online 9 January 2026. University of Queensland. STR4 resistance mapped to a genomic region on chromosome 5. URL: https://doi.org/10.1093/hr/uhag001

Khoury, C.K. et al. “Crop genetic erosion: understanding and responding to loss of crop diversity.” New Phytologist, 2022, vol. 233(1), pp. 84–118. Review of a century of evidence on crop diversity loss, the basis of the “75 percent lost” claim, and the case for both ex situ and in situ conservation. URLs: https://doi.org/10.1111/nph.17733 · https://genebanks.cgiar.org/wp-content/uploads/2021/10/Khoury_cropgeneticerosion_GROW_2021_10_18.pdf

Food and Agriculture Organization of the United Nations, The Third Report on the State of the World’s Plant Genetic Resources for Food and Agriculture, 2025. Some 6,000 plant species cultivated for agriculture; no single global diversity-loss percentage. URL: https://doi.org/10.4060/cd4711en

Food and Agriculture Organization of the United Nations, “Biodiversity to nurture people,” chapter 3 of Harvesting Nature’s Diversity (World Food Day publication), 1993. About 150 plant species cultivated on a meaningful scale; the FAO estimate of 75 percent of crop genetic diversity lost. URL: https://www.fao.org/4/v1430e/V1430E04.htm

International Potato Center (CIP), Genebank documentation, cipotato.org. Collection figures: some 4,670 traditional landrace cultivars and more than 2,500 accessions representing 140 wild potato species. URLs: https://cipotato.org/genebankcip/potato/ · https://cipotato.org/genebankcip/potato-wild/

ICARDA, “ICARDA’s seed retrieval mission from Svalbard Seed Vault” (2015) and “ICARDA deposits vital seeds in Svalbard Vault” (2018). First withdrawal of 38,073 samples in 2015: 8,840 accessions to Lebanon and 29,233 to Morocco; Aleppo genebank of 141,000 accessions. URLs: https://icarda.org/media/news/icardas-seed-retrieval-mission-svalbard-seed-vault · https://icarda.org/media/press-release/icarda-deposits-vital-seeds-svalbard-vault-securing-food-generations-come

Svalbard Global Seed Vault, “Withdrawal of ICARDA Aleppo seeds accomplished,” 2019, and Crop Trust, “More Than 50,000 Seeds Withdrawn From Seed Vault,” 2017. Withdrawals in 2015, 2017 and 2019; re-deposits since 2017. URLs: https://www.seedvault.no/2019/09/11/withdrawal-of-icarda-aleppo-seeds-accomplished/ · https://www.croptrust.org/news-events/news/more-than-50000-seeds-withdrawn-from-seed-vault-1/

Food and Agriculture Organization of the United Nations, Markets and Trade: Bananas, fao.org. Latin American share of global banana export origin; Cavendish production volume as just under half of global production. URL: https://www.fao.org/markets-and-trade/commodities-overview/bananas-tropical-fruits/bananas/en

Alliance of Bioversity International and CIAT, International Musa Germplasm Transit Centre (ITC), Leuven, Belgium. More than 1,700 accessions of edible and wild banana species, held in trust under FAO auspices and freely available under the Multilateral System; conserved in vitro and cryopreserved, with a safety duplicate at IRD Montpellier. URL: https://alliancebioversityciat.org/services/genebanks/international-musa-germplasm-transit-centre

Food and Agriculture Organization of the United Nations. “Fighting the deadly disease that is killing the world’s most exported fruit.” FAO Transboundary Plant Pests and Diseases, June 13, 2025. Bananas critical to food security and livelihoods of around 400 million people; global banana trade worth about USD 10 billion a year; TR4 described as spreading through Asia, Africa, the Middle East, and Latin America with no cure available absent replacement with resistant varieties. URL: https://www.fao.org/transboundary-plant-pests-diseases/news/detail/fighting-the-deadly-disease-that-is-killing-the-world-s-most-exported-fruit/en

