How the major diseases and pests of cacao spread, what they cost where that was measured, and what has been tried against them. Most management trials are small and local, and several control methods have been measured by insects caught or disease reduced without anyone measuring the crop saved.
38 studies1981 to 2024Each checked against its published abstract
What these studies add up to
The big diseases are regional and have moved by dated steps: witches' broom into Bahia in 1989 by two separate introductions, the aggressive black pod species across Ghana from 1985, frosty pod rot as two clonal genotypes.
Swollen shoot is not one virus. Sequencing has found up to ten species, the Nigerian one unique to Nigeria, and no single test detects them all.
Losses, where measured, are large and local: 71% of pods in untreated plots in one wet Cameroonian year; every seedling infected and over half dead in one planting with vascular-streak dieback.
Fungicide works against black pod. Biocontrol fungi sit between fungicide and nothing, and gave incomplete protection against frosty pod rot.
Removing diseased pods or shoots gave different results in three places: more yield in Peru, less disease without clearly more production in Cameroon, and low production in Bahia.
Resistance has failed before. The main source against witches' broom was overcome within years, and resistance to vascular-streak dieback rarely came with good yield.
Replanting after swollen shoot clearance often puts new cocoa beside infected old cocoa; a belt of another crop almost stopped the virus in one trial.
In one young planting every seedling caught vascular-streak dieback and over half died
How does vascular-streak dieback spread and how fast does it build up in young cocoa?
How
Inoculation studies and field observations of spore dispersal and disease increase in young plantations. The abstract does not name the country or give plot sizes.
Finding
Disease increase in young plantations occurred at a rate of 0.31 per unit per month; in one block 100% of seedlings were infected and 55% were killed within 33 months of planting. The disease was most damaging in seedlings younger than 18 months. Sporophore production and infection rate were correlated with rainfall.
Does not establish: It does not quantify yield loss in mature cocoa; the abstract says only that there was evidence of substantial loss.
Limits: The worst figure comes from one block of susceptible seedlings, decades ago.
What quarantine measures prevent vascular-streak dieback moving in cocoa planting material?
How
Quarantine practice and laboratory examination in Papua New Guinea: isolation of clonal material in screened shade houses, microscopy of budwood and pod stalks, and germination of seed from diseased branches. No sample numbers in the abstract.
Finding
Isolation of clonal material in screened shade houses for 6 months allowed adequate opportunity to detect the disease. Hyphae were found in the stalks and placentae of pods from diseased branches, but seeds from them germinated normally and there was no evidence of seed transmission.
Does not establish: It does not prove seed can never carry the fungus, only that no transmission was seen.
Limits: One country's quarantine experience in the 1980s, without sample sizes in the abstract.
At which stage of pod development does feeding by Helopeltis theivora cause crop loss?
How
Field study in Peninsular Malaysia relating feeding lesions to survival across five post-flowering developmental stages, with artificially imposed attacks and a life table from flower bud to ripe pod. No sample sizes in the abstract.
Finding
Cherelle initiates invariably abscised after only one feeding lesion. Yield was most affected by death of cherelles; older cherelles tolerated more damage. The yield of full-sized pods was unaffected by artificially imposed very severe attack.
Does not establish: It gives no percentage crop loss and does not cover the African mirid species.
Limits: One species in one country, partly with artificial infestations.
What was done to eradicate and then contain the first recorded outbreak of witches' broom in Bahia, and why did it fail?
How
Descriptive account of the eradication attempt during 1989 and the later containment efforts. The abstract gives no data.
Finding
The paper describes the attempts during 1989 to eradicate the first recorded outbreak in Bahia and the subsequent efforts to contain spread, and evaluates the factors that worked against successful eradication.
Does not establish: The abstract gives no numbers for area treated, trees destroyed or production lost, so none can be drawn from it.
Limits: A narrative account by people involved, with no figures in the abstract.
Does more frequent removal of diseased pods pay for itself where moniliasis, witches' broom and black pod occur together?
How
Comparison of weekly and fortnightly phytosanitary pod removal in three fields in eastern Peru. The abstract gives no plot sizes or duration.
