Progeny were generated in two spawning seasons (2002 and 2003) in a fully pedigreed population based on the sixth generation of GIFT (Genetically Improved Farmed Tilapia) established in Malaysia. Two lines were created (Selection and Control), and two production environments (cages and ponds) were used to grow out the fish. Live weight at harvest (LW) was the trait under selection, treating the production environments as a fixed effect in the model. In the analysis presented here LW in cages and in ponds was treated as two different traits (LWC and LWP, respectively). Response to selection was estimated for LWC and LWP by three methods by comparing: (i) The least squares means of LW for the Selection and Control lines in the progeny of the 2003 spawning season, (ii) The estimated breeding values for LW between the progeny of the 2002 spawning season and those of the Selected line in the 2003 spawning season, and (iii) The estimated breeding values of the Selection and Control lines in progeny of the 2003 spawning season. For LWC, expressed as a percentage of the overall least squares mean in the population, the estimates were 9.2, 9.6 and 12.7 for methods (i), (ii) and (iii), whereas for LWP they were 8.2, 7.9 and 10.4, respectively. It was concluded that selection response was being achieved in both environments and that there was not enough evidence to justify the conduct of separate genetic improvement programs for cage and pond environments.
Studies on nutrient transport in rivers is still in its infancy in Malaysia. This study investigates the effects of different land-use activities on the transport of nutrients and sediments into the Timah Tasoh Reservoir, northern Malaysia, during the low flood periods (January-June 2001). Nutrient retention and sedimentation were observed in the littoral zone of the reservoir resulting in suspended sediment concentrations (TSS) and nutrients being almost one third lower in the main body of the reservoir compared to the incoming sources. Nitrate concentrations were also reduced by 20%. Mean sedimentation rates in the reservoir were 300 mg day -1 , ranging from 120 mg day -1 in the middle of the reservoir to a maximum of 800 mg day -1 near the littoral zones.
We mainly (but not exclusively) draw upon research and development work carried out by The WorldFish Center (WorldFish). We review the current state of development of selection programs that have had a main focus on growth rate and body traits. There is evidence of sustained gains of 10 to 15 per cent per generation over more than six generation. To date, these gains have not been accompanied by any undesirable correlated response. The prospects for altering sexual dimorphism and the shape of the fish appear to be very limited, however. We also examine the issue of the appropriate environment for selection. Not surprisingly, experimental evidence on genotype by environment interaction suggests that this is more likely to be of importance when the environments in question are markedly dissimilar. We argue that no universal guidelines can be prescribed regarding the need for more than one selection program to cope with different production environments, but that instead, each case should be examined in its own right. Finally, we discuss traits likely to be candidates for inclusion in future, more elaborate, breeding objectives for Nile tilapia, and comment on selection methods that may be implemented in the future.