{ "id": "coffee-wilt-disease", "url": "https://coffearium.org/entries/coffee-wilt-disease/", "license": "CC-BY-4.0", "title": "Coffee wilt disease", "summary": "A soil-borne fungal wilt that kills whole coffee trees in Africa. How Fusarium xylarioides blocks the water supply, why separate strains attack robusta and arabica, and how uprooting and resistant robusta clones answer it.", "domain": "farm-management", "kind": "Disease", "stage": "farm", "topics": [ "history", "science" ], "topicSections": {}, "lang": "en", "status": "verified", "reviewer": "Coffearium verification pass", "lastVerified": "2026-10-23T00:00:00.000Z", "updated": "2026-10-23T00:00:00.000Z", "added": "2026-10-23T00:00:00.000Z", "brief": [ "Coffee wilt disease (CWD), also called tracheomycosis, is a vascular wilt caused by the fungus Fusarium xylarioides. Infected trees wilt and die.5,1", "It was first observed in 1927 on Coffea excelsa near Bangui, Central African Republic, and destroyed plantations across West and Central Africa by the 1950s.1,5,6,15", "It re-emerged on robusta in the Democratic Republic of the Congo (DRC) in the 1970s and was confirmed in Uganda in 1993.1,6,21,10", "Two host-specific populations cause the current epidemics, one on robusta in Uganda, Tanzania and the DRC and one on arabica in Ethiopia.2,4", "Control rests on uprooting and burning diseased trees and on resistant robusta varieties; Uganda’s national research organisation has released seven.5,10,19" ], "quickFacts": [ { "label": "Pathogen", "value": "Fusarium xylarioides (sexual stage Gibberella xylarioides)" }, { "label": "Type", "value": "Vascular wilt (tracheomycosis); colonises the xylem" }, { "label": "First observed", "value": "1927, Coffea excelsa, near Bangui, Central African Republic" }, { "label": "Hosts today", "value": "Robusta (Uganda, Tanzania, DRC); arabica (Ethiopia)" }, { "label": "Known range (Oct 2026)", "value": "Present in Ethiopia, Uganda, DRC and Tanzania (EPPO, updated March 2024)" }, { "label": "Uganda", "value": "14.5 million trees destroyed since 1993 (reported 2003)" }, { "label": "Main control", "value": "Uprooting and burning; resistant robusta clones" } ], "aka": [ "CWD", "Tracheomycosis", "Fusarium xylarioides", "Gibberella xylarioides" ], "linkTerms": [ "coffee wilt disease", "coffee wilt", "tracheomycosis", "Fusarium xylarioides", "Gibberella xylarioides", "xylarioides" ], "related": [ "coffea-canephora", "liberica-and-excelsa", "ethiopian-landraces", "coffee-berry-disease", "coffee-leaf-rust", "coffea-arabica", "integrated-pest-management", "coffee-nematodes" ], "contested": [ "Survival in soil: a 2006 CABI project report states that soil from affected sites stayed infective to newly planted seedlings for many months8; a 2021 book chapter states that infectivity in soil lasted a few months and then declined.5 Both come from the same CABI-led research programme and give different durations; no peer-reviewed study consulted for this entry measured soil survival.", "Strictness of host specialisation: a 2006 cross-inoculation study found that an Ethiopian arabica strain killed arabica seedlings but caused no wilt on robusta lines, while robusta strains attacked robusta20; Ugandan host–pathogen studies cited in a 2021 chapter found host specialisation that was “not clear cut”.5", "Losses in Uganda: a 2003 paper reports 14.5 million trees destroyed since 199311; a 2022 study states that the disease wiped out about 60% of Uganda’s robusta trees in the early 1990s, an estimated loss of US$259 million19; a 2006 conference abstract gives a still higher tree count.21 The figures cover different periods and are not directly comparable.", "First report in Tanzania: a 2009 CABI synthesis and a 2006 conference abstract give different years for the first report in north-western Tanzania.6,21", "One species or several: a 2021 chapter reports a proposal to split the pathogen into four species, one for each outbreak5; the 2021 and 2024 genome studies treat the strains as populations and lineages of one species, F. xylarioides.2,1" ], "corrections": [ { "date": "2026-10-23T00:00:00.000Z", "change": "Dates resting only on CABI programme sources (1938–1945, 1939, 1945, 1947, 1958, programme 2000–2007, release of the resistant clones in 2010) were removed from the text and the timeline; the release of seven resistant varieties now rests on a peer-reviewed study without a year.", "reason": "Sources from the same CABI programme are not independent; a number resting only on them needs a Tier A source (editorial ruling)." }, { "date": "2026-10-23T00:00:00.000Z", "change": "The 2002–2003 regional survey percentages (Uganda and Ethiopia) and the chart built on them were removed; soil and wood survival times were given without numbers.", "reason": "They appear only in a conference paper, a project report and a chapter from the same CABI-coordinated programme." }, { "date": "2026-10-23T00:00:00.000Z", "change": "The Uganda figure of 4.8% was removed (the paper does not state its base), and the contested loss figure close to half of Uganda’s trees was replaced by the 2022 study’s statement of about 60% of robusta trees.", "reason": "The 