Invasive Species Compendium

Detailed coverage of invasive species threatening livelihoods and the environment worldwide


Conopomorpha cramerella
(cocoa pod borer)



Conopomorpha cramerella (cocoa pod borer)


  • Last modified
  • 18 November 2019
  • Datasheet Type(s)
  • Invasive Species
  • Pest
  • Preferred Scientific Name
  • Conopomorpha cramerella
  • Preferred Common Name
  • cocoa pod borer
  • Taxonomic Tree
  • Domain: Eukaryota
  •   Kingdom: Metazoa
  •     Phylum: Arthropoda
  •       Subphylum: Uniramia
  •         Class: Insecta

Don't need the entire report?

Generate a print friendly version containing only the sections you need.

Generate report


Top of page
The adult is a small brown moth, ca 7 mm in length. It has a wingspan of about 12 mm characterized by bright yellow patches at the tips of the forewings. The moths have very long antennae which are swept backwards in their natural resting position.
CaptionThe adult is a small brown moth, ca 7 mm in length. It has a wingspan of about 12 mm characterized by bright yellow patches at the tips of the forewings. The moths have very long antennae which are swept backwards in their natural resting position.
CopyrightHo Cheng Tuck/Tropical Press Sdn Bhd
The adult is a small brown moth, ca 7 mm in length. It has a wingspan of about 12 mm characterized by bright yellow patches at the tips of the forewings. The moths have very long antennae which are swept backwards in their natural resting position.
AdultThe adult is a small brown moth, ca 7 mm in length. It has a wingspan of about 12 mm characterized by bright yellow patches at the tips of the forewings. The moths have very long antennae which are swept backwards in their natural resting position.Ho Cheng Tuck/Tropical Press Sdn Bhd
C. cramerella moth at rest.
CaptionC. cramerella moth at rest.
CopyrightJ. Mumford
C. cramerella moth at rest.
AdultC. cramerella moth at rest.J. Mumford
Cocoa pod borer damage. The larva feeds on the tissue surrounding the cocoa beans and on the placenta; the bean proper is seldom attacked.
TitleDamage symptoms on cocoa pod
CaptionCocoa pod borer damage. The larva feeds on the tissue surrounding the cocoa beans and on the placenta; the bean proper is seldom attacked.
CopyrightHo Cheng Tuck/Tropical Press Sdn Bhd
Cocoa pod borer damage. The larva feeds on the tissue surrounding the cocoa beans and on the placenta; the bean proper is seldom attacked.
Damage symptoms on cocoa podCocoa pod borer damage. The larva feeds on the tissue surrounding the cocoa beans and on the placenta; the bean proper is seldom attacked.Ho Cheng Tuck/Tropical Press Sdn Bhd
Cocoa pod borer damage: clumped cocoa beans fermenting.
TitleDamage on cocoa
CaptionCocoa pod borer damage: clumped cocoa beans fermenting.
CopyrightJ. Mumford
Cocoa pod borer damage: clumped cocoa beans fermenting.
Damage on cocoaCocoa pod borer damage: clumped cocoa beans fermenting.J. Mumford


Top of page

Preferred Scientific Name

  • Conopomorpha cramerella Snellen

Preferred Common Name

  • cocoa pod borer

Other Scientific Names

  • Acrocercops cramerella Snellen
  • Gracillaria cramerella Snellen
  • Zarathra cramerella Snellen

International Common Names

  • English: cacao moth; cocoa moth; javanese cocoa moth; rambutan borer; ram-ram borer
  • Spanish: polilla javanesa del cacao
  • French: teigne javanaise du cacaoyer

Local Common Names

  • Germany: Motte, Javanische Kakao-
  • Indonesia: penggerek buah kakao

EPPO code

  • AROCCR (Conopomorpha cramerella)

Taxonomic Tree

Top of page
  • Domain: Eukaryota
  •     Kingdom: Metazoa
  •         Phylum: Arthropoda
  •             Subphylum: Uniramia
  •                 Class: Insecta
  •                     Order: Lepidoptera
  •                         Family: Gracillariidae
  •                             Genus: Conopomorpha
  •                                 Species: Conopomorpha cramerella

Notes on Taxonomy and Nomenclature

Top of page Although the earliest report of damage caused by this pest was noted by Jansen (1860), the cocoa pod borer was first taxonomically described at the beginning of the 20th century and named Acrocercops cramerella (Snellen, 1904). Recently the generic placement of this species has been revised and it is now known as Conopomorpha cramerella (Bradley, 1986).


Top of page Adult

The adult is a small brown moth, ca 7 mm in length. It has a wingspan of about 12 mm characterized by bright yellow patches at the tips of the forewings. The moths have very long antennae which are swept backwards in their natural resting position. Moths in flight have an appearance similar to that of large, slow-flying mosquitoes.


Eggs are yellow-orange, flattened and just visible to the naked eye (ca 0.5 x 0.2 mm). Rectangular indentations cover the surface of the egg. On hatching, eggs become translucent, the shell being whitish but darkened by faeces within.


First-instar larvae are translucent white in colour and ca 1 mm long. Late-instar larvae are ca 10 mm long and creamy coloured while still inside the pod, but greenish after they emerge to pupate.


Pupae lie beneath a light-brown waterproof silken membrane tightly drawn over a depression on a pod surface or leaf.


