1.

The rooting ability of branch cuttings was evaluated for 100 tree species (including 41 families and 78 genera) collected in a tropical rainforest in Sarawak, Malaysia. Leafy cuttings of natural forest saplings were planted in a non-mist propagation system with IBA treatment. During the 6-month experiment, 66 species were rooted with an overall mean rooting percentage of 37.7% (range 0¯100%). Species in the families Dipterocarpaceae and Lauraceae had a low rooting ability, whereas those in Euphorbiaceae, Rubiaceae, and Annonaceae had a high rooting ability. Differences in rooting ability were related to species-specific mature sizes, diameter growth rates and habitat preferences. Species of smaller mature sizes and faster diameter growth rates showed better rooting ability. Species whose forest saplings sprouted more vigorously after experimental felling rooted better than those that showed less vigorous sprouting. Species whose habitats were on lower elevations, concave slopes, and/or clay-rich soils rooted significantly better than those that preferred opposite habitats or habitat generalists that showed no significant habitat preference. The implications of these relations are discussed from the viewpoint of saplings' adaptation to physical damage in their natural habitats.

Indigenous species are replacing exotic fast-growing species in the rehabilitation of degraded tropical rainforests. To properly restore degraded ecosystems, particularly in the moist tropics, a remarkable number of species are required. For example, a tropical rainforest in Sarawak, east Malaysia, contained over 1000 tree species within an area of approximately 50 ha (Condit et al., 2000). Thus, it is important to maintain a regular planting stock of various indigenous tree species to restore the diversity of tropical rainforests. However, the irregular flowering and fruiting habits (i.e., "general flowering"; Appanah and Ashton) and the recalcitrant nature of the seeds of many species (Tompsett, 1987) make seed collection problematic in Southeast Asian rainforests. Collection of wildings has been used to supplement seedling stock, but mortality is high without proper handling (e.g., Palmiotto, 1993).

To overcome the unpredictable supply of seeds, many studies have been conducted investigating the ability of cuttings to propagate vegetatively, and successful rooting has recently been reported for a considerable number of  Southeast Asian rainforest species (reviewed by Dick and Aminah, 1994). However, most of the species chosen are economically important timber trees, including members of the Dipterocarpaceae, which dominate the forests. Little information is available regarding the vegetative propagation of cuttings from species that are economically less important. In order to successfully rehabilitate the diversity of degraded forests, we need more information on the rooting ability of non-timber tree species, especially since the majority of species in tropical rainforests are non-timber species.
 
The study selected 100 study species (41 families and 78 genera) from the LTER plot. Voucher specimens of the species are kept at the Forest Research Center of the Forest Department of Sarawak, Kuching. Branches were collected from saplings of the study species (n=2-10 per species; height 2-6 m) under the closed canopy of the forest around the LTER plot and transferred to the nursery (0.5-1 km away from the collecting sites) in sealed .plastic bags containing 0.2 l water. Branches were collected in June-July 1997, December 1997, and January 1998. Ten leafy cuttings of 2-4 nodes (length~15cm) were prepared for each species. Two to four leaves (depending on leaf size) were left attached to each cutting, and all other leaves were removed. Large leaves were trimmed to approximately 30 cm2 each. We adopted a non-misting cutting propagation method: cutting beds were constructed of cylindrical plastic containers (50 cm in diameter and 25 cm deep) containing water-saturated fine river sand and sealed with transparent plastic. The cutting beds were kept under 50% natural shade for 6 months. Cuttings of the same species were planted in the same cutting bed, thus, there was no replication for each species. Commercial rooting powder (Serbajadi Rooting Powder, Perniagaan Serbajadi, Malaysia; active ingredient IBA) was applied to the base of each cutting. Preparation and planting of cuttings were completed within 8 h from the time of branch collection. The cutting beds were watered 1-3 times during the experiment when the sand appeared to be dry. All cuttings were harvested after 6 months. The number and length of all roots thicker than 1 mm at the base were measured on each cutting.

Of 100 study species, 66 successfully rooted during the experiment from at least one cutting. Rooting percentage of rooting species differed from 10 to 100%. Among the six families including >3 study species, Dipterocarpaceae and Lauraceae had smaller proportions of rooting species: 5/16 and 1/4, respectively. In contrast, the majority of the study species successfully rooted in Euphorbiaceae (16/17), Rubiaceae (5/5) and Annonaceae (5/7) (see Appendix A).

Next page