Arias, P., Dankers, C., Liu, P. and Pilkauskas, P. The World Banana Economy, 1985–2002. FAO Commodity Studies No. 1. Rome: FAO, 2003. “Almost all bananas traded worldwide are Cavendish”; the Cavendish replaced the Gros Michel in international trade because of its resistance to Panama disease. URL: https://www.fao.org/4/y5102e/y5102e04.htm

“‘World’s first’ GM banana approved by regulators in Australia and New Zealand,” AgNavigator, 19 February 2024. Approval of the TR4-resistant QCAV-4 Cavendish. URL: https://www.agnavigator.com/Article/2024/02/19/World-s-first-GM-banana-approved-by-regulators-in-Australia-and-New-Zealand

Science Friday, “Why Don’t Banana Candies Taste Like Real Bananas?” History of artificial banana flavor and isoamyl acetate. URL: https://www.sciencefriday.com/articles/why-dont-banana-candies-taste-like-real-bananas/

Ó Gráda, C. “The Lumper Potato and the Famine.” History Ireland 1(1). Lumper reliance concentrated among the poor; more varietal diversity than often assumed; all other varieties commonly sown also succumbed to the blight. URL: https://historyireland.com/the-lumper-potato-and-the-famine-11/

Teagasc, “Breeding for resistance,” TResearch, Winter 2020. The Lumper as Ireland’s main and especially susceptible potato variety; major resistance genes later bred in from the wild Mexican species Solanum demissum. URL: https://teagasc.ie/wp-content/uploads/media/website/publications/2020/TResearch_Winter2020_BreedingResistance_Web.pdf

“Decline in the Population of Ireland,” Thom’s Almanac 1862, via Library Ireland, and RTÉ, “The truth about Trevelyan,” 2020. Census population fall of 1.6 million between 1841 and 1851, emigration figures, and at least one million famine deaths. URLs: https://www.libraryireland.com/articles/DeclinePopulationThom1861/ · https://www.rte.ie/history/the-great-irish-famine/2020/0902/1162846-the-truth-about-trevelyan/

Food and Agriculture Organization of the United Nations, International Treaty on Plant Genetic Resources for Food and Agriculture: overview and the Multilateral System. In force since 29 June 2004; 64 crops and forages covered; mandatory monetary benefit-sharing triggered when a commercialized product restricts further research and breeding. URLs: https://www.fao.org/plant-treaty/overview/en/ · https://www.fao.org/plant-treaty/areas-of-work/the-multilateral-system/overview/en/ · https://www.fao.org/plant-treaty/areas-of-work/the-multilateral-system/faqs-mls/en/

Earth Negotiations Bulletin, Vol. 9 No. 880, Summary of the Eleventh Session of the Governing Body of the ITPGRFA, 24–29 November 2025. The twelve-year process to enhance the Multilateral System, including user-based payments, ended without agreement. URL: https://enb.iisd.org/sites/default/files/2025-12/enb09880e.pdf

ProMusa, “Tropical race 4 – TR4.” Fusarium wilt on Cavendish in Taiwan first observed in 1967; Cavendish plantations in Indonesia and Malaysia decimated in the early 1990s; TR4 in Australia’s Northern Territory by the end of the 20th century; Mozambique 2013; Colombia 2019. URL: https://promusa.org/Tropical+race+4+-+TR4

Hort Innovation, “New research to future-proof Australian bananas against Fusarium wilt,” 24 March 2026. Levy- and government-funded University of Queensland project (BA24004) on the genetic basis of TR4 resistance and breeding markers. URL: https://www.horticulture.com.au/news-and-events/latest-news/2026/new-research-to-future-proof-australian-bananas-against-fusarium-wilt

AgroFair, “Fusarium TR4: AgroFair supports the search for resistant banana varieties.” Four TR4-resistant CIRAD varieties planted in Peruvian trial fields in 2025 alongside Cavendish, first bunches in early 2026, evaluation through 2027. URL: https://www.agrofair.nl/portfolio-items/fusarium-tr4-agrofair-supports-the-search-for-resistant-banana-varieties/

CGIAR System Organization, Charter and governance pages. Integrated Partnership Board as governing body of the System Organization and the Integrated Partnership; each Center an independent legal entity with its own board. URL: https://www.cgiar.org/integrated-partnership-board

ADVERTISEMENT