The efficiency of a very simple trap to capture adults of O. cassina was tested in several experiments conducted in a commercial oil palm plantation. The trap is formed by a disposable plastic bag and a food bait. Adults may enter the trap but can not abandon it once inside.

The use of these traps allowed to conduct studies on the life cycle of the insect and particularly to follow the dynamics of the adult population. The traps have become an important part of an integrated management of the pest. When traps are placed timely, a large proportion of females still carrying eggs are captured. These traps along with the promotion of the activity of biological control agents, and a sound agronomic management of the plantation may keep the attacks of O. cassina under control. This trap is now used commercially in Costa Rica.

Introduction

Opsiphanes cassina is an important defoliator of the oil palm in Tropical America (Genty et al . 1978, Chinchilla 1993, Mexzón y Chinchilla 1996). Some important attacks have occurred in Costa Rica in plantations located on the Pacific coast. Counting larvae feeding on leaf number 17 has served as a guide to start a chemical intervention trying to lower such population (Rhainds et al . 1993). Goods results have been obtained applying commercial products based on Bacillus thuringiensis . An additional measure taken has been to promote the growth of some plant species that harbor some of the natural enemies of the pest.

The adult population of the insect can be attracted and killed in poisoned food baits. A common trap uses ripe fruits (bananas, melons, pineapple, sugar cane etc.) poisoned with an insecticide such as Lannate. These traps are placed within a container or placed directly on the ground. Females visit these traps looking for sugars and nitrogen compounds needed for egg development.

Bait distribution of such traps in the field is expensive, and poisons used are a threat to wild life, including parasitoids of important pests, and even to humans, particularly children that may feel attracted by the fruits when still fresh. Besides this, most adults abandon the trap after feeding and die in the nearby vegetation where they are eaten by ants and other animals, which make impossible to document if the females had or not lay the eggs, and the total number of insects eliminated.

This paper describes several field experiments done to optimize a trap to capture the adults of O. cassina . The trap was used to determine the period of adult emergence and to know other important aspects of the dynamic of the adult population.

Materials and methods

Fig. 1. Trap model to capture Opsiphane cassina adults

Trap design. The basic design consists of a transparent plastic bag (100 x 60 cm), which comes from the lining of the fertilizer bags. The bait is placed internally and the mouth of the bag is rolled downwards so it can be hung from a leaf base on the palm. A properly placed trap (mouth collapsed leaving a narrow entrance) will allow the butterflies to enter but not to leave since they have a zigzag type of fly which makes them to hit the walls and eventually tire and die. Butterflies may crawl inside or let them fall by bending their wings (Fig. 1). The original idea about such type of trap came from observations made by a field worker.

Trial I 

The basic trap design was compared with a modification consisting in placing a waist in the bag. A hollow piece of a bamboo stem (about 10 cm long) was placed inside the bag and hold on place by wrapping it from the outside with a string. The mouth of the bag was kept partially open by placing inside an oval-shaped wire. 

During the first trial, holes were made at the bottom of the bags to drain excess liquids, however it was realized that such holes negatively affected trap efficiency since many butterflies did not enter the trap, but preferred to feed from liquids oozing out the trap.

Six different food baits were tested in both types of traps (the basic and modified designs) which were used at two densities (4 and 8 traps/ha).

Trials were conducted during the months of September and October 1998 (rainy season) right after an outbreak of the pest. Traps were placed approximately two weeks after first adults had emerged from pupae.

Baits 

Baits tested were: 

1. sugar cane 
2. sugar cane with molasses (100 ml)
3. sugar cane with an insecticide 
4. sugar cane, molasses and insecticide 
5. molasses with yeast 
6. molasses with yeast and insecticide

Sugar canes were cut longitudinally and approximately one kilogram of pieces (15-20 cm long) was used per trap. For treatments with insecticide, the bait was submerged during three hours in Carbaryl (Sevin 80), 5g/l.

Molasses (100 ml) were throughout mixed with 15 g of yeast (Red Start™ ) and put in a plastic cup. Special care was taken so the yeast would not form a crust on the surface. The cup was placed inside the trap (plastic bag). Treatments with molasses and insecticide were shaken for two minutes before being placed in the cups or applied on the sugar cane. All baits were changed every other week.

Fig 2. Larva, male and pupa of Opsiphanes cassina

Treatments were arranged as a split plot within a complete randomized design. Large plots were two densities (4 and 8 traps/ha), and small plots were trap type and baits. The area covered by traps placed at a density of 4/ha was 12 ha, and that covered by the other density was 8 ha. Traps were placed at approximately 1.6 m height on the palm stems.

Trapped insects were counted every 2-3 days. Males were separated from females due to its smaller size, darker color, tapering abdomen (round in females), presence of two yellow spots on both sides of the abdomen, and finally because males have a brush of hairs on the anterior wings (Fig 2). The presence of eggs in the females was documented. Larval and eggs counts on leaf number 17 were taken as complementary data. Statistical analysis were done only for data taken the first 13 days, period during which the largest captures were obtained.

Trial II

The basic design of traps was used without the drainage holes. Five baits were tested: 

1. sugar canes with molasses (30 ml of molasses and 1 kg sugarcane) 
2. sugar cane with diluted molasses (1l. molasses in 10 l. water where canes were soaked for one hour) 
3. sugar cane with molasses and 5 g of yeast 
4. sugar cane with yeast 
5. molasses with yeast (30 ml molasses with 5g yeast mixed and placed in a plastic cup 

Treatments were arranged in a CRBD with 12 replications in an area of 15 ha.