Finding
Weekly removal decreased moniliasis by 26-41%, black pod by 35-66% and witches' broom on pods by 14-57% compared with fortnightly removal. Yield rose from an average of 504 to 660 kg per hectare per year, and weekly removal was 32% more profitable.
Does not establish: It does not show that pod removal controls these diseases on its own, or that the economics transfer to other labour markets.
Limits: Three fields in one region, with labour costs of the 1990s. The comparison is between two removal frequencies, with no untreated control described.
Do isolates of the witches' broom fungus from Bahia come from one introduction or more?
How
RAPD analysis of 46 isolates of Crinipellis perniciosa from cocoa in Bahia and the Brazilian Amazon, using 258 RAPD loci.
Finding
The disease was first reported on cocoa in Bahia in 1989. The Bahian isolates separated into two major groups that differed in their similarity to Amazonian isolates, and their geographic distribution suggests two independent focal points of introduction.
Does not establish: It does not establish who or what brought the fungus to Bahia, or whether the introductions were accidental or deliberate.
Limits: 46 isolates and a marker type (RAPD) that later methods have superseded. The inference of two introductions rests on clustering and geography.
What does removing diseased pods do to black pod progress and to cocoa production under Phytophthora megakarya?
How
Field experiments over two successive years in two smallholders' plots in Cameroon, with and without phytosanitary pod removal, in a randomised complete block design.
Finding
Removing diseased pods reduced the black pod rate by 22% and 31% at the two sites in the first year and by 9% and 11% in the second year. Total production was higher with pod removal but the difference was not significant.
Does not establish: It does not show that sanitation alone protects a crop: the authors say it would need to be combined with other methods.
Limits: Two plots over two years, with a smaller effect in the second year. No fungicide comparison in the same trial.
Which chemical and biological agents, applied how, improve healthy pod yield under frosty pod rot?
How
Two field trials at La Lola, Costa Rica, comparing fungicides and hyperparasitic fungi applied with motorised mistblowers and hydraulic sprayers. The abstract gives no plot or tree numbers.
Finding
Copper hydroxide as prophylactic applications at 1500 g a.i. per hectare showed the most consistent but incomplete improvement in healthy pod yield. Isolates of Clonostachys byssicola and Trichoderma asperellum and two triazole fungicides made no significant improvement. The authors estimate copper may be cost effective when farm-gate prices exceed about $1.25 per kg.
Does not establish: It does not show that biological control fails in general; a later trial with a different formulation reported a yield gain.
Limits: One research station, two trials, and no yield figures in the abstract. The price threshold is from the cocoa economy of the time.
Is cacao swollen shoot virus present in Litimé, Togo's main cocoa-producing area?
How
Disease note: symptomatic leaves collected during 2000 in Litimé, Togo, confirmed by immunosorbent electron microscopy and PCR, with one isolate fully sequenced.
Finding
Swollen shoot foci were detected for the first time in Litimé at the end of the 1990s; the disease then spread rapidly and approximately 60% of the plots in Litimé contain infected trees. The Litimé isolate shared only 77% nucleotide identity with the earlier Togolese isolate. The note dates the disease to 1922 in Ghana and 1955 in Togo, in Kloto.
Does not establish: The 1922 date for Ghana differs from the 1936 date given by Ghanaian authors and by EPPO, so the note does not settle when the disease was first seen.
Limits: A short disease note; the abstract does not say how the 60% figure was surveyed.
How genetically variable is Moniliophthora roreri across its range in tropical America, and where is its diversity greatest?
How
AFLP and ISSR profiles of 94 isolates of M. roreri from across its range in Central and South America, with sequence data from the nuclear rDNA internal transcribed spacer region.
Finding
The highest genetic diversity was in Colombia, not Ecuador as previously believed. The fungus fell into five genetic groups; two have a wide range and the others appear endemic to Colombia or north-western Ecuador. The authors speculate that central or north-eastern Colombia is the centre of origin.
Does not establish: It does not show when or by what route the fungus reached each country, and human-mediated spread is discussed, not measured.
Limits: The origin is stated as speculation from diversity patterns. The abstract gives no dates for the spread into any country.
