The Plant-Pollinator Community in a Lowland Dipterocarp Forest

Abstract

Tropical pollination biology at the community level in forests was studied first in the Neotropics (Bawa et al. 1985; Kress and Beach 1994). In a tropical rain forest in La Selva, Costa Rica, medium-sized to large bees and small diverse insects are the main pollinators in the canopy, while hummingbirds and euglossine bees are prevalent in the forest understory (Janzen 1971a; Stiles 1978; Endress 1994; Kress and Beach 1994; Rinco ´n et al. 1999). In West Malaysia, however, plant-pollinator communities are expected to be different from those in the Neotropics, because plant reproductive phenology, fauna, and flora differ greatly between the two areas. The phenomenon known as general flowering, or GF, occurs inWest Malaysia and, as might be expected, this has consequences for the coevolutionary processes between plants and pollinators. More than 80% of the emergent and canopy tree species bloom in short periods of three to four months at irregular intervals, usually of 2 to 10 years (Ashton et al. 1988; Appanah 1993). During the remainder of the time, often for several years, both floral resources and pollinators become relatively rare. Therefore, pollinator shortages might occur unless there is a rapid response to the general flowering with population growth and adult activity (Ashton et al. 1988). According to these authors, thrips are capable of such a response. Thrips maintain a low population density using floral resources in gaps during generally flowerless seasons, and they have a short generation time of around two weeks (Appanah and Chan 1981) and high fecundity. Thus, when a general flowering starts, they can increase in numbers quickly by using the massive floral resources. However, thrip pollination of diptercarps is only known for the genus Shorea, section Mutica in the Malay Peninsula (Appanah and Chan 1981). Are many trees that bloom in GF pollinated by thrips, or are there other types of pollinators that can quickly respond to the general flowering? This is the first question that we address. Appanah (1990) provided one clue to the answer. Carpenter bees (Xylocopa) shift foraging areas in GF from forest edges, much like the thrips, to the inside of closed forests. However, we question whether such shifts of foraging areas could provide sufficient pollinator populations. Bawa (1990) stated that long-distance pollen flow is intensified in species-rich tropical rain forests, because conspecific plants are spatially isolated from each other. Hummingbirds and euglossine bees are among the most important long-distance pollinators in the Neotropics (Kress and Beach 1994), but they are absent in Southeast Asia. From La Selva, Costa Rica, 1287 species of wild flowering plants have been recorded (Hartshorn and Hammel 1994). The exact number of plant species in Lambir, Sarawak is unknown, but even when restricted to trees (1 cm dbh) found in a 52 ha plot, over 1173 species have been recognized, and it is likely that the total number of flowering plants exceeds 2000. In and around the Canopy Biology Plot (8 ha), 999 species of flowering plants have been collected (Nagamasu and Momose 1997). Thus, species richness is very high, and conspecific plants are considered to be spatially isolated from each other. It would be expected that long-distance pollinators have important roles in the species-rich lowland dipterocarp forest. If so, what types of long-distance–specific pollinators are there? This is our second question. Momose et al. (1998c) collected or observed flower visitors of 270 plant species of 73 families (see Appendix A). Based on that study, we describe pollination syndromes, describe the plant-pollinator community in this lowland dipterocarp forest, and try to answer the above questions. Figs (Ficus spp., Moraceae) are not included, and their pollination is discussed in Chapter 10.