2003 paper gives “4.8% of losses” without a base; the 2006 abstract gives 45%, not close to half, and carries no number alone." }, { "date": "2026-10-23T00:00:00.000Z", "change": "Chemical control, the 2005 isolate description and the historic lineages’ hosts were corrected to match the sources.", "reason": "Copper stem paints worked in the DRC and Uganda but not Ethiopia, separately from herbicides; the 36 isolates came from coffee and other woody hosts; the historic lineages were isolated from various Coffea species." } ], "sources": [ { "id": "s1", "title": "Horizontal transfers between fungal Fusarium species contributed to successive outbreaks of coffee wilt disease", "authors": "Peck, L. D.; Llewellyn, T.; Bennetot, B.; O’Donnell, S.; Nowell, R. W.; Ryan, M. J.; Flood, J.; Rodríguez de la Vega, R. C.; Ropars, J.; Giraud, T.; Spanu, P. D.; Barraclough, T. G.", "year": 2024, "publisher": "PLOS Biology 22(12), e3002480", "url": "https://storre.stir.ac.uk/retrieve/4df74757-d6c9-4145-8c92-dd0abdb66a64/Peck%202024%20PLoS%20Biol%20-%20Horizontal%20transfers%20fungal%20Fusarium%20species%20contributed%20to%20successive%20outbreaks%20of%20coffee%20wilt%20disease.pdf", "doi": "10.1371/journal.pbio.3002480", "tier": "A", "quote": "Yet by the 1970s, coffee wilt disease re-emerged on robusta coffee in central Africa", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s2", "title": "Historical genomics reveals the evolutionary mechanisms behind multiple outbreaks of the host-specific coffee wilt pathogen Fusarium xylarioides", "authors": "Peck, L. D.; Nowell, R. W.; Flood, J.; Ryan, M. J.; Barraclough, T. G.", "year": 2021, "publisher": "BMC Genomics 22, 404", "url": "https://link.springer.com/article/10.1186/s12864-021-07700-4", "doi": "10.1186/s12864-021-07700-4", "tier": "A", "quote": "the robusta population arose from the initial outbreak", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s3", "title": "Temperature contributes to host specialization of coffee wilt disease (Fusarium xylarioides) on arabica and robusta coffee crops", "authors": "Zhang, X.; Peck, L. D.; Flood, J.; Ryan, M. J.; Barraclough, T. G.", "year": 2023, "publisher": "Scientific Reports 13, 9327", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10250448/", "doi": "10.1038/s41598-023-36474-w", "tier": "A", "quote": "CWD first emerged in west and central Africa in the 1920s", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s4", "title": "Current knowledge of coffee wilt disease, a major constraint to coffee production in Africa", "authors": "Rutherford, M. A.", "year": 2006, "publisher": "Phytopathology 96(6), 663–666", "url": "https://www.apsnet.org/publications/phytopathology/2006/June/Pages/96_6_663.aspx", "tier": "A", "quote": "two genetically and biologically distinct forms are responsible for current problems", "quotes": [], "abstractOnly": true, "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s5", "title": "Coffee wilt disease", "authors": "Flood, J.", "year": 2021, "publisher": "Burleigh Dodds Science Publishing, Burleigh Dodds Series in Agricultural Science (open-access chapter)", "url": "https://library.oapen.org/bitstream/20.500.12657/61487/1/9781786768391_web.pdf", "doi": "10.19103/AS.2021.0096.25", "tier": "B", "tierNote": "Review chapter in an agricultural science book series; carries numbers only together with a second independent source or a Tier A source.", "quote": "CWD, also called Tracheomycosis, is a vascular wilt disease.", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s6", "title": "Coffee wilt: the problem and the project (ch. 1)", "authors": "Phiri, N.; Baker, P. (eds)", "year": 2009, "publisher": "In: Coffee Wilt Disease in Africa – A synthesis of the work of the Regional Coffee Wilt Programme 2000–2007, CABI, pp. 1–8", "url": "https://assets.publishing.service.gov.uk/media/57a08b41ed915d3cfd000c12/Coffee_CH01.pdf", "tier": "B", "tierNote": "Programme synthesis report of CABI, an intergovernmental organisation; chapter authors not stated in the copy consulted.", "quote": "This strategy proved highly successful in eradicating the disease from Côte d’Ivoire and DRC", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s7", "title": "Executive summary", "authors": "Phiri, N.; Baker, P. (eds)", "year": 2009, "publisher": "In: Coffee Wilt Disease in Africa – A synthesis of the work of the Regional Coffee Wilt Programme 2000–2007, CABI", "url": "https://assets.publishing.service.gov.uk/media/57a08b3fe5274a27b2000a51/Coffee_ExecSummary.pdf", "tier": "B", "tierNote": "Same report as source 6; the two parts count as one source for the number rule.", "quote": "A breeding programme in Uganda has screened thousands of Robusta plants for resistance to CWD.", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s8", "title": "Promotion of current knowledge on pests of coffee in East Africa. Final Technical Report, project R8513 (ZA0726)", "authors": "Rutherford, M.", "year": 2006, "publisher": "CABI Bioscience UK Centre, for the UK Department for International Development Crop Protection Programme", "url": "https://assets.publishing.service.gov.uk/media/57a08c19ed915d622c00110b/R8513CoffeeProjectFTR.pdf", "tier": "B", "tierNote": "Project