Sumary

A marked reduction in the number of male inflorescences in anthesis per area, during the dry season in a young oil palm plantation, was associated with a reduction in the population of the pollinator E. kamerunicus . Consequently, pollination of female inflorescences in anthesis was very deficient, which caused a marked drop in fruit set and an increase in bunch failure values 5-6 months later. This sequence of events probably also occurred during the previous dry season, which caused a pronounced increase in bunch failure during the months of August and September. Rainfall levels two years previously were associated with a change in sex ratio: more abundant and better distributed rain caused an increase in sex ratio (more female inflorescences). The associated effect was a large reduction in the number of male inflorescences in anthesis two years later. The inverse situation (smaller sex ratio: a larger proportion of male inflorescences), occurred when rainfall levels and rain distribution was rather poor two years before the anthesis period. The 1994 dry season had these characteristics, and therefore, during the dry season of 1996 there was a fair number of male inflorescences in anthesis, accompanied by a large and active population of pollinators, which guaranteed a good fruit set 5-6 months later. The rather severe problems of fruit set (and a high proportion of bunch failure) affecting this young oil palm plantation were most probably caused by a particular combination of soils and climate that favored a very high sex ratio, which caused a drastic drop in male inflorescences and pollinator population during some periods of the year. The relative isolation of these plots from other adult plantations prevented the wind from alleviating the severe shortage of available pollen during some periods. Over time, the fruit set problems have gradually disappeared as palms age, sex ratio is smaller and palm height increases, which means that wind plays a more important role as a pollen carrier.

Introduction

The oil palm is pollinated by several species of the genus Elaeidobious in the Curculionidae family (Syed 1979). However, as the palms gain height, wind may play a more important role as a pollen carrier (Hardon and Turner 1967, Syed 1979).

Two insects coexisted in Costa Rica as the main pollinators of the oil palm ( Elaeidobius subvittatus and Mystrops costaricensis ). Due to some apparent pollination problems during some parts of the year, E. kamerunius was introduced in 1985, and it is now the most prevalent pollinator throughout Central America (Syed 1885, Chinchilla 1991). E. kamerunicus seems to guarantee suitable levels of pollination in all commercial plantations where palms of different ages coexist. The situation may differ in some isolated new plantations, where adequate pollination may become difficult if the initial sex ratio is too high in some commercial new genotypes. The lack of male inflorescences in anthesis has a drastically negative effect on the population of E. kamerunicus , since this weevil only completes its life cycle on male oil-palm inflorescences (Syed 1978, 1982).

A young oil palm plantation was established in southern Costa Rica, in an area relatively isolated from other adult plantations, between 1989 and 1991. Initially, it was noted that sex ratio was exceptionally high, and in 1994, fruit set was abnormally low, and many bunches failed to develop to maturity. A sample of 1421 bunches harvested in mid August showed that only 26% could be considered normal, and the rest had a low fruit set (30% or less). Many of the bunches had no fruits at all (bunch failure).

This type of situation, also described by Turner (1981), clearly indicated a poor pollination of the female flowers while in anthesis, which occured about five and a half months before the bunch was ready to be harvested. In a "normal" bunch, nearly 60% of the flowers are expected to be pollinated and develop as normal fruitlets. Bunch failure may result from a poorly pollinated female inflorescence.

This study documents the relationships observed between the abundance of male inflorescences in anthesis, pollinator population, fruit set, and the amount and distribution of rainfall in a young oil palm plantation in Costa Rica.

Materials and methods

The study was conducted in two commercial plots planted in 1989 and 1991 with Deli x AVROS crosses. These plots form part of a new project (approximately 500 hectares in 1994), which initially developed in relative isolation from other adult plantations. The main nucleus of adult plantations is located about 60 km away.

This area has a well defined dry season, mean temperatures of 27.5°C (maximum: 32.3°C; minimum: 22.8°C), and an average of 8.7 hours of sunshine a day (maximum: 9.5; minimum: 7.5). Estimates of pollinator population and fruit set were obtained following the methods described in Syed (1986) and Chinchilla and Richardson (1990). Fruit set, number of weevils per male inflorescence, and number of male inflorescences were determined once a month. Sampling was done in approximately 490 randomly selected palms in each of two plots.