Top of page

C. cramerella occurs only in South-East Asia and the western Pacific, and its origin within this area has been the subject of speculation since the turn of the century (Mumford, 1986).

After the appearance of C. cramerella at Manado in the 1840s, the cocoa industry in North Sulawesi declined rapidly (Toxopeus and Giesberger, 1983). The industry began to move elsewhere, to the Philippines and to Java. In about 1880 cocoa was introduced into central Java on a plantation scale to replace coffee (Hall, 1949), although there had been some earlier trade in cocoa between Java, Sulawesi and the Philippines (Wardojo, 1980). C. cramerella infestations soon appeared in the areas where cocoa was planted from seed pods carried from previously infested regions. Long-distance movement of the pest must almost certainly have taken place through the movement of infested pods, as there is no indication that the moths can fly long distances.

It is possible that a cocoa form of C. cramerella was introduced into Sabah, East Malaysia during the post-1977 'cocoa boom', with pods brought from Indonesia or the Philippines. A similar introduction from a previously infested area may have occurred in Malacca and Negeri Sembilan in West Malaysia prior to the discovery of C. cramerella in those states in 1986. However, in both cases it is also possible that there were local adaptations of rambutan-feeders or nam-nam-feeders to cocoa.

The infestation in Sabah spread rapidly from Tawau (south-east Sabah) across the entire state during a 3-year period (Tay, 1987), and it soon spread to Sarawak. The initial infestations discovered in West Malaysia were contained for several years, thanks to vigorous quarantine efforts based on experience from east Malaysia (Chin, 1987; Ling et al., 1987). Unfortunately scattered outbreaks began to appear in other parts of West Malaysia during the first half of 1988 and it was not possible to extend this vigorous containment and eradication effort to a wider area.

C. cramerella is presently spreading through northern and central Sulawesi as smallholder production in that area is increasing. There is considerable speculation that this infestation has been introduced from Sabah, but the early history of C. cramerella in Sulawesi could also account for the new infestations.

Live C. cramerella borers are known to travel long distances within rambutan fruits. Rambutans exported from Thailand to Saudi Arabia containing live pupae were found in a Riyadh supermarket (Mumford and Ho, 1988).


Distribution Table

Top of page

The distribution in this summary table is based on all the information available. When several references are cited, they may give conflicting information on the status. Further details may be available for individual references in the Distribution Table Details section which can be selected by going to Generate Report.

Last updated: 10 Jan 2020
Continent/Country/Region Distribution Last Reported Origin First Reported Invasive Reference Notes


BruneiPresentEPPO (2014)
ChinaPresent, LocalizedEPPO (2014)
IndiaPresentCABI (Undateda)Present based on regional distribution.
-BiharPresentMukherjee and Rajeev Ranjan (2005)
-Himachal PradeshPresentRanjeet Bhatia and Divender Gupta (2003)
-West BengalPresentNair and Sahoo (2006)
IndonesiaPresent, WidespreadEPPO (2014); UK, CAB International (1984)
-Maluku IslandsPresentUK, CAB International (1984); Waterhouse (1993); EPPO (2014); CABI (Undated)
-Irian JayaPresentWaterhouse (1993); EPPO (2014)
-JavaPresentUK, CAB International (1984); Waterhouse (1993); EPPO (2014); CABI (Undated)
-SumatraPresentUK, CAB International (1984); Waterhouse (1993); EPPO (2014)
-SulawesiPresentUK, CAB International (1984); Waterhouse (1993); EPPO (2014); CABI (Undated)
MalaysiaPresent, LocalizedEPPO (2014)
-SabahPresentUK, CAB International (1984); EPPO (2014)
-SarawakPresentUK, CAB International (1984); EPPO (2014)
-Peninsular MalaysiaPresent, WidespreadUK, CAB International (1984)
PhilippinesPresentUK, CAB International (1984); Waterhouse (1993); EPPO (2014)
Sri LankaPresentMumford and Ho (1988)
TaiwanPresent, LocalizedEPPO (2014)
ThailandPresentMumford and Ho (1988); Waterhouse (1993)


AustraliaAbsent, EradicatedDAFF (2014); EPPO (2014)via PestLens
-Northern TerritoryAbsent, EradicatedUK, CAB International (1984); EPPO (2014)
-QueenslandAbsent, EradicatedDAFF (2014); Australia Biosecurity Queensland (2011); EPPO (2014)via PestLens
Papua New GuineaPresentUK, CAB International (1984); EPPO (2014)
SamoaPresentUK, CAB International (1984); EPPO (2014)

Risk of Introduction

Top of page Quarantine and surveillance for C. cramerella remains a problem in areas of South-East Asia in which the pod borer does not presently affect cocoa (Thailand, Sri Lanka, Sumatra, Irian Jaya and Papua New Guinea). Rambutan or nam-nam borers are already known from Thailand, Sri Lanka and New Britain (Papua New Guinea). C. cramerella has also been found from unspecified hosts, in Western Samoa and the Northern Territory of Australia early in the 20th century (Mumford, 1986). Live borers can travel long distances within rambutan fruit: healthy C. cramerella pupae were found on Thai rambutans in a supermarket at Riyadh, Saudi Arabia in 1986 (JD Mumford, Imperial College, Ascot, UK, personal communication, 1996), but no other cases of C. cramerella getting that far through quarantine have been reported.