report of CABI to a government funder, without stated review.", "quote": "90% and 30% of the farms in Uganda and Ethiopia, respectively, were found to be infected", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s9", "title": "Surveys to establish the spread of coffee wilt disease, Fusarium (Gibberella) xylarioides, in Africa", "authors": "Oduor, G. I.; Phiri, N.; Hakiza, G. J.; Abebe, M.; Asiimwe, T.; Kilambo, D. L.; Kalonji, M. A.; Pinard, F.; Simons, S.; Nyassé, S.; Kébé, I. S.", "year": 2005, "publisher": "20th International Conference on Coffee Science (ASIC), Bangalore, India, 2004, pp. 1252–1255", "url": "https://agritrop.cirad.fr/529389/", "tier": "B", "tierNote": "Conference proceedings without stated peer review.", "quote": "The highest incidence was in Uganda, where 90.3% of farms had CWD", "quotes": [], "abstractOnly": true, "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s10", "title": "Current research trends on coffee wilt disease tracheomycosis", "authors": "Hakiza, G. J.; Birikunzira, J. B.", "year": 2000, "publisher": "Uganda Journal of Agricultural Sciences 5(1), 17–21", "url": "https://journal.naro.go.ug/index.php/ujas/article/view/52", "tier": "A", "quote": "Disease incidence varies from a few infected trees", "quotes": [], "abstractOnly": true, "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s11", "title": "Evaluation of potential sources of inoculum for the coffee wilt epidemics in Uganda", "authors": "Kangire, A.; Olal, S.; Kabole, C.", "year": 2003, "publisher": "Uganda Journal of Agricultural Sciences 8, 37–40", "url": "https://journal.naro.go.ug/index.php/ujas/article/download/426/373/689", "tier": "A", "quote": "Cumulatively, this disease has destroyed a total of 14.5 million trees since it was first reported in 1993", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s12", "title": "Response of robusta coffee populations to Fusarium xylarioides infection under screen house condition in Uganda", "authors": "Musoli, P.; Olal, S.; Nabaggala, A.; Kabole, C.", "year": 2003, "publisher": "Uganda Journal of Agricultural Sciences 8, 31–36", "url": "https://journal.naro.go.ug/index.php/ujas/article/download/425/372/687", "tier": "A", "quote": "15.4% of the seedlings survived first round of infection.", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s13", "title": "Gibberella xylarioides (anamorph Fusarium xylarioides), a causative agent of coffee wilt disease in Africa, is a previously unrecognized member of the G. fujikuroi species complex", "authors": "Geiser, D. M.; Lewis Ivey, M. L.; Hakiza, G.; Juba, J. H.; Miller, S. A.", "year": 2005, "publisher": "Mycologia 97(1), 191–201", "url": "https://pure.psu.edu/en/publications/gibberella-xylarioides-anamorph-fusarium-xylarioides-a-causative-/", "doi": "10.1080/15572536.2006.11832853", "tier": "A", "quote": "These isolates fell into two morphologically and phylogenetically distinct groups.", "quotes": [], "abstractOnly": true, "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s14", "title": "An investigation into Fusarium spp. associated with coffee and banana plants as potential pathogens of robusta coffee", "authors": "Serani, S.; Taligoola, H. K.; Hakiza, G. J.", "year": 2007, "publisher": "African Journal of Ecology 45(s1), 91–95", "url": "https://agris.fao.org/search/en/records/65dfabfc4c5aef494fe43602", "doi": "10.1111/j.1365-2028.2007.00744.x", "tier": "A", "quote": "Also recovered from banana roots and pseudo-stem", "quotes": [], "abstractOnly": true, "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s15", "title": "A review of the indigenous coffee resources of Uganda and their potential for coffee sector sustainability and development", "authors": "Davis, A. P.; Kiwuka, C.; Faruk, A.; Mulumba, J.; Kalema, J.", "year": 2023, "publisher": "Frontiers in Plant Science 13, 1057317", "url": "https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2022.1057317/pdf", "doi": "10.3389/fpls.2022.1057317", "tier": "A", "quote": "later caused widespread damage to Liberica and robusta coffee across large areas of tropical Africa", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s16", "title": "Fusarium xylarioides (GIBBXY) – Distribution", "authors": "EPPO", "year": 2024, "publisher": "EPPO Global Database, European and Mediterranean Plant Protection Organization (distribution last updated 6 March 2024)", "url": "https://gd.eppo.int/taxon/GIBBXY/distribution", "tier": "A", "quote": "Present, widespread", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s17", "title": "Fusarium xylarioides (GIBBXY) – Categorization", "authors": "EPPO", "year": 2024, "publisher": "EPPO Global Database, European and Mediterranean Plant Protection Organization", "url": "https://gd.eppo.int/taxon/GIBBXY/categorization", "tier": "A", "quote": "A1 list", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s18", "title": "NARO-Kituza Robusta 6 (KR6)", "authors": "World Coffee Research", "year": 2026, "publisher": "World Coffee Research, Coffee Varieties Catalog", "url": "https://varieties.worldcoffeeresearch.org/varieties/kr-6", "tier": "A/B", "tierNote": "A for