Weevil population was determined during anthesis in 10 male inflorescences randomly selected from the sample of palms where fruit set was obtained. Sampling was done between 5 and 8 in the morning, when 9 spikelets (3 each from the basal, medium and apical portions of the inflorescences) were cut and saved in order to later count the insects.

Fruit set was estimated in 12 randomly selected ripe bunches (before any loose fruit was detected). The spikelets were separated from the stalk, leaving a sample of 15-25 spikelets from where fruitlets were classified as normal, partenocarpic or failures (undeveloped white fruitlets).

Results and discussion

Seasonal fluctuation in the number of male inflorescences

In both plots studied, the number of male inflorescences in anthesis dropped gradually between October 1994 and January 1995. In September 1995, an average of 6.1 and 20.7 inflorescences/ha was observed in each of the plots ( Fig. 1 ). No inflorescences were seen during February and March in the sampling area, and only a few were found in April. During the following nine months, the number varied between approximately one and 27 ( Fig. 1 ).

The number of male inflorescences in anthesis was again rather low during the dry season of 1995, but increased considerably in the following dry period. This was considered to be the effect of a previous water deficit that had lowered the sex ratio.

Rainfall is the main source of water for this plantation. However, during the dry season, some water may infiltrate from a large river (Térraba) that flows around part of the plantation. A large mound protects the plantings from overflows of this river. Subterranean movement of water from the river to the plantation is facilitated by the medium and light textures of these alluvial soils.

To better understand the effect of rains on sex ratio, we must consider not only total amount of rainfall to November (which determines the amount of water that can be stored by the soil as a reserve), but also the contribution of scattered rain showers during the otherwise dry season (months between December and March) ( Table 1 ).

Water reserves in the soil were high during the dry season of 1992-93 as a result of high precipitations during all the previous rainy season (424 mm only in November) and moderate rains (total of 344 mm) between December 1992 and March 1993 (97, 111, 71 and 65 mm in December, January, February and March 1993) ( Fig. 2) . During this dry period it was noted that the water level of the Terraba river did not recede much.

The natural conditions created by a good aeration of these light-textured alluvial soils and the good water availability during the dry season were extremely favorable for the plants, which responded with a abnormally high sex ratio. The final result of this was an extremely low number of male flowers in anthesis two years later (basically February and March 1995).

The situation in 1993 was quite different. Only 100 mm of rainfall were recorded in November, and 176 mm during the first months of the following year, which were also poorly distributed (110 mm in December, 5.9 mm in January, 20 mm in February and 40 mm in March) ( Fig. 2 ). The low rainfall had the effect of reducing the water level of the Terraba river, which in turn reduced the amount of water available through lateral infiltration. This rather severe dry season lowered the sex ratio, which caused a marked reduction in the number of female inflorescences in anthesis two years later (February and March of 1996).

An additional stress factor that affects sex ratio in favor of masculinity is poor soil aeration caused by excessive rains, as has been observed in the main oil palm production area in southern Costa Rica (Chinchilla and Richardson 1991). It is believed that the high precipitations during 1993 (particularly during August and September), caused a reduction in the sex ratio. In October 1993, the mean number of sunshine hours per day was just 4.4. These conditions ( Fig. 2 ) are associated with an increase in the number of male inflorescences two years later (October 1995).

In order to better understand the relationships between the amount of rainfall and sex ratio, male inflorescences were counted weekly between January and June 1996 ( Table 2 ), and the information was compared with the rainfall during the various preceding weeks. However, this effort was unsuccessful, since there was little or no rainfall over a period of several weeks, which generated many zeros and prevented us from establishing any clear statistical correlation.

