Darna trima
(nettle caterpillar)

D. trima has a limited regional distribution being found in Indonesia, Malaysia and Singapore only. Records of the occurrence of D. trima in China and Thailand probably refer to other Darna species, most likely D. furva (see Holloway et al., 1987).

D. trima has a very wide host range covering many plant families.

The life cycle of D. trima has been reported by several authors and is summarized in a table by Holloway et al. (1987). As an example, the life cycle reported by Tiong and Munroe (1977) is as follows: eggs 2-3 days, larvae 30-33 days, pupae 12-14 days and adults 7-10 days.

On oil palm, eggs are laid singly or in groups of up to four (Holloway et al., 1987) and are glued to the abaxial surface of the leaflets (Tiong and Munroe, 1977). The first-instar larva feeds on the abaxial leaf epidermis and mesophyll layer resulting in small (1.5-2.0 x 2.0-3.0 mm) lacerated patches but the adaxial epidermis is left intact (Tiong and Munroe, 1977). The second and later instars are able to consume the whole thickness of the leaflet but not the midrib. The mature larva pupates within a cocoon on the lower part of the frond near the base of the leaflets, in the leaf axils, amongst the epiphytes commonly found on the trunk, or among the debris at the base of the palm (Holloway et al., 1987). The adult emerges through a circular opening in the cocoon.

Outbreaks of limacodids have been observed to occur suddenly (Wood, 1982). An infestation of D. trima seems to start as a big pocket (Syed and Shah, 1977). Attacks may occur at particular times of the year, e.g. at the beginning of the year in Peninsular Malaysia. They are known to recur in the same locality year after year.

Rainfall is believed to play an important role in the population dynamics of many tropical insects. Increases in the populations of insect pests of oil palm are often associated with periods of low rainfall. Rainfall may help to regulate oil palm pests by making conditions conducive for the promotion of certain pathogens, e.g. non-occluded virus in D. trima (Syed and Shah, 1977).

Parasitoids and predators are an important mortality factor in the dynamics of limacodid populations and certain plantation practices can decimate these natural enemies. The detrimental effects of insecticides, particularly the non-selective ones, are well known (Syed and Shah, 1977; Wood, 1987). Less well known are the indirect effects of indiscriminate weed control or loss of ground cover due to fire; certain types of flowering weeds are necessary for maintaining the natural enemy fauna (Syed and Shah, 1977).

Of the various species of parasitoids of D. trima found in Sumatra, Trichogrammatoidea thoseae has been singled out as being of major importance because it accounts for a high proportion of egg mortality (Desmier de Chenon et al., 1990). However, its short life cycle of 12-15 days requires that alternative hosts are available so that it may persist in sufficiently high numbers.

Other potentially effective candidates for use in biological control of limacodids are members of the tribe Euplectrini, family Eulophidae (Desmier de Chenon et al., 1990). Among the species in the tribe is Platyplectrus orthocraspedae. The larvae of this predator are ectoparasitoids of the early instars of Limacodidae and the adults prey on later instars.

Asopin bugs belonging to the genera Eocanthecona and Cantheconidea are regarded as good candidates for biological control of leaf-eating caterpillars of oil palm (Desmier de Chenon et al., 1990). These predators lay their eggs on the oil palm, and nymphs and adults live on young and mature palms (even the tall ones). They feed on a wide range of prey, including all the lepidopteran families with the exception of bagworms. Their fecundity is high and their life cycle is short at 2 months.

Ants (species unspecified) have been observed to account for considerable mortality in the host prepupae present on the ground (Young, 1971).

Viruses can give rise to epizootics which may quickly terminate outbreaks (Wood, 1968; Syed, 1971; Holloway et al., 1987). A review of virus diseases of Limacodidae is given by Entwistle (1987). Viruses identified from D. trima collected from oil palm in Sarawak were a granulosis virus and a mixture of small non-occluded RNA viruses (Tinsley, cited by Tiong and Munroe, 1977) the latter almost certainly including a member of the Nudaurelia b group (Entwistle, 1987). Three types of viral particles were identified from infected caterpillars of D. trima collected in Indonesia. They were Nudaurelia b viruses, Picornaviruses and Baculoviruses (Desmier de Chenon et al., 1987).

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.

Integrated Pest Management

Pardede (1992) found that an IPM approach integrating the application of Bacillus thuringiensis, conservation of the natural enemy population (in particular Cantheconidea javana) and collection and destruction of cocoons and adults of D. trima, can rapidly bring an outbreak under control.

Cultural Control

Many cultural practices are modified towards enhancing the effectiveness of indigenous natural enemies (see Biological Control).

Biological Control

Tiong and Munroe (1977) studied the effects of spraying a virus suspension on an outbreak of D. trima in Sarawak. Dead larvae were macerated and strained to produce an inoculum containing a mixture of granulosis virus and small RNA viruses. The suspension was more effective when applied with mistblowers than with knapsack sprayers: more than 77% of the larvae died within 4 days, with almost complete mortality by the 12th day. The cost of using the virus was about 40% of that using chemical treatment; virus suspension appeared to provide a more durable control.