During the first cycle of adult emergence, a density of four traps per hectare was used, which was reduced to a half for the next cycle, when the only bait used was molasses with yeast.

All experiments were carried out in a commercial oil palm plantation (Deli x AVROS, 16 years old), located on the South Pacific coast of Costa Rica, in an area that had suffered from recurrent attacks by O. cassina . 

Results and discussion

Trap design and trap density

The basic design of trap captured significatively more insects (48412 vs. 41913, P=0.05) than the modified trap (with a waist and a oval-shaped wire at the mouth of the bag). Many butterflies were able to escape from the modified trap, but not from the basic design.

Captures were significatively increased when a higher density of traps was used (8 vs. 4 traps/ha). Any particular density of traps to be used will depend on the size of the population expected based on field observation of last instar larval density and parasitism. Field experience and data from other experiments indicate that trap density can be reduced down to two traps per hectare if good control was achieved during the previous generation of adults.

Baits

Baits were significatively different ( Fig. 3 and Fig. 4 ). The use of an insecticide may not be necessary and may even reduce captures; the insecticide with sugar cane and molasses may have had a repellent effect. The best bait tested was the mixture of molasses and yeast. This bait has the advantage of being of easy preparation and the ingredients can be easily found. Females find in yeast a source of energy and amino acids needed for egg development. 

The condition of the traps must be closely monitored. Baits must be kept moist and should be changed every 8-10 days, other wise they lose effectiveness. It is also important to change any damaged bags. The mouth of the bags may close completely and prevent the butterflies from entering, particularly during periods of heavy rains. The mixture of molasses with yeast needs moisture for the components to ferment and being more attractive. During dry periods, this mixture hardens and lose effectiveness.

There are still many ways to improve this trap. Some volatiles from ripen fruits could substitute the fruits themselves. Actually, and after these experiments were finished, the trap was even improved by placing the yeast and molasses within a half-liter plastic botle. Volatiles from the bait may be released through small holes made in the upper part of the bottle. The capped bottled is hung with a string from a leaf base first and then it is placed within the plastic bag. This arrangement allows the replacement of the plastic bag when necessary without the need to discard the bait, which may last for the whole period of adult emergence.

Dynamics of the adult population of the insect

Adults of O. cassina are active for about 7-10 days. During this period they look for food, a mate, and in the case of the females, a place to lay their eggs. The pattern of visits to the traps is illustrated in figure 5 and figure 6 . It is apparent from these data that adult emergence is not simultaneous in the whole area.

The total period of adult emergence from pupae took about a month, and followed a normal distribution (peak of emergence at about half the period). This pattern was very clear during the second trial when traps were placed before the very first adults emerged from pupae ( Fig. 7 ). During the first trial, when traps were placed late, the second part of the cycle of emergence is basically observed ( Fig. 8 ).

The first two weeks of adult emergence are the critical period when traps should be already placed in the field. During this period a high proportion (85-100%) of females captured still had not lay all their eggs ( Fig. 9 ). This percentage is reduced to 75% after the peak of captures passes. The fact that during the first weeks males predominates ( Fig. 10 ), may increase the chances of a female to be fertilized. Another hypothesis to be tested is that males may need a longer period to sexually mature.

After some time, the proportion of males and females gets close to 1:1 and may even change in favor of females. This behavior was observed in the first trial. The smaller proportion of females with eggs visiting the traps during the second portion of the emergence cycle can not be explained in terms of late visits since the longevity of the adult is rather short.

The period between two consecutive cycles of adult emergence was approximately a month: final of one cycle and beginning of the next ( Fig. 7 ). This gives an average of two months for the larval stage. During this period between cycles it is important to follow the development of the larval stage and pupation so the traps can be timely placed. A few strategically placed traps may be let permanently in the field to detect the emergence of the first adults.

Trap captures and larval and eggs counts on the leaves.

The number of larvae and eggs on the leaves was inversely related to the number of captures in traps placed in the same area ( Fig. 11 and Fig. 12 ). A larger number of eggs was found on the leaves within the first two weeks of adult emergence. A larval peak was observed about four weeks after traps were placed. To reduce the number of eggs being lay, the traps must be placed before adults start to emerge and lay eggs, and the best bait and trap density must be used. All these will drastically reduce the size of the next generation of larvae.

The rapid decline of larvae and eggs on the leaves ( Fig. 11and Fig.12 ) is probably due to depredation and natural parasitism suffered by these first stages in the life cycle of the insect. The period of egg eclosion in O. cassina is between 5 and 15 days (Genty et al . 1978).

About two months after larval counts were initiated, the population fell drastically on leaf 17. This was an indication that the natural enemies of the pest were being very effective regulating the population and that the previous adult trapping had reduce the population to levels where this was possible.

References

Chinchilla, C. 1993. Fauna perjudicial en palma aceitera. ASD de Costa Rica, Costa Rica, pp. 13-20.

Genty, P.; Desmier de Chenon, R.; Morin, J.P. 1978. Las plagas de la palma aceitera en América Latina. Oléaginéux (número especial). 33(7): 350-351.

Mexzón, R.; Chinchilla, C. 1996. Enemigos naturales de los artrópodos perjudiciales en la palma aceitera en América Central. ASD Oil Palm Papers. 13: 9-33.

Rhainds, M.; Chinchilla, C.; Gries, G. 1993. Desarrollo de un método de muestreo para larvas de Opsiphanes cassina Feldes en palma aceitera. Manejo Integrado de Plagas (Costa Rica). 30:15-18.