Variation in tree species floristic composition and local-scale distribution in tropical forests is thought to be associated with habitat. This study compared the spatial distribution pattern and spatial genetic structure of Shorea parvifolia spp. parvifolia in a lowland and a hill dipterocarp forest and postulated how the spatial structure of a tropical tree species may be influenced by habitat heterogeneity. The significant spatial aggregation of S. parvifolia spp. parvifolia in both the lowland and hill dipterocarp forests indicated that seed dispersal was limited. The higher degree of habitat heterogeneity (e.g. topography, habitat type and resource availability) in hill dipterocarp forests compared with lowland dipterocarp forests, as established by previous research, may have contributed to the higher level of aggregation of S. parvifolia spp. parvifolia in the hill dipterocarp forest in the present study. The significant spatial genetic structure observed in the lowland dipterocarp forest may be due to restricted pollen and seed dispersal. In contrast, the absence of significant spatial genetic structure in the hill dipterocarp forest may indicate extensive pollen flow, and consequently, energetic and generalist pollinators (e.g. social bees Apis spp. and Trigona spp.) may be involved in the pollination of dipterocarps in the hill dipterocarp forests of Peninsular Malaysia.

Modelling based on paternity analysis of one hill and three lowland dipterocarp timber species in Malaysia indicated that more pollen donors sired progeny in Shorea leprosula and S. parvifolia than in S. curtisii and S. maxwelliana populations, because they start reproducing earlier. The relatively greater male fecundity of small S. leprosula and S. parvifolia trees also contributed to more rapid recoveries of outcrossing pollination following reductions in selective logging simulations. Furthermore, pollination by these trees significantly increases numbers of adult trees that contribute to mating as males, and hence maintenance of higher genetic diversity in S. leprosula and S. parvifolia populations. The variation in male fecundity of the dipterocarp species might be linked to variations in their turnover rates, wood density and (hence) optimal logging regimes. More specifically, species with low wood density (S. leprosula and S. parvifolia) grow faster, start reproducing at smaller sizes and can be harvested with the current selective logging protocol with minimal loss of diversity. In contrast, species with high wood density (S. curtisii and S. maxwelliana) grow more slowly, start reproducing at larger sizes and should be harvested with a less intensive protocol to maintain outcrossing pollination rates and genetic diversity in subsequent generations. The results showed that selective logging guidelines should be customised for specific groups of dipterocarp timber species according to their reproductive parameters and wood density.

The in-house production of fluorescently labelled internal size standard offers the advantage of cost saving over the commercial size standard in microsatellite genotyping. Based on the reported in-house internal size standard protocol, we have improved the method by generating 21 DNA fragments (in a standard named as HM-400) with each size similar to that of the commercial size standard. The consistent amplification of the correct fragment size was optimised via primer modulation for non-templated nucleotide addition by Taq DNA polymerase. A total of six microsatellite loci were used to assess the accuracy of HM-400 and the mean standard deviation of the size data was 0.19. The differences between the fragment size means for samples sized using HM-400 and commercial size standard were small with an average of 0.29 bp. The production cost of HM-400 was only 10% of the cost of commercial size standard.

This paper reports the isolation and characterization of 24 polymorphic microsatellite markers in an important tropical timber species, Koompassia malaccensis (Leguminosae). The primers were designed from a genomic library enriched for dinucleotide (CT) repeats and screened on 24 samples from a natural population. The number of alleles detected per locus ranged from two to 13 while the observed heterozygosity ranged from 0.042 to 1.000. Significant departure from Hardy-Weinberg equilibrium (P < 0.05) was detected in two loci. These microsatellite markers were tested across 13 timber species of the same family. The amplification success appeared to be associated with taxonomy classification at the genus but not subfamily levels.

This study reports the isolation and characterization of 15 polymorphic microsatellite loci in an endangered tropical tree Hopea bilitonensis (Dipterocarpaceae). The primers were designed from a genomic library enriched for dinucleotide (CT) repeats and screened on 24 adult trees from a natural population. The number of alleles ranged from two to eight, and the observed heterozygosity ranged from 0.042 to 1.000. These loci will allow mating system and population analyses to be carried out in this species.