Can sanitation plus fungicide directed at the trunk control black pod on pods higher in the canopy?
How
Field studies in two cocoa-growing areas of Ghana, one affected only by Phytophthora palmivora and one predominantly by P. megakarya. Sanitation was followed by systemic, semi-systemic or contact fungicides on the trunk. No plot numbers in the abstract.
Finding
Sanitation combined with fungicide application on the trunk significantly reduced black pod incidence in the canopy. Ridomil 72 plus at 3.3 g per litre and Kocide DF at 10 g per litre as sprays, and 40 ml of Foli-R-Fos 400 injected twice a year, performed better than the other treatments. Pod position mattered more in the P. megakarya area.
Does not establish: It does not separate the effect of sanitation from that of fungicide.
Limits: The abstract gives no incidence figures or yields, and sanitation was not tested without fungicide.
What is known about the cause, spread and management of vascular-streak dieback of cacao?
How
Review by two researchers of the disease. Not a systematic review; no new data.
Finding
The fungus was identified as the cause of the disease in Papua New Guinea in the 1960s and the disease now occurs throughout Southeast Asia and parts of Melanesia. Spores remain viable for a few hours, so very few new infections occur beyond 80-m from diseased trees. Transmission through seed or infected cuttings has not been demonstrated.
Does not establish: It gives no loss figures, and the fungus's original wild host is described as unidentified.
Limits: A narrative review reflecting knowledge to 2007; later work renamed the fungus and reported new symptoms.
Where are populations of the cacao wilt fungus genetically diverse, and where do they look like recent introductions?
How
Microsatellite markers and nuclear and mitochondrial DNA fingerprints compared among isolates of Ceratocystis cacaofunesta from western Ecuador, Costa Rica, Colombia, and Rondônia and Bahia in Brazil. Isolate numbers are not in the abstract.
Finding
All marker types showed limited variation in the Colombian, Costa Rican and Bahian populations, as expected for introduced populations that passed through recent bottlenecks. The Rondonian and western Ecuadorian populations had diversity similar to natural populations. The putatively introduced populations were closer to the Rondonian one.
Does not establish: It does not date the arrival of the disease in Bahia or show how it travelled.
Limits: Introduction is inferred from low diversity; no dates or routes are given.
How do rainfall, a chemical fungicide and a Trichoderma biocontrol agent affect black pod caused by Phytophthora megakarya?
How
Three consecutive years in one smallholder's plantation in Cameroon, with weekly observations on trees sprayed with the fungicide Ridomil, with Trichoderma asperellum strain PR11, or left untreated.
Finding
Disease rates were 1.73%, 47.1% and 71.23% in the fungicide, PR11 and untreated plots in 2004, and 0.67%, 11.35% and 34.04% in 2005. Pod rot incidence correlated positively with rainfall after a 1-week lag, and immature pods were the most susceptible. The authors conclude the biocontrol was promising but not as effective as chemical methods under high disease pressure.
Does not establish: The untreated rates are from one farm in two years and are not a national loss estimate.
Limits: A single plantation, so plots are not replicated across sites. Disease pressure differed sharply between the two years reported.
How can cacao genotypes be reliably screened for resistance to Ceratocystis cacaofunesta?
How
Inoculation of cut stems of seedlings and of ten cacao genotypes propagated by cuttings, in Brazil, scoring incidence, mortality, a disease index and lesion heights.
Finding
TSH-1188 was confirmed as the reference genotype for resistance and ICS-1 as the susceptible standard. Of the clones, CEPEC-2008 was resistant, CEPEC-2002 and CEPEC-2007 moderately resistant, and CEPEC-2009, CCN-10, CCN-51, HW-25, PH-16 and SJ-02 susceptible.
Does not establish: It does not measure field losses from the disease.
Limits: Young plants inoculated through wounds, which may not match infection of mature trees in the field.
Which combination of biocontrol, fungicide and broom removal best manages witches' broom in Bahia?
How
Three-year field study in Bahia, Brazil, with Trichoderma stromaticum, copper hydroxide and phytosanitary broom removal applied alone or in combination, against untreated controls. No plot numbers in the abstract.
