Paramasius distortus (Curculionidae) is morphologically similar to M. hemipterus, and both insects are common visitors to fresh trunk wounds and leaf cuts in oil palm. P. distortus and R. palmarum also frequent traps baited with sugar cane and the male aggregation pheromone of M. hemipterus.

In Central America, R. palmarum is commonly found to be contaminated with the red ring/little leaf nematode, and reduction of the adult population of this insect through the use of traps has been associated with a significant drop in the incidence of this disease in oil palm (Morales and Chinchilla 1990; Oehlschlager et al. 1992; Chinchilla et al.1992). A similar response has been observed in coconut plantations (Santos et al. 1998).

However, within the plantation under study, there were areas in which the disease reduction rate was lower than the plantation average. It was then important to determine if this situation was partly due to the presence of some other vector whose population remained high.

Besides R. palmarum, no other insect has been found to be a vector of the nematode B. cocophilus in Central America in spite of a systematic search over the years, specially in M. hemipterus and P. distortus (Chinchilla et al. 1996). However, it is possible that the proportion of the insects infected with the nematode is so low that it does not show up in routine samples. Still, even a very low contamination rate could be important, considering the high populations that these insects reach in oil palm plantations.

The relative importance of M. hemipterus as a potential vector of the Red Ring/Little Leaf nematode could be evaluated, then, in an indirect manner, by measuring the effect that the reduction of the insect population through the use of traps had on disease incidence.

The objective of this work was to compare, on a commercial scale, the incidence of the Red ring/Little Leaf syndrome in two areas, in which two densities of traps were set in an attempt to reduce the M. hemipterus population.

Materials and Methods
The study was carried out on a commercial oil palm plantation belonging to the Palma Tica company, located on the southern Pacific coast of Costa Rica. The area has an average annual precipitation of 31.8 °C and 21.2 °C respectively, and a relatively dry season running from January to March.

The M. hemipterus adults were caught in traps made of plastic recipients (10 l) containing carbofuran-treated sugar cane as well as the species male aggregation pheromone made up of a mixture of methyl-nonanol and methyl-heptanol (Peres et al.1997). This pheromone (Meta-Lure®) and the Rhynko-Lure® (the aggregation pheromone of R. palmarum were obtained from ASD de Costa Rica, S.A. (Apdo. 30-1000 San José, Costa Rica, ph. 506-257-2666).

The traps were set during the first week of May, 1995, and capture data were collected every two weeks until the middle of April, 1996. Three areas were chosen, in which one of three treatments was used. The first treatment consisted of 36 traps, baited only with the M. hemipterus pheromone and distributed over an area of 36 hectares (1 trap/ha). The second treatment was placed in an area of 26 ha, and consisted of a combination of traps baited only with the M. hemipterus pheromone (27 traps), only with the R. palmarum pheromone (7 traps), and a combination of both pheromones in the same trap (5 traps). Thus, there was a density of 1.23 traps/ha with Meta-Lure®, and 0.46 traps/ha with Rhynko-Lure®. A third area (159 ha) was used as a control treatment, in which a commercial trap for R. palmarum with a density of one Rhynko-Lure® trap every five hectares was used. The first two areas had had, since 1992, commercial R. palmarum traps (Rhynko-Lure®) at a density equal to that in area three. These traps were removed before the new traps were placed.

The bi-weekly captures for each trap (M. hemipterus, P. distortus y R. palmarum) were set out to dry on absorbent paper for a period of 24 hours. The initial idea was to count each insect from the traps individually, but this turned out to be impractical in the case of M. hemipterus and P. distortus due to the high number of captures. In order to simplify the process, captures were estimated according to the volume of insects. The number of insects was estimated using a regression curve previously established between the number of insects and the volume taken up, and was expressed as an average capture per trap, as the number of traps was different in each area. Red Ring incidence was expressed as the number of diseased in 142 palms.

During the period August, 1994 to August, 1995, a sample was taken from each insect population in search of individual carriers of the nematode B. cocophilus. Each month a group of insects was collected from each species (P. distorus and M. hemipterus), and they were macerated separately and placed in a modified Baermann funnel. The search for nematodes in R. Palmarum was carried out by cutting each insect individually and placing it in a Baermann funnel.