Hosts/Species Affected

Top of page Sapindaceous and leguminous species (rambutan, pulasan, kasai and nam-nam) are most likely to be the original hosts of C. cramerella as they are indigenous to the region to which the moth is restricted, whereas cocoa is a recently introduced species. Wood (1980) notes that rambutan and nam-nam produce fruits with pulp similar to cocoa. However, they tend to be more seasonal than cocoa and so may not provide the right conditions for permanent establishment. This could limit populations in local areas.

Host Plants and Other Plants Affected

Top of page
Plant nameFamilyContext
Cola acuminata (cola)SterculiaceaeMain
Nephelium lappaceum (rambutan)SapindaceaeMain
Pometia pinnata (fijian longan)SapindaceaeWild host
Theobroma cacao (cocoa)MalvaceaeMain

Growth Stages

Top of page Fruiting stage


Top of page External damage to the pod caused by the cocoa pod borer is seen as entry and exit holes created by the tunnelling larvae on the husk, and overall premature or uneven ripening (yellowing) of pods caused by internal feeding activity. If pods are cut open, characteristic tunnels and scarification caused by feeding commonly cause beans to stick together. Harvested beans clump together in severe infestations and may be impossible to extract from damaged pods.

List of Symptoms/Signs

Top of page
SignLife StagesType
Fruit / internal feeding
Seeds / internal feeding

Biology and Ecology

Top of page The eggs of C. cramerella may be laid anywhere on the surface of host-plant pods although there appears to be some preference for primary or secondary furrows of pods. Eggs are not laid on pods less than 5 cm in length, and are most commonly laid on pods 2-6 weeks from maturity (assuming maturity as the first sign of yellowing). On hatching, the first-instar larva tunnels through the floor of the egg shell and bores perpendicular to the pod surface until it reaches the sclerotic layer of the husk. At this point the larva either tunnels directly through, or tunnels along the surface of the sclerotic layer for up to several centimetres before penetrating the husk. Once inside the pod, tunnelling becomes less directional, the inner surface of husks, pulp of beans and placentae being randomly tunnelled into and fed upon. In younger pods, early-instar larvae could penetrate the testa of developing beans, and in some cases cotyledons could be attacked. Such feeding causes scarification and subsequent sticking of the beans. Damage to the funicles of pods often results in beans being malformed and under-sized, significantly reducing quality and thus the value of the processed beans. Feeding in pods also causes them to yellow or ripen unevenly and prematurely, confusing ripeness standards for harvesting.

The entire larval stage takes 14-18 days to complete, with 4-6 instars. The larvae then tunnel out through the pod wall, leaving an easily identifiable exit hole.

Once outside the pod, larvae crawl or lower themselves by a silk thread to a suitable site for pupation. The pupation site could be in a furrow of the pod, or green dried leaves and other debris. Once at this site the larvae spin oval-shaped cocoons and enter a short prepupal stage before forming obtect pupae. The pupal stage normally takes 6-8 days to complete. It is also at this stage that the pest is most likely to be transported by man to other cocoa-growing areas through movement of pods, leaves and other objects to which pupae are attached.

The moths are most active at night, mating and laying of eggs being carried out at this time. A female can normally produce 50-100 eggs in her lifetime. During the day, adult moths normally rest underneath horizontally inclined branches of the cocoa tree, and their cryptic coloration blending with the resting place makes them very difficult to spot. Adult longevity is 1-30 days, but adults generally live for ca 1 week. In total, the entire life cycle takes ca 1 month to complete.

Natural enemies

Top of page
Natural enemyTypeLife stagesSpecificityReferencesBiological control inBiological control on
Acrocerilia pachynervis Parasite Larvae
Beauveria bassiana Pathogen Pupae
Bracon Parasite Larvae
Ceraphron aguinaldoi Parasite Pupae Sabah cocoa
Chelonus chailini Parasite Larvae
Chrysonotomyia Parasite Pupae
Diaglyptidea roepkei Parasite
Dolichoderus thoracicus Predator Malaysia
Goryphus fasciatipennis Parasite
Goryphus javanicus Parasite
Goryphus mesoxanthus Parasite
Nesolynx thymus Parasite Pupae
Paecilomyces fumosoroseus Pathogen
Paecilomyces lilacinus Pathogen
Paraphylax fasciatipennis Parasite Philippines cocoa
Pediobius soror Parasite
Photoptera erythronota Parasite
Trichogrammatoidea bactrae fumata Parasite Eggs Malaysia; Sabah; South East Asia cocoa
Trichogrammatoidea cojuangcoi Parasite Eggs
Trichospilus pupivora Parasite

Notes on Natural Enemies

Top of page Natural mortality from predation in the field appears to be greatest from general predators of pupae, in particular Iridomyrmex spp. Normal predation levels are in the range 40-60%. Recent effort has been directed towards use of Dolichoderus and Anoplolepis spp., whose nests can be augmented, as additional sources of predation. This has met with limited success.

Two major parasitoid-based control programmes have been attempted. Introductions of exotic parasites from Sri Lanka to Malaysia and from west to east Malaysia, by the International Institute of Biological Control (IIBC), were unsuccessful in the 1980s due to poor establishment. Large-scale mass production of Trichogrammatoidea species in Malaysia and the Philippines by both Agriculture Departments and private industry have had much greater success and have given approximately 50% reduction in losses, but at costs much higher than spraying insecticides. The use of egg parasites has been practised on many estates, but has generally been discontinued due to costs and organisational difficulties involved in maintaining the insectaries.