Another approach for utilizing viruses to control leaf-eating caterpillars infesting oil palm in North Sumatra was described by Sipayung et al. (1990). For general field treatment at least 300 g diseased caterpillars are required per hectare. During a pest outbreak a mixture of a sublethal dose of a pyrethroid insecticide (less than 10% of the normal dosage) and the virus is recommended; the pyrethroid increases the susceptibility of the caterpillars to the virus, resulting in faster infection and higher mortality.

Various workers have suggested the conservation or establishment of suitable plants to enhance the numbers and reproductive potential of parasitoids and predators. Syed and Shah (1977) believe that extensive destruction of weeds, either deliberately or inadvertently, is a major indirect cause of pest outbreaks in oil palm in Sabah. They recorded several beneficials visiting the weeds Euphorbia geniculata, E. prunifolium, Ageratum conyzoides and A. mexicanum. They argued for the maintenance of certain weeds, especially the flowering ones, to help support the natural enemy fauna. Desmier de Chenon et al. (1990) has made several recommendations in the same vein; less competitive non-crop plants, such as members of the Euphorbiaceae, should be grown in the borders of plantation blocks to provide sources of food for the parasitoids. In Sumatra larvae of the sawfly Neostromboceros luchti is an important source of food for predators like Eocanthecona furcellata and some reduviids. These sawfly larvae are frequently found feeding on the fern Diplazium asperum which is common in oil palm plantations, and the conservation of this plant should provide an alternative source of food for the predators to tide them over periods when leaf-eating caterpillars of oil palm are scarce. The tachinid Chaetexorista javana with its high fecundity, good searching ability and wide host range can be a very useful parasitoid. However, the adults like to feed on flowers in sunny situations and this feeding is necessary to enable their eggs to mature. The lack of suitable flowering plants in sunny locations should be rectified in oil palm plantations. Pardede (1992) recommends the planting of Pueraria javanica [P. phaseoloides] in young oil palm plantings. In addition to its other beneficial effects, this leguminous cover crop acts as a host for caterpillars of Lamprosema [Omoides] diemenalis, which in turn, is an alternative prey for Cantheconidea javana, an important predator of D. trima.

Chemical Control

Improper use of insecticides has been repeatedly blamed for causing outbreaks through their effect on the natural enemy fauna (Wood, 1971; Syed and Shah, 1977). Rational use of insecticides is achievable through: (1) early detection of infection so that control can be effected when the pest population is still restricted to small areas; (2) monitoring the pest population to time insecticide application when the stage most vulnerable to insecticides predominates; (3) permitting and encouraging the operation of natural mortality factors whenever possible. A decision to implement chemical control should preferably be based on some threshold level (see Field Monitoring/Economic Threshold Levels). Chemical control should be applied when most of the eggs have hatched and while the larvae are in the early stages of development (Wood, 1987). The area sprayed should include the lightly infested field margins.

A wide range of insecticides are effective against limacodids and a comprehensive list is given by Wood (1987). Insecticides chosen should have a selective action so that effective control is achieved with minimal effect on the natural enemies. Selectivity can be achieved through specificity, residual action, and route of uptake. Examples of insecticides that achieve selectivity through specificity and that are known to be effective against limacodids are those based on Bacillus thuringiensis and insect growth regulators. Organophosphates such as trichlorfon and quinalphos have a certain degree of selectivity because of their relatively short residual effect. Regarding route of uptake, systemic insecticides are suitable for the purpose of achieving selectivity in oil palm. As well as its selectivity, trunk injection has a number of advantages (Khoo et al., 1983). Ground spraying using trichlorfon is popular (Wood, 1987; Norman and Basri, 1992) although there are suitable alternatives such as leptophos, quinalphos and aminocarb (Wood et al., 1977). Aerial application, often of trichlorfon, has been used on a number of occasions to control outbreaks of leaf-eating caterpillars in oil palm. In general, aerial application is not a popular option in Malaysia (Norman and Basri, 1992). The pros and cons of this method of application have been discussed by Wood and Nesbit (1969) and Wood et al. (1973b).

Field Monitoring/Economic Threshold Levels

Various systems of field monitoring are practised (Wood, 1976). Wood (1987) recommends a two-stage system. In the 'alert' stage, field workers keep a look-out for frond damage, and if it is detected the area is examined to determine if the damage is caused by an active population of the pest. If so, the 'enumeration' stage proceeds to estimate the number of caterpillars on sampled fronds.

As a example of a system of field monitoring in practice, Hoong and Hoh (1992) carried out 'detection' followed by 'census'. Detection entails monthly monitoring of palms along harvesting paths for presence of pests; in addition workers in the field are encouraged to report the occurrence of pests. Soon after a pest is detected a census is carried out to determine the pest species, its abundance and distribution. Usually, one palm in every ten within a row and one row in every ten is selected; one frond is sampled from each of the upper, middle and lower levels of the canopy. Wood (1987) discussed the practical difficulties of deciding on threshold levels. Nevertheless, levels of 10-12 larvae per frond are set for the larger species and 30-80 larvae for the smaller species. For D. trima, the 'critical pest hazard level' has been arbitrarily set at 10 larvae per frond (Hoong and Hoh, 1992). If this level is exceeded a detailed assessment is immediately carried out to determine the predominant stage and health of the pest so that control can be timed for best results. Control measures are implemented when the number of healthy larvae exceeds the critical level.