Finding
Alternating fungicide with biocontrol gave higher pod yields and consistently lower pod losses than untreated controls. Broom removal reduced pod losses but total pod production per tree was consistently low whenever it was used. The biocontrol fungus alone did not reduce disease on pods, though it reduced vegetative broom formation.
Does not establish: It does not show that sanitation pruning is harmful elsewhere; a Peruvian trial of pod removal reported a yield gain.
Limits: One region and one three-year trial; the abstract reports directions of effect without figures.
Do wild cacao populations from the Brazilian Amazon carry resistance to witches' broom distinct from Scavina 6?
How
43 families from natural cacao populations of the Brazilian Amazon evaluated under nursery or field conditions by artificial or natural infection, with field evaluation over a nine-year period.
Finding
The abstract states that resistance from the most widely used source, Scavina 6, has been overcome after a few years of deployment, and the trial showed its natural field infection increasing. Accessions from several river basins had the least disease, the two most outstanding being CAB 0208 and CAB 0214 from the Purus river basin.
Does not establish: It does not show that the new sources will last; the same abstract records how the previous main source failed.
Limits: Resistance is judged by disease incidence in one programme; yields and bean quality of the resistant accessions are not reported in the abstract.
How likely is cocoa pod borer to reach other cocoa islands of Papua New Guinea, and how long should surveillance continue after an eradication?
How
Monte Carlo simulation of dispersal from East New Britain Province to Bougainville and New Ireland under different scenarios, and a contingency model of sampling effort after an incursion.
Finding
The pest emerged in East New Britain, Papua New Guinea, in 2006. The model suggested it is far more likely to establish on New Ireland than on Bougainville. At a 1% infestation level, sampling should continue for 2.5-2.7 years after eradication is claimed.
Does not establish: It does not record whether the pest reached either island.
Limits: Scenario modelling: the authors say it identifies trends and that uncertainty about cocoa movements between islands strongly affected the outputs.
Does damage by the mirid Helopeltis sulawesi change attack by the cocoa pod borer and the resulting yield?
How
Two-year survey of herbivore damage and yield on 10 focal trees in each of 43 cacao plantations in Indonesia, plus a two-choice oviposition experiment.
Finding
The proportion of pods affected by both pests was significantly lower than expected, because egg-laying pod borer females avoided pods damaged by the mirid. Marketable yield was highest at a mirid incidence of 51%.
Does not establish: It does not show that mirids in general are harmless; it concerns one mirid species in Indonesia.
Limits: One region and one pair of pest species. The yield optimum is a fitted value from survey data.
Can simple field scores of cocoa pod borer infestation be turned into estimates of yield loss?
How
Models built from infestation and yield data in mixed-variety plantations in South Sulawesi, Indonesia, for six yield variables, and validated against an independent dataset from the same province. The number of pods is not in the abstract.
Finding
Models using a four-point severity grade performed markedly better than presence or absence models. The best fits were for dry weight per pod and pod value, with 62% and 69% of the variance accounted for.
Does not establish: The abstract gives no figure for how much yield the pest takes, so the paper cannot be cited for a percentage loss.
Limits: Models from one province; explanatory power was poor for some variables and validation showed slight under-estimation.
Can SNP markers from RNA sequencing reveal the genetic diversity and origin of Moniliophthora roreri?
How
SNP markers developed from RNASeq of 13 isolates; 88 polymorphic SNPs then used to genotype 172 M. roreri isolates from cacao in an international collection.
Finding
The 172 isolates resolved into 37 distinct genotypes. No heterozygosity was found, indicating clonal reproduction. The upper Magdalena Valley of Colombia had the highest diversity, with 20 genotypes of which 13 occurred only there, and is given as the possible centre of origin.
Does not establish: The statement that the disease causes yield losses up to 80% is background in the abstract, not something this study measured.
Limits: Diversity in a culture collection reflects where isolates were collected. A centre of diversity is inferred to be a centre of origin, which the abstract words as possible.
Does formulation change how well Trichoderma isolates manage frosty pod rot in the field?