Broeckmans (1975) described a similar situation to the one observed in this study. He noticed a close relationship between rainfall during the dry season and the sex ratio two years later. It is generally accepted that previous situations of stress result in an increase in the number of palms that subsequently produce male inflorescences (Hardon and Corley 1976). In Nigeria, it was found that the maximum number of male inflorescences were produced about 20 months after a severe dry season (Hardon 1973), and abnormally high peaks of male inflorescences were observed in Malaysia, 19-21 months after periods of water stress (Corley 1973).

Elaeidobius kamerunicus population

The adults of E. kamerunicus lay their eggs in the male inflorescences of the oil palm, where the developing larvae find shelter and food (Syed 1978).

The number of E. kamerunicus per spikelet increased from October 1994 to January 1995, but no insects were observed between February and June of this last year. In 1996, March was the only month in which no insects were found ( Fig. 3 ).

The fluctuation in the number of male inflorescences in anthesis is what causes major variations in weevil population in commercial plantations. However, it should be noted that during some months, although male inflorescences in anthesis were found, these were completely free of E. kamerunicus (for example in April and May 1995). This indicates that when the weevil population drops below a certain level, it needs at least a month to recover and to show up again in the samplings.

The apparent increase in the number of weevils per spikelet found between October 1994 and January 1995 could be the result of more individuals being concentrated in a reduced number of inflorescences. This concentration of insects in a few inflorescences may be negative for the species, since it may lead to an increase in the mortality rate and to a greater risk of disease transmission (Syed 1978).

The opposite situation (an increase in the total weevil population per area, but a reduction in the number of weevils per spikelet), was associated with a larger number of male inflorescences in anthesis per hectare from April 1995.

Pollination and fruit set

Fruit set at harvest time was lowest in August and September 1995 ( Fig. 4 ), a response associated with a reduction in the pollinating weevil population that occurred about five months earlier (when male inflorescence in anthesis per area was very low). August and September was also a period with a high proportion of bunch failure.

During 1996 the phenomenon repeated itself, when the weevil population and male flowers in anthesis were both very low during the dry season ( Fig. 4 ). These events were once again associated with a drop in fruit set, which in one of the lots studied reached nearly 48% between September and October. The decrease in fruit set was not so pronounced in the other lot, since the male inflorescence population did not drop too much during the dry season (maintained an average of 10 inflorescences/ha).

A reasonably high fruit set 5-6 months after a decline in the pollinating weevil population may indicate that wind plays a more important role as a pollen carrier as the palms became taller.

Fruit set problems in these young plantings have become less severe as the palms have grown older. This is thought to be the result of a combination of lower sex ratios (more male inflorescences) and taller plants that facilitate wind pollination.

References

Broekmans, A. 1957. Growth, flowering and yield of the oil palm in Nigeria. J. of Waifor, 2(7):187-220.

Chinchilla, C.; Richardson, D. 1991. Pollinating insects and the pollination of oil palm in Central America. ASD Oil Palm Papers, Costa Rica. 1-18.

Corley, R. 1973. Oil palm physiology. In Advances in oil palm cultivation. The Incorporated Society of Planters, Kuala Lumpur, Malaysia. Oxford University Press. 469 p.

Hardon, J. 1973. Assisted pollination in oil palm: a review. In Advances in oil palm cultivation. The Incorporated Society of Planters, Kuala Lumpur, Malaysia. Oxford University Press. p 184-195.

Hardon, J.; Corley, R. 1976. Pollination. In Oil Palm Research. Elsevier Scientific Publishing Company, Amsterdam, The Netherlands. p 300-305.

Syed, R. 1978. Studies on pollination of oil palm en West Africa and Malaysia. Report of The Comm. Inter. Biological Control, CAB. 38 p.

Syed, R. 1984. Los insectos polinizadores de la palma africana. Palmas (Colombia), 5:19-64.

Syed, R. 1985. Report on visit to Costa Rica and Honduras for Compañía Bananera de Costa Rica and Tela Railroad Company of Honduras. Harrisons Fleming Advisory Services Ltd. 9 p.

Turner. P. 1981. Oil Palm Diseases and Disorders. The Incorporated Society of Planters, Kuala Lumpur, Malaysia. Oxford University Press. 280 p.