Natural parasitism is detected at significant levels only at the peak-crop period when C. cramerella numbers are highest.


Top of page C. cramerella has become the most important insect pest of cocoa in many parts of South-East Asia over the past 150 years. C. cramerella causes losses to cocoa by boring in the placental tissues and the wall of the pod, disrupting the development of the beans. Feeding results in pods that may ripen prematurely, with small, flat beans, often stuck together in a mass of dried mucilage. The beans from seriously infested pods are completely unusable, and in heavy infestations over half the potential crop can be lost. In light infestations there is no loss, but control may still be needed to prevent higher infestations from developing.

For information on economic threshold levels, see Control.

Detection and Inspection

Top of page The cryptic coloration and night activity of the adult moths renders their detection within a cocoa stand very difficult, although specimens can sometimes be seen resting on the undersides of horizontally inclined branches during the day. Inspection methods are usually based on assessing pods for larval damage. Exit holes or larval debris on the pods are a clear indication of the presence of the pest. Within a crop, uneven or premature ripening of pods, recognised by a yellowing of the surface, is also a good indication of internal feeding activity by the larvae.

Once pods are opened, characteristic tunnels caused by larvae and their presence within such tracts is confirmatory evidence of the pest's presence. Extensive feeding within the pod commonly leads to individual beans sticking together, which often means that bean extraction from the pod is difficult.

Pupae can be found on the pods or on leaf litter, and this is often a primary source of pest infestation.

Similarities to Other Species/Conditions

Top of page C. oceania, C. sinensis and C. litchiella are three congeneric species found in South-East Asia which are very similar to C. cramerella. They can be differentiated from C. cramerella by wing pattern and male and female genitalia (Bradley, 1986).

Prevention and Control

Top of page

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Harvesting (Rampasan)

In the early days of the 20th century Roepke (1913) considered rampasan to be the only feasible control method. Research on the life cycle and oviposition habits of C. cramerella in the early 1980s confirmed that removing all pods longer than 6-7 cm from a field for 6 weeks would break the life cycle of C. cramerella, as female moths will not usually lay eggs on younger pods. No evidence was found for easy transfer from one host species to another, so the main problems for rampasan are the proximity of other cocoa in which C. cramerella is not controlled, and the difficulty of practising complete rampasan in normal estates. Without heavy pruning, complete elimination of a population of C. cramerella through rampasan is extremely unlikely. For further information see Mumford (1986); Mumford and Ho (1988); Wood et al. (1992).

Mumford (1986) and Mumford and Ho (1988) observed that the great majority of C. cramerella larvae emerge from cocoa pods after they become ripe, and that if pods are picked at the earliest stage of ripeness then almost 90% of larvae will still be inside the pods. If pods are broken quickly and the husks destroyed, larval mortality will be very high and a good degree of control may be achieved. Alternatively, unbroken pods could be kept in plastic bags for several days, either to contain emerging larvae or to kill them due to the high temperature inside the bags. Although this method has never been conclusively tested, Wood et al. (1992) did confirm effectiveness in large trial areas over a period of several seasons. Thorough harvesting is required at intervals of 14 days or less in order to be sure to collect pods before they mature and before larvae emerge in the field. Unsatisfactory results may arise from incomplete harvesting.

However, the greatest benefit from frequent, complete harvesting is likely to be gained if it is done during the low-crop period through the first half of the rising crop. Complete, frequent, regular harvesting (CFRH) is to some extent an extension of the principle of rampasan, but aimed at larval emergence rather than oviposition. While CFRH does not completely break the life cycle, it should help to keep the C. cramerella population low (as occurs at the low-crop period), and thorough early harvesting helps to keep the C. cramerella population relatively low during the time when the rising peak crop is most susceptible to oviposition (about 2 months after the low-crop point).

An alternative would be to abandon harvesting during the low crop, and at the first sign of the rising crop to begin very intensive CFRH for several months. The economic implications of this idea would need to be tested in trials.

In C. cramerella-infested areas in the southern Philippines, some cocoa has been planted at very high densities in hedges with access for small tractors between pairs of rows. Trees have been kept to a low height so that all harvesting can be done within easy reach. Mechanization allows frequent, regular harvesting, and the hedge-like structure of the crop (1-m squares within the double rows, and 2-3 m between rows for mini-tractor access) allows very complete harvesting. Under this sytem, infestations of C. cramerella were at insignificant levels during the late 1980s, without any other form of control being applied.

Day (1985) developed the untested idea that increasing the amplitude of the cocoa crop cycle could reduce the damage from C. cramerella. If C. cramerella populations can be reduced to very low levels during the low-crop period, a natural rampasan effect could be induced. This could be achieved by use of plant-growth regulators to induce synchronous flowering at certain times during the year.

Each of these forms of cultural control poses economic and management difficulties. CFRH must be seen as a form of pest control, not just as more labour-intensive harvesting, and may require compensatory management procedures. An alternative could be to employ normal piece-rate harvesting gangs, followed through the fields by specialized flat-rate-payment 'clean-up' gangs. More synchronous cropping, whether achieved by plant-growth regulators or selection of more seasonal cropping cultivars or clones, creates problems with more seasonally varied labour demands.