Genetic diversity study of Koompassia malaccensis based on 19 populations from 18 forest reserves located throughout Peninsular Malaysia was reported. The average number of samples analyzed per population was 24. Genetic diversity assessment was based on six polymorphic microsatellites. Overall, all the populations showed high levels of genetic diversity. The allelic richness ranged from 6.0 (Pekan) to 9.3 (Lenggur) whereas the gene diversity ranged from 0.683 (Pekan) to 0.859 (Lenggur). The estimated coefficient of population differentiation (Rst) was 0.07, implying that 93% of the genetic diversity was partitioned within population, with only 7% distributed among populations. Further analysis using more microsatellites will be carried out.

In this study, we report on the development of expressed sequence tags-derived simple sequence repeats (EST-SSRs) for Shorea leprosula, an important timber species in Malaysia. A total of 3000 EST sequences were generated from cDNA libraries representing the inner bark tissue of S. leprosula. Homologue searches were performed using blast search against non redundant databases (blastn and blastx). The search for SSRs in S. leprosula EST sequences revealed 154 clones of EST sequences consisting of di- and tri-nucleotide repeat motifs, with ≥6 bases. Redundancy analysis of the 154 sequences revealed 135 ESTs belonging to 63 clusters and 19 ESTs being singletons. Hence, a total of 82 unique sequences (2.7% of all EST sequences investigated) represented potential candidates for EST-SSR marker development. Primer design was possible only for 24 out of the 82 sequences. This was basically due to the sequences contained either too little DNA sequence flanking the SSR or the sequences were inappropriate for designing primer. The screening on 24 primer pairs flanking the di- and tri- SSRs were subsequently fluorescently end-labeled and assayed to detect polymorphisms in the 24 adult individuals from Pasoh Forest Reserve. Out of the 24 primers screened, 15 primers were found to be polymorphic, five primers were found to be monomorphic and the remaining four primers were found to be none specific. For polymorphic loci, the number of alleles detected varied from 2 to 17. This might indicate that in terms of levels of polymorphism, EST-SSRs are comparable to non coding SSRs. The conserved nature of the flanking regions also indicate that EST- SSRs can be suitable markers for cross-species and even cross-generic amplifications in other dipterocarps. Hence, the EST-SSRs developed from this study will provide a valuable source of new polymorphic markers which can be used for genetic studies and functional biodiversity assessments of the Dipterocarpaceae.

A population genetic investigation of Dipterocarpaceae species from the Pasoh Forest Reserve (FR), a lowland dipterocarp forest, and the Sg. Lalang FR, a hill dipterocarp forest was carried out. Two species with contrasting breeding systems (Shorea leprosula a diploid and S. ovalis ssp. sericea an autotetraploid) from the 50-ha plot in Pasoh were studied. Three species with different habitat preferences (S. curtisii - habitat specific: dominates the ridges, S. leprosula - habitat specific: dominates the valleys and S. macroptera - habitat generalist: common on the ridges as well as in the valleys) from the 33 ha plot in Sg. Lalang were studied. Levels of genetic diversity were estimated for all the studied species from both populations, and patterns of tree distribution as well as spatial genetic structure were evaluated for several diameter classes in each population. The results suggested that all the studied species maintain relatively high levels of genetic diversity, and that the strongest evidence for spatial clumping of individuals and spatial autocorrelation of alleles occurs among trees of the smallest diameter class in lowland dipterocarp forest. Such structure is weaker or non-existent among larger trees in lowland dipterocarp forest. For hill dipterocarp forest, the spatial genetic structure analyses revealed significant spatial genetic structure only at large-diameter trees of all the three species. Given the high degree of environmental heterogeneity in hill dipterocarp forest, for habitat specific species ( S. curtisii and S. leprosula), the spatial genetic structure at the large-diameter trees could be explained by intense selection in favour of certain genotypes at the small-diameter trees. As a result, seedlings with suitable genotypes for a specific habitat will survive and subsequently become adults. The lack of spatial genetic structure but significant aggregation at the small-diameter trees of all the three species might indicate extensive pollen flow. Hence, if seed dispersal is restricted but pollen flow is extensive, no spatial genetic structure but significant spatial aggregation will be observed at the small-diameter trees, regardless of whether the species is habitat specific or habitat generalist. The inferred extensive pollen flow also allows us to postulate that energetic and generalist pollinators might be involved in the pollination of dipterocarps in the hill dipterocarp forests. Twenty polymorphic microsatellite markers were developed