catalogue data (breeder, disease resistance, yield potential); B otherwise.", "quote": "Resistant to coffee wilt disease (CWD).", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s19", "title": "Cup quality profiles of Robusta coffee wilt disease resistant varieties grown in three agro-ecologies in Uganda", "authors": "Mulindwa, J.; Kaaya, A. N.; Muganga, L.; Paga, M.; Musoli, P.; Sseremba, G.; Wagoire, W. W.; Bitalo, D. N.", "year": 2022, "publisher": "Journal of the Science of Food and Agriculture 102, 1225–1232", "url": "https://nru.uncst.go.ug/bitstreams/e9cd1625-147c-4068-b00c-c3634a7491ee/download", "doi": "10.1002/jsfa.11460", "tier": "A", "quote": "variety KR4 was hardly influenced by environmental variations", "quotes": [], "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s20", "title": "Tracheomycosis (Gibberella xylarioides). A menace to world coffee production: evidenced by cross inoculation of historical and current strains of the pathogen", "authors": "Girma, A.; Flood, J.; Hindorf, H.; Bieysse, D.; Sarah, S.; Rutherford, M.", "year": 2007, "publisher": "21st International Scientific Colloquium on Coffee (ASIC), Montpellier, France, 2006, pp. 1268–1276", "url": "https://agritrop.cirad.fr/540119", "tier": "B", "tierNote": "Conference proceedings without stated peer review.", "quote": "strains isolated from Canephora attack C. canephora", "quotes": [], "abstractOnly": true, "accessed": "2026-10-23T00:00:00.000Z" }, { "id": "s21", "title": "Coffee Wilt Disease: a major constraint to coffee production in Africa", "authors": "Bieysse, D.; Lepoint, P.; Hakiza, G.; Kalonji, A.; Maraite, H.", "year": 2006, "publisher": "Ninth European Fusarium Seminar (CIRAD record)", "url": "https://nru.uncst.go.ug/handle/123456789/8718", "tier": "B", "tierNote": "Conference abstract; used only to show a disagreement.", "quote": "CWD was first observed on Coffea excelsa in Central Africa Republic (CAR) in 1927.", "quotes": [], "abstractOnly": true, "accessed": "2026-10-23T00:00:00.000Z" } ], "markdown": "## What coffee wilt disease is\n\nCoffee wilt disease (CWD) is a fungal disease that kills whole coffee trees. It is also called tracheomycosis, and it is a vascular wilt.5 The fungus, *Fusarium xylarioides*, colonises the xylem, the vessels that carry water from the roots to the leaves.1\n\nThe fungus enters mainly through the roots and the lower stem, helped by wounds.5 Inside, it blocks the uptake of water and breaks down cell walls for food.1 The tree wilts and dies.1,9\n\nAs of the EPPO record of March 2024, the disease is present only in Africa.16 It has attacked excelsa and liberica coffee, robusta (*Coffea canephora*) and, in Ethiopia, arabica (*C. arabica*).5,6,15 It attacks the whole plant through its water supply.1,5\n\n## Signs on the tree\n\nA wilting tree shows signs from the leaves down to the wood and roots.5\n\n- **Leaves.** Wilt and turn inwards, then fall; the branch dies back.5\n- **Berries.** Ripen prematurely.5,8\n- **One branch.** The disease may attack a single branch while the others stay healthy for a while.11 Symptoms then spread to the whole plant.5\n- **Bark.** The trunk can swell, with vertical or spiralling cracks in the bark.5\n- **Perithecia.** Small blackish-brown fruiting bodies of the sexual stage form in the cracks; dead stems can bear them in large numbers.5,11\n- **Wood.** Under the bark the wood shows a blue-black stain, a diagnostic sign according to a 2021 chapter.5\n- **Roots and lower stem.** The main entry points, often through wounds.5\n\nYoung plants can die within weeks of the first signs; mature trees take months.5,21 In 2003, Ugandan researchers called a therapeutic cure “still unknown”.11\n\n## How the fungus spreads\n\nThe fungus spreads between trees close together. Neighbouring trees often fall ill, which suggests contact between roots or spread over short distances.5 Spread did not appear to be linked to wind direction.5\n\nPeople move it too. In a trial reported by CABI, trees cut with a machete used on diseased trees developed the disease. Trees cut with a disinfected machete did not.8 Infected wood stays infective for a time, and so does soil after diseased trees are removed (see Contested).5,8\n\nThe fungus has been isolated from banana roots in Uganda.14,5 Leaves are not an entry point, at least in greenhouse tests.11\n\n## A century of epidemics\n\n
\n
\n
    \n
  1. 1927. First observed on Coffea excelsa near Bangui, Central African Republic.15,1,5,6
    2. \n
  2. 1940s–1950s. First wave: excelsa, liberica and robusta destroyed across West and Central Africa; controlled by the 1960s.1,3,15,21
    3. \n
  3. 1957. First observed on arabica, in Ethiopia.3,5,6
    4. \n
  4. Late 1970s. Re-emerges on robusta in north-eastern DRC.1,6,21
    5. \n
  5. 1993. Confirmed in Uganda, in Bundibugyo district near the DRC border.10
    6. \n
  6. 2021. Genome study of historical strains: the robusta population descends from the first outbreak.2
    7. \n
  7. 2024. Genome study links outbreaks to genes taken up from other Fusarium species.1
    8. \n
\n