Mechanical Control (Sleeving)

The idea of sleeving pods with bags of plastic or other materials to prevent egg laying originated in Indonesia. Thin plastic bags with open bottoms for ventilation, placed on very young pods (less than 7 cm long) and left throughout the pod maturation period, can result in virtually complete protection from C. cramerella. The main problems are that bags are sometimes placed on pods too late, or that insufficient ventilation may result in rots, and that this method is expensive in terms of labour. It is impractical to put bags on pods that cannot be reached from the ground, so only the lowest pods can be protected. This can result in greater levels of infestation in the upper-level pods, and in some cases the concentration of infestations in the upper levels can give greater losses than if a lower level of infestation was spread over all the pods in the field. Sleeving is unlikely to be a feasible option for cocoa growers, other than those with young trees, inexpensive family labour, or those in a desperate position to ensure a partial crop.

Biological Control

Dutch entomologists in the early part of the 20th century advocated that ants (the large black ant, Dolichoderus sp., and the weaver ant, Anoplolepis sp.) should be encouraged to prey on C. cramerella prepupal larvae on emergence from the pod and on pupae, and to disturb adults. Research during the 1980s suggested that most predation was actually due to the small 'sugar' ants, Iridomyrmex spp., which are much more difficult to augment or manipulate (Day, 1985). A long-term study by Day showed that ant predation was almost constant at around 40% of pupae each month throughout a 4-year period (Day, 1985). Predation levels were independent of C. cramerella pupal density, as would be expected from generalist predators.

Low levels of natural pupal parasitism were found in Sabah, Malaysia, but little progress was made in finding alternative hosts for the parasites. No practical control effort was made using indigenous pupal parasites, as there was no indication of significant parasitism. However, due to the presence of natural predators and parasites of C. cramerella, and due to concern for the natural control of other cocoa pests such as stem and branch borers and bagworms, selective spraying was recommended so that natural enemies (and pollinators) were not disrupted.

Control by egg parasites had originally been dismissed because of the small size and dispersed distribution of C. cramerella eggs. However, considerable effort by Lim and his colleagues demonstrated the presence of a parasitic wasp (Trichogrammatoidea sp.) in Sabah (Lim and Chong, 1987). An intensive programme of rearing these wasps on an alternative host (Corcyra cephalonica) in laboratories and commercial breeding rooms, and releasing them into the field, gave surprisingly good levels of control, but at prohibitive costs for about 12 500 wasps/ha/day.

Exploration was also undertaken to find exotic natural enemies to import into Sabah, centring at first on Peninsular Malaysia. A relatively low level of C. cramerella was known to exist on rambutan and other non-cocoa hosts in West Malaysia, and it was possible that natural enemies were partially responsible for keeping the populations at low levels. Parasites were likely to be more specific than predators, so attention centred on wasps rather than ants. Searches for parasites were extended by IIBC's Malaysian research station to Indonesia, the Philippines, Sri Lanka and Papua New Guinea, and two wasp species (Ceraphron and Ooencyrtus) were introduced in the field in small numbers to Sabah, while a third species (Nesolynx) failed to establish in field cages. None of these has established successfully; general predation was cited as a possible explanation for the failure of these introductions.

Host-Plant Resistance

Early Dutch interest in host-plant resistance focused on the surface of the pod. Eggs were generally found in the furrows on the pod surface, smoother pods being less attractive than deeply ridged cultivars, though this appears not to have been tested or developed. Day (1985) found no obvious differences in egg laying on different cultivars, but considerable differences in larval mortality were found inside pods of different cultivars. Greater mortality occurred in pods with either thicker or harder stony endocarp layers in the pod wall. Larval survival was as much as 10 times greater in soft/thin-walled cultivars. Azhar and Lim (1987) studied several dozen cultivars in seed gardens and concluded that the hardness of the sclerotic layer was the important factor, and they confirmed that very wide ranges of larval survival occurred in existing cultivars in Sabah. Sime Darby Plantations, Malaysia followed up these observations by planting a block of particularly hard-walled cultivars in an effort to demonstrate the effect at Segaliud in Sabah, but unfortunately the block was uprooted due to low cocoa prices, before it gave conclusive results (JD Mumford, Imperial College, Ascot, UK, personal communication, 1996).

A major constraint to the adoption of host-plant resistance in the mid-1980s was the time and cost involved in replanting existing stands with hard-podded cultivars. This was exacerbated by the susceptibility of several hard varieties (AML, PA) to vascular streak dieback (Oncobasidium theobromae), variation in pod properties within individual cultivars and the general lack of interest in planting new areas while cocoa prices were falling in relation to alternative crops. Recent innovations, particularly at BAL Estates in Sabah, in clonal tissue culture and in grafting clonal tissue onto mature stems, may allow partial host-plant resistance to play a much greater part in C. cramerella control in the future (JD Mumford, Imperial College, Ascot, UK, personal communication, 1996).

Chemical Control

Observations in the 1950s and 1960s showed that cocoa in South-East Asia could not survive continuous blanket sprays of broad-spectrum insecticides, because of destructive outbreaks of secondary pests freed from their normal control by natural predators and parasites (Conway, 1973; Toxopeus and Giesberger, 1983). The long-term practical use of insecticides for control of C. cramerella requires that the chemical is applied selectively, either to undersides of the lower branches where natural enemies are less common, or at times when natural enemies are less abundant. Several developments in the 1980s have improved chemical control, without resulting in serious secondary pest outbreaks.