Results and discussion
Relation between number of insects and volume occupied
The regression curve (y=4.519x - 6.0458, R = 0.95, n= 295, x = volume in ml, y = number of insects)describes the relation between the total number of individuals (M. hemipterus and P. distortus) and the volume occupied (Fig. 1). An average of 71% of the captures corresponded to the former species, with the other 29% corresponding to the latter. This relation was fairly consistent throughout the period of study.

M. hemipterus and P. distortus populations
The average monthly number of captures of these insects per trap in the area with one Meta-Lure® trap per hectare was statistically equal to that found in the area with a density of 1.5 traps/ha. After a year of trapping at these two densities, there did not seem to be a tendency towards a reduction in these insect populations in the areas under study, which suggests that a greater trap density was necessary to achieve a significant impact on the population. The situation was different with R. palmarum, in which it was found that, depending on the initial insect population, a significant reduction could be achieved by placing a trap every one to five hectares. An experimental trap density of 6 traps/ha reduced the population by half in only four months (Oehlschlager et al. 1992). On the commercial plantation, an average density of 0.2 Rhynko-Lure® traps per hectare was used, and a population reduction of more than 95% was achieved within a few years.

Seasonal variation
A seasonal variation was observed in the number of M. hemipterus and P. distortus captures, similar to that of R. palmarum in other studies in oil palm in the area (Morales and Chinchilla 1990).

The number of captures fell with the beginning of the rainy season, and showed a tendency to increase toward the end of this period, reaching a peak in maximum captures during the dry season (Fig. 2). In the period of highest precipitation, between the months of May and October (accumulations of 2,822 mm), captures were 538 insects per month per trap in the first area (Meta-Lure® traps) and 488 in the second area (Meta-Lure® and Rhynko-Lure® traps). In the period of lowest precipitation, between November and April (842 mm), average captures per month per trap were 1,280 and 1,150 insects in the first and second areas, respectively.

In the first area, average monthly captures per trap of M. hemipterus and P. distortus varied from 177 insects (at the beginning of the rainy season in May) to 2,087 (for January, during the dry season). In the second area, similar variation was observed (189 in May and 1,768 in January). The correlation between amount of rain and captures per trap per month was negative (r= -0.76, P = 0.004) in area one and -0.80 (P=0.02) in area two. The average R. palmarum captures were very low throughout the period of study (fewer than two insects per trap every other week), which was attributed to the intensive trapping program that had been carried out against this insect on the plantation since the end of 1993.

Boscán and Godoy (1988) found that the M. hemipterus population visiting traps (banana pseudu stem) located on a commercial plantation went down during the periods of highest precipitation.

Red Ring/Little Leaf incidence
No correlation was observed to exist between M. hemipterus and P. distortus populations captured and disease incidence in either of the two areas with M. hemipterus traps. In area two, for example, taking into account a three-month incubation period, the correlation between monthly M. hemipterus captures and Red Ring incidence yielded an "r" value of -0.22 (P=0.6). In area one, the correlation was positive (r=0.7), but it reached a low level of significance (P=0.08). During the first half of the year, no treatment effect was expected, given that the incubation period of the disease is at least three months, and the traps were set in May. In the period between August, 1995 and April, 1996 (Fig. 3.), when treatment effects were expected, it was found that the disease incidence decreased 36% in area one and 13% in area two (with a higher density of Meta-Lure® traps). Incidence did not vary greatly in the control area (Rhynko-Lure® traps).

During one part of the period of study, the disease incidence reduction was lower in the area with the greatest M. hemipterus trap density. Nevertheless, neither of the two trap densities used consistently reduced the population of this insect, which showed little change throughout the period of study.

In the entire study area (including the control area, without M. hemipterus traps), the disease continued the tendency to decrease begun in 1993 when the general trapping program with the R. palmarum male aggregation pheromone was begun. In the period prior to treatment establishment, as well as during the period of study, R. palmarum captures had decreased to less than two per trap per two-week period. Initial captures, which were associated with a high disease incidence, varied greatly among traps, but were generally higher than 30 insects per trap per week.