Coffee wilt disease from the first observation to the genome studies, 1927–2024. Accent markers are disease events, hollow markers research. Dates to scale; the first wave is drawn as 1940–1960 and the late-1970s marker is placed at 1978; markers schematic.1,2,3,10,21
\n
\n\n**First wave.** The disease was first seen in 1927 on excelsa coffee in the Central African Republic.15,5,6 A 2009 CABI synthesis traces its spread to Cameroon, the DRC, Côte d’Ivoire and Guinea in the following decades.6 By the 1950s, coffee in West and Central Africa had been devastated, on excelsa, liberica and robusta.1,15 Producers uprooted and destroyed affected trees over large areas and replanted with resistant *C. canephora*. A 2009 CABI synthesis calls this highly successful in Côte d’Ivoire and the DRC.6 By the 1960s the disease was under control.3\n\n**Second wave.** From the 1970s the disease re-emerged on robusta in Central Africa.1,3 It was confirmed in Uganda in 1993, and by the end of 1998 it had spread to 18 districts.10 By the 1990s it was causing wide damage to arabica in Ethiopia, according to a 2021 genome study.2 A 2024 study describes its severity on arabica as low at first, with signs of an increase.1 As of 2021, two populations of the fungus were active: one on robusta in Uganda, Tanzania and the DRC, one on arabica in Ethiopia.2\n\n## Control: uprooting and resistant clones\n\nNo therapeutic cure was known in 2003, and chemical control is difficult.11,5 Control therefore removes the sources of the fungus and replaces susceptible trees.5,6\n\n- **Uprooting and burning.** Infected trees are uprooted and burned on the spot, stumps included.5,10\n- **Clean tools.** Machetes used on diseased trees should be cleaned, for example with a disinfectant or in a fire, according to the CABI report.8\n- **Moving material.** In Uganda, rules restricted the movement of unhulled coffee and banned coffee husks as mulch.10\n- **Replanting.** Uganda advised replanting on new land; when old sites can safely be replanted is not settled.10,8\n- **Resistant varieties.** Uganda screened its robusta for resistance, and its national research organisation released seven resistant varieties.19,7\n\nThe Regional Coffee Wilt Programme, whose synthesis CABI published in 2009, combined research, surveys and farmer training in several African countries.5,6 Integrated pest management, which combines such measures, has its own entry.\n\n\n\n## The pathogen and its names\n\n**Names.** The fungus was isolated from excelsa in the DRC by Steyaert and named *Fusarium xylarioides* from that work.5,6,13 The sexual stage, *Gibberella xylarioides*, was described by Heim and Saccas.5,15 EPPO uses *Fusarium xylarioides* as the preferred name.16 Some recent papers, such as a 2023 review of Ugandan coffee resources, still write *Gibberella xylarioides*.15\n\n**Species complex.** A 2005 DNA study sequenced 36 isolates previously identified as *F. xylarioides* or related fusaria from coffee and other woody hosts, and 12 isolates from the Ugandan outbreak; they fell into two distinct groups.13 The first belongs to the *Gibberella fujikuroi* species complex; the second contained four lineages related to *Fusarium lateritium*.13 The authors concluded that unrelated species are associated with coffee wilt, which matters for its epidemiology, causation and control.13\n\n**One species or four.** According to a 2021 chapter, a 2005–2006 proposal split the pathogen into four species: one for the Ethiopian arabica outbreak, one for the robusta epidemic in the DRC, Uganda and Tanzania, an asexual form for Guinea and *G. xylarioides* for the original excelsa disease.5 The genome studies of 2021 and 2024 treat these as populations of one species (see Contested).1,2\n\n**How the tree dies.** The fungus colonises the xylem. Wilting follows because the pathogen blocks water uptake, while it also degrades cell walls for nutrients.1 According to the 2021 chapter, colonisation of the vascular system triggers host responses that disrupt water conduction.5\n\n## Two host-specific populations\n\nThe current epidemics come from two populations that differ in host and place.2,4 A 2006 review describes two genetically and biologically distinct forms.4 A 2006 CABI report names them: one specific to robusta, the other to arabica.8\n\n| Population | Host | Where (as of 2021) | Source |\n|---|---|---|---|\n| Robusta | *C. canephora* (robusta) | Uganda, Tanzania, DRC | Peck et al. 20212 |\n| Arabica | *C. arabica* | Ethiopia | Peck et al. 20212 |\n| Historic (two lineages) | Various *Coffea* species | Earlier outbreaks | Peck et al. 20241 |\n\n**Cross-inoculation.** In a 2006 cross-inoculation study, an Ethiopian arabica strain killed seedlings of all arabica cultivars tested and caused no wilt on robusta lines.20 Strains from robusta were pathogenic to all robusta lines tested.20 A historical excelsa strain killed *C. liberica* lines and also affected arabica and robusta, to a lesser degree.20 Ugandan studies found host specialisation that was not clear cut (see Contested).5\n\n**Temperature.