Day (1985) and Mumford and Ho (1988) noticed that adult C. cramerella rest by day on the undersides of lower branches in cocoa, and this provides an ideal site for selective spraying as it is relatively dry, dark, inert and accessible. Relatively small amounts of contact insecticide, either pyrethroid or carbamate, applied to these resting sites during the low-crop period, kept C. cramerella populations below economic damage levels during subsequent peak populations in several trials. Sprays applied at peak crop periods had little effect on infestations in the following months, when falling pod numbers could obscure differences in percentage infestation. As with cultural controls, heavy pruning and height maintenance ensure that all the lower branches are accessible. Branch spraying was advocated only in the low-crop period because this was believed to give the greatest returns, as it protected the susceptible peak crop developing in the several months immediately following spraying. Applications at the peak crop did not provide good returns, as C. cramerella populations naturally fall after the peak crop anyway, and the protection afforded by the sprays only applies to the relatively low crop maturing in the 3 months after the peak.

Aerial spraying was tried in Sabah, without good results. The pesticide is mainly deposited on the shade trees where C. cramerella moths may spend some time between 21.00 and 06.00 h, based on the relatively high catches of moths in pheromone traps placed above the cocoa canopy. However, sunlight and rain are likely to reduce the persistence of the pesticide very quickly. Aerial spraying is not possible in hilly areas or on estates with irregular high jungle shade. Aerial spraying during the C. cramerella quarantine-eradication effort in West Malaysia greatly reduced mosquito infestations in the estates. The best return from pod spraying is likely to come from applications during the early stage of a rising crop, soon after the low-crop period.

Wood et al. (1992) did not find much effect from spraying alone. A possible explanation for the apparent effect of frequent spraying, either in trials or in commercial practice, is that fields that are frequently sprayed are also likely to be frequently harvested and may also be partially pruned to make spraying easier or more practical, whereas unsprayed control areas may not be harvested as often or as completely, especially if branches are not trimmed.

Pheromonal Control

Day (1985) and Beevor et al. (1986) showed that female C. cramerella moths attract males by releasing a pheromone, which was identified and synthesized in the laboratory. Many more male moths were trapped using synthetic pheromones than could be caught in traps baited with female C. cramerella moths; these traps could be used for both monitoring and control. Control is achieved if a large proportion of male moths are trapped before they can mate, thus reducing the overall mating success of female C. cramerella moths and lowering fecundity.

A large-scale (200-ha) trial was conducted over 4 years at BAL Estates in Sabah during the mid 1980s, to test the effect of pheromone trapping as a control method (Beevor et al., 1993). Trials and commercial-scale control using traps were also undertaken on several other estates in Sabah. Trap densities of four and eight per ha were used to give economic control. Results of these trials indicated that losses were reduced by about one-third in trapped areas compared with untreated areas. A major problem with pheromone-trapping trials was in finding adequate untreated control areas. The pheromones were shown to catch moths as far as 800 metres from infested cocoa, and so the size of both treated and untreated areas needed to be very large to give good comparisons. Neighbouring areas were also subject to pruning, harvesting and spraying actions which may have distorted results.

In 1987 a race of C. cramerella was found in West Malaysia that did not respond to the pheromone used in Sabah. Analysis showed that the mixture of pheromone molecules in the new race was very different from the original formula, and the new mixture was shown to catch C. cramerella moths in West Malaysia. The new mixture also caught large numbers of C. cramerella moths in Sabah, casting doubt on the future of mass trapping and raising questions about the origins and interactions of the two races. There was an immediate problem of keeping the two mixtures separated; when used together in the same trap no moths of either type would be caught. More importantly, however, there was a fear that further races could develop or appear which would require frequent changes of pheromone components, adding to costs and limiting efficacy. As a result, pheromones are not used to control C. cramerella at present, although occasional commercial interest is shown in developing new pheromone mixtures.

Integrated Pest Management

The most immediate reductions in C. cramerella are likely to come about through better managed harvesting and spraying. Both of these rely on well-pruned trees kept to a height low enough to collect and/or spray all pods. See sections on Harvesting (Rampasan) and Chemical Control. Longer term control may be improved by grafting or replanting with hard-walled clones. Further releases of exotic natural enemies may provide additional partial control, if suitable parasites can be found.

Field Monitoring/Economic Threshold Levels

Day (1985) presented a damage function for C. cramerella relating loss of yield to percentage infestation in harvested pods. The function indicated almost no loss with infestations up to approximately 60%, and rapidly increasing losses with higher infestations (over half the crop lost in very heavy infestations). While this is a reasonably practical way of estimating damage, the relationship is inevitably inaccurate at higher infestation levels. Low infestations (20-40%) consist of a small proportion of pods each with only one larva inside, whereas high infestations (80-100%) could have many pods with 2, 3 or up to 30 larvae inside. Percentage infestation does not give a good measure of the actual number of larvae in the pods at high percentages, while it does at low percentages. However, for setting priorities for control action this is probably adequate: infestations high enough to be in the inaccurate area would certainly need control anyway, while at lower infestations when there is more doubt about the immediate need for control, the function appears to be reasonably accurate.