How
Field study between May 2009 and April 2011: two Trichoderma isolates in three formulations applied monthly at 180 mL per tree, with water and copper oxychloride controls. The abstract does not give the site or the number of trees.
Finding
Averaged over the 2 years, the invert corn oil formulation increased yield to 30·7% healthy pods compared with 9·7% healthy pods in the water control. The oil formulation of isolate DIS-70a gave the largest yield increase of all treatments, including the fungicide control.
Does not establish: It does not show that a commercial product exists or that the result holds at farm scale; the authors call it a promising model.
Limits: Even the best treatment left roughly seven pods in ten unhealthy. Site and replication are not in the abstract.
How far has Phytophthora megakarya spread in Ghana, and do other plants on cacao farms harbour it?
How
Annual surveys of cocoa-growing districts of Ghana from 1985 to 2012; in July to October 2008, isolations from roots of economic plants and from cacao at 139 locations in 22 districts, giving 1459 Phytophthora isolates from 22 plant species.
Finding
P. megakarya had spread from Akomadan and Bechem in 1985 to 50 more administrative districts in 2012. Only two species were found, P. palmivora and P. megakarya. P. megakarya was isolated for the first time from nine other plants on cacao farms, including cocoyam, plantain, oil palm, avocado, papaya, mango and pineapple.
Does not establish: It does not measure pod losses, and it does not show how the pathogen moved between districts.
Limits: Species were identified by sporangium morphology. Isolation from symptomless roots shows presence, not how much those plants contribute to epidemics.
How diverse is cacao swollen shoot virus in Ghana, and where did it come from?
How
PCR amplification and phylogenetic analysis of infected cacao and alternative host plants collected from the cacao-growing regions of Ghana. No sample size in the abstract.
Finding
Nine CSSV groups were found in Ghana, six of them detected there for the first time, corresponding to at least five putative species. One species was detected only in the alternative host Ceiba pentandra. The abstract states that the disease was first observed in Ghana in 1936 and later discovered in Nigeria, Côte d'Ivoire, Togo and Sierra Leone.
Does not establish: It does not identify an origin: the authors say the spatial distribution makes a single origin difficult to predict.
Limits: No sample numbers in the abstract, and group counts depend on the genome region used.
Can crops immune to swollen shoot virus, planted as a barrier, stop it spreading from an outbreak into new cacao?
How
Field trial at the Cocoa Research Institute of Ghana from 1992 to 2004: citrus, oil palm, kola or cacao planted as a barrier between cacao infected with CSSV strain 1A and 522 test trees per treatment, followed for 7 years after inoculation.
Finding
Visibly infected test trees numbered 1 of 522 behind citrus, 5 of 522 behind oil palm, 89 of 522 behind kola and 134 of 522 behind a cacao barrier. Economic analysis found the barrier crops as profitable as growing cacao.
Does not establish: It does not show how wide a barrier a farmer needs or whether farmers will give land to it.
Limits: One research-station trial with one virus strain. Counts are of visibly infected trees, and infection can be latent.
Do mild strains of swollen shoot virus protect cocoa against severe strains over the long term?
How
Field trials at the Cocoa Research Institute of Ghana with trees inoculated with mild strains N1 or SS365B and uninoculated trees, with growth, mortality and yield analysed between 1996 and 2014. Tree numbers are not in the abstract.
Finding
Trees inoculated with mild strains performed significantly better in the presence of severe strain 1A than uninoculated cocoa, but degenerated over time. Spread of the severe virus eroded the effect, and the protection conferred by the mild strains declined over time.
Does not establish: It does not show cross-protection to be a durable control measure; the trial shows it wearing off.
Limits: Research-station trials; no effect sizes in the abstract.
How does temperature affect growth and survival of the two main black pod pathogens?
How
Laboratory study of pod lesion expansion, mycelial growth, gene expression and metabolite profiles of P. megakarya and P. palmivora at a range of temperatures. Number of isolates not given in the abstract.
Finding
P. palmivora has a higher temperature maximum (34 °C) than P. megakarya (30 °C). In culture P. megakarya lost viability after 3 days at 32 °C, while P. palmivora viability began decreasing after 3 days at 36 °C. The authors judge that elevated seasonal field temperatures in Africa are unlikely to persist long enough to eliminate P. megakarya.