** A 2023 laboratory study grew strains at 10–40 °C.3 Robusta strains grew significantly faster than arabica strains at 20 and 25 °C.3 Arabica strains produced more spores at 15 °C and below than at warmer temperatures.3 The authors propose that temperature contributes to host specialisation, since arabica is grown higher and cooler than robusta.3\n\n## Genomes and borrowed genes\n\n**Historical strains.** Culture collections kept strains from each outbreak. Most strains in the 2024 study came from the CABI collection in Egham, United Kingdom.1 A 2021 study compared six strains collected over 52 years; a 2024 study compared 13 strains spanning six decades.2,1\n\n**Ancestry.** In the 2021 study, the robusta population arose from the initial outbreak, while the arabica population is a divergent sister clade.2 The 2024 study found at least four lineages: one specific to arabica, one to robusta and two historic lineages.1\n\n**Horizontal gene transfer.** Both studies found genes that *F. xylarioides* appears to have acquired from *Fusarium oxysporum*, a related wilt fungus with many host-specific forms.1,2\n\n- **2021.** Fifteen putative effector genes show evidence of horizontal acquisition. A cluster of four sits on a 20 kb scaffold resembling a mobile pathogenicity chromosome of *F. oxysporum*.2 DNA transposons nearby suggest transposition as the route, which the authors offer as a working hypothesis.2\n- **2024.** The populations differ in large genomic regions that closely match several *F. oxysporum* forms; these putative transferred regions range from 20 kb to 200 kb.1 All populations share transposable elements with *F. oxysporum*, including a Starship, a large mobile element that could have mediated the transfer.1\n- **In the plant.** Transferred effector genes were highly expressed during infection of arabica, together with enzymes that degrade plant cell walls.1\n\nThe authors conclude that horizontal transfer likely contributed to the repeated emergence of coffee wilt.1 In 2021 they suggested that this could inform intercropping, to limit effector transfer between closely related *Fusarium* taxa.2\n\n## Spread and survival\n\n**Entry.** Initial entry is considered to be mainly through roots and the lower stem, through wounds from slashing or livestock.5 In greenhouse tests in Uganda, the fungus did not enter through leaves. After stem inoculation, wilt appeared only above the inoculation point.11\n\n**Spores.** Ascospores form in the perithecia on the bark, and all spore types can carry the disease.5 Dry diseased stems produce perithecia in large numbers, especially in the wet season.11\n\n**Short-distance spread.** Infection of adjacent trees suggests root contact or short-distance dispersal, and transmission was not linked to wind direction.5 A 2003 Ugandan study names infected coffee stalks as a possible route.11\n\n**Tools, wood and soil.** In a CABI trial, trees wounded with a machete used on diseased trees developed the disease, and none wounded with a disinfected machete did.8 According to the same report, pieces of infected wood left on soil can also act as a source of infection for months.8 Survival in soil is reported differently (see Contested).5,8\n\n**Other hosts.** In Uganda, *F. xylarioides* was recovered from banana roots and pseudostems along with other *Fusarium* species.14 According to the 2021 chapter, weeds of the mallow and nightshade families in coffee fields could act as alternative hosts, which needs confirmation.5 Insects collected on affected farms did not carry the fungus.5\n\n## Surveys and losses\n\n**Regional survey.** A regional survey carried out within the programme covered coffee areas in Uganda, the DRC, Rwanda, Tanzania, Ethiopia, Cameroon and Côte d’Ivoire.9,5 Uganda had the highest incidence, with most of the surveyed farms affected.9,8,5 The disease was not found in Rwanda, Cameroon or Côte d’Ivoire, and in Tanzania only in the Kagera region bordering Uganda.9,5 The survey percentages appear only in reports of the CABI-coordinated programme itself, a conference paper, a project report and a book chapter, so no independent or peer-reviewed source confirms them; Coffearium does not give them.\n\n| Report | Loss or incidence stated | Source |\n|---|---|---|\n| Laboratory study, 2023 (introduction) | Up to 80% loss in production of plantations without treatment | Zhang et al.3 |\n| Uganda, 2003 | 14.5 million trees destroyed since 1993; 15 of 24 main robusta districts affected | Kangire et al.11 |\n| Uganda, early 1990s (stated 2022) | About 60% of robusta trees wiped out; estimated loss US$259 million | Mulindwa et al.19 |\n| West and Central Africa, by the 1950s | Coffee crops decimated | Peck et al. 20241 |\n| Uganda, 1998 | Spread from one district (1993) to 18 districts | Hakiza & Birikunzira10 |\n\nThese figures use different measures: counts of trees, share of trees or share of production. They refer to different places and decades and are not directly comparable.\n\n## Resistance screening and breeding\n\n**Wild robusta.** According to a 2023 review, wild robusta collected in the Kalangala and Itwara forests of Uganda has a high level of resistance.15\n\n**Screening.