An absolute measure of C. cramerella damage can only be obtained by comparing the yields over an entire cropping period, looking at the overall tonnage produced and especially at the dry bean kg/harvested pod ratio, which shows the proportion of discarded pods and beans across the cropping period. It is important to look at the whole cropping period because losses for the season are made up of a mix of high proportional losses on the low crop and lower proportional losses in the high crop.

Many estates have tried to develop C. cramerella census systems to estimate losses and to plan control programmes. The most common form has been to establish a grid of marked sample trees or rows in each block which are monitored regularly for infestation (clean, infested and very highly infested pods are recorded). Control, usually pod or branch spraying, would then be applied to blocks either over a preset threshold or up to a predetermined number of blocks in priority set by the census results (for instance, the worst 20%). Estates with census systems almost always obtained better control than those without, but an estate manager who can organise a good census plan is also likely to be harvesting more frequently and spraying well-maintained trees. Putting some of the census effort into even better routine harvesting and maintenance may obviate the need for the census.

An important problem with census systems is understanding the significance of the results from individual census rounds. C. cramerella losses result from a combination of numbers of larvae and numbers of pods at an attractive, susceptible stage of maturity (the last 6-8 weeks before ripeness, as a general rule). The infestation in a cohort of mature pods reflects C. cramerella attack over one and half to two generations of moths, it is not a point sample. Furthermore, a 50% infestation one month after the peak will surely rise over the next 2 months, causing higher proportional damage but on a falling crop, while a 50% infestation 6 weeks before a peak will almost certainly fall, as pod numbers are likely to increase faster than C. cramerella populations in normal crop cycles. Therefore, the interpretation of any single census figure (percentage infestation) depends on a prediction of the numbers of pods reaching maturity over the following 3 months. This means that in addition to the fairly simple task of monitoring percentage infestations, the pod-age structure should also be monitored, which is not as easy.

Pheromone traps provide another possible form of monitoring, but they only provide a relative measure of populations, as the range of attraction is unknown and variable. The problem of two races of C. cramerella (see Pheromonal Control) greatly reduces the value of pheromone traps for estimating numbers of moths, as it is not clear whether both types are equally attracted, nor whether other races remain undetected. Pheromone traps may, however, be useful for quarantine detection in uninfested areas.


Top of page

Australia Biosecurity Queensland, 2011. Cocoa pod borer. Communique., Australia: Department of Agriculture, Fisheries and Forestry, Australian Government, 1.

Azhar I; Lim DHK, 1987. An investigation on the use of host plant resistance and crop manipulation in the management of the cocoa pod borer. Proc. Symposium on Management of the Cocoa Pod Borer, March 1987, Malaysian Plant Protection Society, Kuala Lumpur, Malaysia, 83-101.

Beevor PS; Cork A; Hall DR; Nesbitt BF; Day RK; Mumford JD, 1986. Components of female sex pheromone of cocoa pod borer moth, Conopomorpha cramerella. Journal of Chemical Ecology, 12(1):1-23

Beevor PS; Mumford JD; Shah S; Day RK; Hall DR, 1993. Observations on pheromone-baited mass trapping for control of cocoa pod borer, Conopomorpha cramerella, in Sabah, East Malaysia. Crop Protection, 12(2):134-140

Bradley JD, 1986. Identity of the South-East Asian cocoa moth, Conopomorpha cramerella (Snellen) (Lepidoptera: Gracillariidae), with descriptions of three allied new species. Bulletin of Entomological Research, 76(1):41-51

Chin F, 1987. Views on the eradication programme for the cocoa pod borer in Peninsular Malaysia. Proc. Symposium on Management of the Cocoa Pod Borer, March, 1987, MAPPS, Kuala Lumpur, Malaysia, 176-180.

CIE, 1984. Distribution Maps of Pests, No. 458. Wallingford, UK: CAB International.

Conway GR, 1973. Ecological aspects of pest control in Malaysia. In: The Careless Technology, Eds. Farvar, MU, Milton, London: JP. Stacey, 467-488.

DAFF, 2014. Cocoa pod borer. Queensland Department of Agriculture, Fisheries and Forestry (DAFF). January 16, 2014., Australia: Queensland Department of Agriculture, Fisheries and Forestry (DAFF).

DAFF, 2014. Eradication of cocoa pod borer. Australian Department of Agriculture, Fisheries and Forestry (DAFF). January 20, 2014., Australia: Australian Department of Agriculture, Fisheries and Forestry (DAFF).

Day RK, 1985. Control of the cocoa pod borer. PhD. thesis. London, UK: Imperial College, University of London.

Entwistle PF, 1972. Pests of cocoa. London, UK: Longman, 779 pp.

EPPO, 2014. PQR database. Paris, France: European and Mediterranean Plant Protection Organization.

Hall CJJ van, 1949. Cacao. In: van Hall CJJ, van de Koppel C, Uitgeverij NV, van Hoev W, Gravenhage S, eds. De Landbouw in de Indische Archipel IIB, 272-346.

Jansen AJF, 1860. Aantekingen betreffende de kakao cultuur in de residentie Manado. Natuurkundig Tijdschrift voor Nederlandsch Indie 20 vierde serie deel VI, 289-307.

Kalshoven LGE; Laan PA van der (Reviser and translator), 1981. Pests of crops in Indonesia (revised). Jakarta, Indonesia: Ichtiar Baru, 701 pp.