Does not establish: It does not predict that warming will remove P. megakarya from West Africa; the authors judge elimination unlikely.
Limits: Culture and detached-pod work, not field epidemics. The number of isolates per species is not stated.
How many distinct badnavirus species are associated with cacao swollen shoot disease?
How
Illumina sequencing of cacao samples from the Cocoa Research Institute of Ghana museum collection and field samples from Côte d'Ivoire and Ghana; 20 new complete genomes reconstructed. Species were separated at a 20% threshold of nucleotide divergence in the RT/RNase H region.
Finding
The authors conclude there are seven new species associated with the disease, which with the three already described makes ten. A sample from Sri Lanka with similar leaf symptoms held a further new virus, named cacao bacilliform SriLanka virus.
Does not establish: It does not show that the species differ in the damage they cause or in how they spread.
Limits: Species counts depend on the demarcation threshold and on which samples were sequenced. The title says more than ten, the abstract text says ten.
Does shade from agroforestry trees change the severity of swollen shoot symptoms and cocoa yield?
How
Observational study along a gradient of shade tree abundance on smallholder cocoa farms in Ghana, measuring symptom severity, capsid damage, yield, light and soil fertility. The number of plots is not in the abstract.
Finding
Symptom severity and yield both followed quadratic curves, lowest and highest in plots with 54% and 39% shade respectively. Simulated optimal shade levels for the two overlapped between 45% and 53%. The abstract states that the only effective treatment for infected cocoa is to cut and replant.
Does not establish: It does not show that shade prevents infection or slows spread between trees.
Limits: Observational, so shade may stand in for other farm differences. Severity of symptoms was measured, not infection or spread.
Which local and hybrid cacao clones in Sulawesi resist vascular-streak dieback, and how heritable is the resistance?
How
Participatory trials in three provinces of Sulawesi, Indonesia: 2.5-year-old clonally propagated trees evaluated for disease severity from 2010 to 2012 and re-evaluated in 2014.
Finding
Resistance rankings were consistent across trials and broad-sense heritability was estimated as 0.67–0.92. Two progeny clones, KW617 and ICCRI03, were as resistant as their resistant parents, PBC123 and Scavina 6. Resistance in farm selections did not usually co-exist with yield and bean quality.
Does not establish: It does not show that a clone combining resistance, yield and quality is available to farmers; the authors say new crosses are required.
Limits: Severity scores under natural infection on young trees over a few years.
A mirid mass-trapping trial found too little damage to say whether trapping worked
Sarfo and colleagues, 2018. Entomologia Experimentalis et Applicata. doi:10.1111/eea.12699
Question
What density of pheromone traps is best for mass trapping cocoa mirids?
How
Replicated field experiment on mature cacao in Ghana: pheromone traps deployed for 3 months at nine densities from 20 to 300 traps per hectare plus an untreated control, with mirid populations assessed by insecticide knockdown on 400 trees and searches on 1 200 trees.
Finding
Traps caught 781 Sahlbergella singularis and 235 Distantiella theobroma. The optimal density for S. singularis was 150 traps per hectare. Population density was estimated at 220–230 mirids per hectare, 63% of them D. theobroma. Too few pods and shoots were damaged to establish whether trapping controlled mirids.
Does not establish: Efficacy against damage was not established, as the abstract itself states. The abstract also says insecticide spraying has been the principal control since 1954.
Limits: One three-month trial at low pest density.
Which virus is associated with swollen shoot symptoms on cacao in Nigeria?
How
18 leaf samples collected in 2017 from symptomatic cacao trees in Osun and Oyo, Nigeria; PCR amplification and sequencing of nine complete genomes.
Finding
The nine genomes shared 86 to 99% nucleotide identity with each other but only 71 to 75% with 21 previously reported swollen shoot badnaviruses, and are named as a new species, Cacao red vein-banding virus. Primers designed from it amplified a fragment from 16 of 18 field samples.
Does not establish: It does not show how widespread this species is in Nigeria or whether it occurs in neighbouring countries.