** Uganda screened seedlings in a screen house by dipping their roots in spore suspensions. In a 2003 study of 119 robusta cultivars from the national germplasm collection, plus survivors from wilt hot spots, 15.4% of the seedlings survived the first round of inoculation.12 Survivors were inoculated again, and those that survived were to be propagated vegetatively for field trials.12\n\n**Selection and release.** According to the programme synthesis, further screening and agronomic trials reduced the candidates to a short list for release to farmers.7 A 2022 peer-reviewed study states that the effort resulted in the release of seven resistant varieties, named Kituza Robusta (KR).19 The release year is given only in reports of the CABI-coordinated programme, so it is not stated here.\n\n**Named varieties.** World Coffee Research lists NARO-Kituza Robusta 6 (KR6), bred by Uganda’s National Coffee Research Institute (NaCORI), as resistant to coffee wilt disease.18 Its yield potential is given as 2,650 kg green coffee per hectare.18 KR8, KR9 and KR10 are listed as resistant too.18\n\n**Ethiopia.** Ethiopian breeders select wilt-tolerant arabica from local landraces, as the entry on Ethiopian landraces describes.\n\n## Chemical control\n\nAccording to the 2021 chapter, chemical control of this pathogen is difficult.5 Copper stem paints worked in the DRC and Uganda but not in Ethiopia; herbicide sprays showed promise but were too expensive for many producers.5 The chapter states that wide-scale fungicide use is not advocated, because of the cost to smallholders and environmental concerns.5 Coffearium gives no product or dose advice.\n\n## Distribution and quarantine\n\n**Distribution as of October 2026.** The EPPO table (last updated 6 March 2024) gives:16\n\n| Status in the EPPO database | Countries | Source |\n|---|---|---|\n| Present, widespread | Ethiopia, Uganda | EPPO16 |\n| Present, restricted distribution | Democratic Republic of the Congo, Tanzania | EPPO16 |\n| Absent, pest no longer present | Angola, Burkina Faso, Cameroon, Central African Republic, Congo, Côte d’Ivoire, Eswatini, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Niger, Nigeria, Senegal, South Africa, Sudan, Togo, Zimbabwe | EPPO16 |\n| Absent, invalid record | Kenya | EPPO16 |\n| Absent, unreliable record | Rwanda, El Salvador | EPPO16 |\n\n**Listings.** EPPO records the fungus on the A1 list of the Asia and Pacific Plant Protection Commission (APPPC) since 1988, and on the A1 list of Brazil since 2018.17 The 2006 cross-inoculation study called for strong quarantine policies in all coffee-producing countries.20\n\n## What it means for the cup and the buyer\n\nCoffee wilt disease kills trees, so its effect on supply runs through lost trees and replanting.5,11\n\nResistant clones raised a question of cup quality. A 2022 study cupped the five most commonly grown resistant robusta varieties from three agro-ecological zones in Uganda.19 Most scored 70–79, a premium grade; KR4 scored 80 or more, a specialty grade, and its bean size varied little between sites.19 Blends scored 76–82.19 The entry on *Coffea canephora* covers robusta itself.\n\n## Reference values\n\n| Parameter | Value reported | Source |\n|---|---|---|\n| Pathogen | *Fusarium xylarioides* (sexual stage *Gibberella xylarioides*) | EPPO16; Flood5 |\n| Disease type | Vascular wilt (tracheomycosis); xylem colonised | Flood5; Peck et al. 20241 |\n| First observation | 1927, *C. excelsa*, near Bangui, Central African Republic | Flood5; Phiri & Baker6; Davis et al.15 |\n| First record on arabica | 1957, Keffa, Ethiopia | Zhang et al.3; Flood5 |\n| Re-emergence | 1970s, robusta, Central Africa | Peck et al. 20241; Zhang et al.3 |\n| Uganda, first confirmed | 1993, Bundibugyo district | Hakiza & Birikunzira10 |\n| Entry | Roots and lower stem, through wounds; not through leaves (greenhouse) | Flood5; Kangire et al.11 |\n| Growth temperature (laboratory) | Robusta strains faster at 20 and 25 °C; arabica strains sporulate more at ≤15 °C | Zhang et al.3 |\n| Transferred regions | 20–200 kb, related to *F. oxysporum* | Peck et al. 20241 |\n| Putative effector genes acquired | 15 | Peck et al. 20212 |\n| Uganda losses | 14.5 million trees (1993–2003) | Kangire et al.11 |\n| Screen-house survival, first inoculation | 15.4% of seedlings | Musoli et al.12 |\n| Resistant varieties released (Uganda) | 7 | Mulindwa et al.19 |\n| Range (EPPO, 2024) | Present in Ethiopia, Uganda, DRC, Tanzania | EPPO16 |\n\n## Open questions\n\n- **Soil survival.** How long the fungus stays infective in soil, and so when a site can be replanted, is not settled.5,8\n- **Host range.** How strict the split between robusta and arabica strains is, and whether bananas or weeds act as reservoirs in the field, needs more work.5,14\n- **Species limits.** Whether the outbreaks are one species or several is answered differently in the sources (see Contested).1,2,5\n- **Durability of resistance.** No study consulted for this entry tested whether the Ugandan resistant clones stay resistant over decades or against strains from other countries.\n- **Excelsa today.** As of 2023, no coffee wilt disease had been reported on excelsa in Uganda, but its level of resistance needs research, as the entry on liberica and excelsa notes.15\n\n## Key studies\n\n
\n