Lim GT; Chong TC, 1987. Biological control of cocoa pod borer by periodic release of Trichogrammatoidea bactrae fumata Nagaraja in Sabah, Malaysia. Proc. Symposium on Management of the Cocoa Pod Borer, March, 1987, Malaysian Plant Protection Society, Kuala Lumpur, Malaysia, 71-80.

Mukherjee U; Rajeev Ranjan, 2005. Comparative bioefficacy of some synthetic insecticides and phytoproducts against litchi fruit borer, Conopomorpha cramerella (Snell.). Plant Protection Bulletin (Faridabad), 57(1/2):23-25.

Mumford JD, 1986. Control of the cocoa pod borer (Acrocercops cramerella): a critical review. In: Pushparajah E, Chew PS, eds. Cocoa and Coconuts: Progress and Outlook. Eds. Incorporated Soc. of Planters, Kuala Lumpur, Malaysia, 277-286.

Mumford JD; Ho SH, 1988. Control of the cocoa pod borer (Conopomorpha cramerella). Cocoa Growers' Bulletin, No. 40:19-29

Nair N; Sahoo AK, 2006. Incidence of Conopomorpha cramerella Snellen (Gracillariidae: Lepidoptera): a serious pest of litchi and its control in West Bengal. Environment and Ecology, 24S(Special 3A):772-775.

Ranjeet Bhatia; Divender Gupta, 2003. Insect and mite pest status of subtropical horticultural crops in Himachal Pradesh. Journal of Insect Science (Ludhiana), 16(1/2):1-8.

Roepke W, 1913. Nieuwe onderzoekingen omtrent de parasieten van de cacaomot. Meded. Proefstat. Midden Java 12:1-21.

Schotman CYL, 1989. Plant pests of quarantine importance to the Caribbean. RLAC-PROVEG, No. 21:80 pp.

Snellen PCT, 1904. Over de ontwikkelingstoestanden van eenige Microlepidoptera van Java. Tijshrift voor Entomologie, 46:79-90.

Tay EB, 1987. Control of cocoa pod borer - the Sabah experience. Proc. Symposium on Management of the Cocoa Pod Borer, March 1987. Kuala Lumpur, Malaysia: Malaysian Plant Protection Society, 7-17.

Toxopeus H; Giesberger G, 1983. History of cocoa and cocoa research in Indonesia. Archives on Cocoa Research, 2:7-34.

Wardojo S, 1980. The cocoa pod borer moth - a major hindrance to cocoa development. Indonesian Agricultural Research and Development Journal, 2:2-4.

Waterhouse DF, 1993. The Major Arthropod Pests and Weeds of Agriculture in Southeast Asia. ACIAR Monograph No. 21. Canberra, Australia: Australian Centre for International Agricultural Research, 141 pp.

Wood BJ; Chung GF; Sim SC; Chee CF, 1992. Trials on control of the cocoa pod borer Conopomorpha cramerella (Snellen) in Sabah by regular complete harvesting. Tropical Pest Management, 38:271-278.

Wood GAR, 1980. The cocoa pod borer-a menace to South East Asia. Cocoa Growers' Bulletin No. 30:25-27.

Wood GAR; Lass RA, 1985. Cocoa. London, UK; Longman, xix + 620pp.

Distribution References

Australia Biosecurity Queensland, 2011. Cocoa pod borer. In: Communique, 1 Australia: Department of Agriculture, Fisheries and Forestry, Australian Government.

CABI, Undated. CABI Compendium: Status as determined by CABI editor. Wallingford, UK: CABI

CABI, Undated. CABI Compendium: Status inferred from regional distribution. Wallingford, UK: CABI

CABI, Undated. Compendium record. Wallingford, UK: CABI

DAFF, 2014. Cocoa pod borer. In: Queensland Department of Agriculture, Fisheries and Forestry (DAFF). January 16, 2014, Australia: Queensland Department of Agriculture, Fisheries and Forestry (DAFF).

EPPO, 2014. EPPO Global database (available online). Paris, France: EPPO.

Mukherjee U, Rajeev Ranjan, 2005. Comparative bioefficacy of some synthetic insecticides and phytoproducts against litchi fruit borer, Conopomorpha cramerella (Snell.). Plant Protection Bulletin (Faridabad). 57 (1/2), 23-25.

Mumford J D, Ho S H, 1988. Control of the cocoa pod borer (Conopomorpha cramerella). Cocoa Growers' Bulletin. 19-29.

Nair N, Sahoo A K, 2006. Incidence of Conopomorpha cramerella Snellen (Gracillariidae: Lepidoptera): a serious pest of litchi and its control in West Bengal. Environment and Ecology. 24S (Special 3A), 772-775.

Ranjeet Bhatia, Divender Gupta, 2003. Insect and mite pest status of subtropical horticultural crops in Himachal Pradesh. Journal of Insect Science (Ludhiana). 16 (1/2), 1-8.

UK, CAB International, 1984. Acrocercops cramerella. [Distribution map]. In: Distribution Maps of Plant Pests, Wallingford, UK: CAB International. Map 458.

Waterhouse D F, 1993. The major arthropod pests and weeds of agriculture in Southeast Asia. Canberra, Australia: ACIAR. v + 141 pp.

Distribution Maps

Top of page
You can pan and zoom the map
Save map