Nine mealybug species were found on cocoa in virus-affected parts of Côte d'Ivoire
N'Guessan and colleagues, 2019. Journal of Applied Entomology. doi:10.1111/jen.12707
Question
Which mealybug species occur on cocoa in Côte d'Ivoire and how abundant are they on young and mature trees?
How
Surveys from 2013 to 2018 of immature and mature cocoa trees, inspected to hand height, in 5 and 29 counties infected with Cacao swollen shoot virus.
Finding
Five species were identified on immature trees and four more on mature trees. On mature trees Formicococcus njalensis made up 63.2% of colonies and Planococcus citri 21.0%. Pods were the preferred organ, with 74.1% of colonies.
Does not establish: It does not show which of these species transmit the virus or how efficiently.
Limits: Trees were inspected only to hand height, and only counties with the virus were surveyed.
Which Phytophthora species cause black pod on cacao in Brazil?
How
40 new isolates from symptomatic pods collected on smallholder farms in 2017, plus ten earlier isolates, characterised by multi-gene sequencing, morphology and growth rates, with pathogenicity tests on pods of four clones.
Finding
19 isolates were Phytophthora palmivora and 31 formed a new species in clade 2, named Phytophthora theobromicola, which induced faster lesion expansion than P. palmivora. The authors state this is the first report of P. theobromicola on cacao and that it seems to be prevalent in field conditions in Bahia.
Does not establish: It does not show the species is new to Brazil as opposed to newly recognised, and it gives no loss figures.
Limits: 50 isolates in all. Prevalence in the field is inferred from the share of isolates, not from a designed survey.
What does the genetic diversity of Moniliophthora roreri across its current range say about its origin and spread?
How
Simple-sequence repeat markers and a public SNP data set, on isolates from areas spanning as much of the current range as possible. The abstract gives no sample size.
Finding
Two hotspots of diversity were found, coastal Ecuador and the inter-Andean Magdalena Valley of Colombia. Neither matches the Amazonian origin of cacao; both were early centres of intense cultivation. Current invasions trace to two genotypes responsible for all known instances in Central America, the Caribbean, Peru and Bolivia. The fungus is reported for the first time in Maynas in the Peruvian Amazon.
Does not establish: It does not identify the centre of origin: the abstract says the origin of the pathogen remains unknown, which differs from earlier papers that proposed Colombia.
Limits: No sample size is given in the abstract. The Maynas record is undated in the abstract and its source in Colombia is described as probable.
Which pheromone dose and formulation best attract male cocoa pod borers, and for how long?
How
Field trapping in Indonesian cocoa plantations comparing six pheromone formulations and several doses, with trapping continued through 2018.
Finding
A 0.1 mg pheromone load was significantly more attractive than lower doses. A 1.0 mg lure was no more attractive in the first month but kept the same weekly average capture efficacy for the whole twenty-seven weeks. The authors propose long-life lures for monitoring large-scale farms.
Does not establish: It does not show that pheromone trapping controls the pest; the authors present it as a monitoring tool.
Limits: The trial measured male moths caught, not pod damage or yield.
Which swollen shoot badnavirus species occur where in Côte d'Ivoire and Ghana, and how well do PCR primers detect them?
How
Field surveillance of cacao plantations in Ghana and Côte d'Ivoire with PCR assays on symptomatic leaf samples, sequencing of positive amplicons and phylogeographic clustering. The number of farms is not given in the abstract.
Finding
Detection efficiency of the primer sets ranged from 0.15% to 66.91% of symptomatic samples. Cacao swollen shoot Togo B virus (588 sequences) was the most prevalent and widely distributed species in both countries. Cacao swollen shoot CE virus (124 sequences) was largely restricted to the border regions of the two countries.
Does not establish: It does not give the share of farms or trees infected, so it is not a measure of how much cacao is diseased.
Limits: Only symptomatic leaves were tested, and the best primer set missed about a third of them, so species that the primers detect poorly are under-counted.
We would like to set two Google Analytics cookies to count page views, and to count them in our own cookie-free measurement as well. They do not identify you and nothing is shared with advertisers. Decline and the site works exactly the same — see cookies for what each one does.