Peck et al., 2024 – Borrowed genes and successive outbreaks. Compared the genomes of 13 historical strains spanning six decades and found four lineages, two of them host-specific. Large regions and effector genes shared with Fusarium oxysporum, and a Starship mobile element, point to horizontal gene transfer as a driver of new outbreaks.

\n

PLOS Biology 22(12), e3002480 · doi:10.1371/journal.pbio.3002480

\n
\n\n
\n

Peck et al., 2021 – Historical genomics of the wilt pathogen. Sequenced six culture-collection strains spanning 52 years. The robusta population descends from the first outbreak, the arabica population is a sister clade, and 15 putative effector genes appear to come from F. oxysporum.

\n

BMC Genomics 22, 404 · doi:10.1186/s12864-021-07700-4

\n
\n\n
\n

Zhang et al., 2023 – Temperature and host specialisation. Grew arabica and robusta strains at 10–40 °C. Robusta strains grew faster at 20 and 25 °C, while arabica strains sporulated more in the cold, consistent with the cooler highlands where arabica grows.

\n

Scientific Reports 13, 9327 · doi:10.1038/s41598-023-36474-w

\n
\n\n
\n

Kangire et al., 2003 – Sources of inoculum in Uganda. Tested how the fungus enters and spreads in robusta. It did not enter through leaves in greenhouse tests, wilt stayed above the point of stem inoculation, and dry diseased stems carried many perithecia; by then 14.5 million trees had been lost.

\n

Uganda Journal of Agricultural Sciences 8, 37–40 · article

\n
\n\n
\n

Geiser et al., 2005 – More than one fungus. Sequenced parts of two genes in 48 isolates from coffee and other woody hosts, 12 of them from the Ugandan outbreak. They fell into two unrelated groups, one in the Gibberella fujikuroi species complex and one near Fusarium lateritium, with consequences for epidemiology, causation and control.

\n

Mycologia 97(1), 191–201 · doi:10.1080/15572536.2006.11832853

\n
" }