Over a period of a year, monthly sample were taken and a total of 3,523 M. hemipterus adults, 912 P. distortus adults, and 284 R. palmarum adults were analyzed. No carriers of the nematode B. cocophilus were found among the first two species. However, approximately 13% of the R. palmarum were found to be carriers. This type of study has been repeated on various occasions in the area of study with similar results: R. palmarum normally turns out to be contaminated, while other insects, including M. hemipterus turn up nematode-free.

The results of these experiments show once again that R. palmarum is possibly the only important vector in the Red Ring/Little Leaf syndrome in Costa Rica and possibly in Central America. M. hemipterus seems to be associated neither with the nematode B. cocophilus nor with the disease.

Bibliography
Boscan, N.; Godoy, F. (1988). Epocas de incidencia de Cosmopolitus sordidus G. y Metamasius hemipterus L. en dos huertos de musaceas en el estado de Aragua. Agronomía Tropical (Venezuela), 38 (4/6): 107-119.

Calvache, H. (1994). Acción de Metamasius hemipterus L. en la transmisión del anillo rojo de la palma aceitera. Palmas (Colombia), 15(4): 17-22.

Chinchilla, C.Ml.; Oehlschlager, C.; Bulgarelli, J. (1996). A pheromone based trapping system for Rhynchophorus palmarum and Metamasius hemipterus. ASD Oil Palm Papers, 12: 11-17.

Chinchilla, C. Ml.; Oehlschlager, A. C.; González, L. M. (1993). Management of the red ring disease in oil palm through pheromone-based trapping of Rhynchophorus palmarum L. 1993 PORIM International Palm Oil Conference. Palm Oil Research Institute of Malaysia, Kuala Lumpur. 19p.

Esser, R.; Meredith, J. (1987). Red Ring Nematode. Nematology. Circular 141. Fla. Agric. Serv. Division of Plant Industry.

Griffth, R. (1968). The mechanism of transmision of the red ring nematode. J. of the Agric. Soc. of Trin. and Tob. 67:436-457.

Mora, L.; Calvache, H.; Avila, M. (1994). Diseminación de Radinaphelenchus cocophilus, agente causal del anillo rojo de la palma aceitera en San Carlos de Guaroa. Palmas (Colombia). 15(1): 15-27.

Morales, J.; Chinchilla, C. Ml.; (1990). Picudo de la palma aceitera y la enfermedad del anillo rojo en una plantación comercial de palma aceitera en Costa Rica. Turrialba (Costa Rica), 40(4): 470-485.

Oehlschlager, A.C.; Chinchilla, C Ml.; González, L. (1992). Manejo del picudo de la palma (R. palmarum) y la enfermedad del anillo rojo, mediante un sistema de trampeo basado en la feromona de agregación . ASD Oil Palm Papers (Costa Rica), 5: 24-30.

Oehlschlager, A.C.; Chinchilla, C. Ml.; González, L.; Jirón, L.; Mexón, R.; Morgan, B. (1993). Development of a pheromone-based trapping system for R. palmarum. J. of Econ. Entomol., 86(5): 1381-1392.

Perez, A.L.; Campos,Y.; Chinchilla, C.Ml.; Oehlschlager, A. C.; Gries,G.; Gries, R.; Giblin-Davis, R.; Castrillo, G.; Peña, J.; Duncan, R.; González, L.; Pierce, H.; McDonald, R.; Andrade, R. (1997). Aggregation pheromones and host kairomones of West Indian sugarcane weevil, Metamasius hemipterus sericeus. Journal of Chemical Ecology, 23(4): 869-888.

Santos, O.; Prado, I; Cibrián, J. 1998. Manejo del mayate prieto, Rhynchophorus palmarum y la enfermedad del anillo rojo de la palma de coco en Guerrero. In. Memorias del XXXIII Congreso Nl. de Entomología. México. P. 564-566.