| dc.description.abstract | Rice, one of the world’s staple is under serious threat by water stress as a result of drought due to the changing climatic conditions. Yield reduction due to drought is increasing by the day and severe losses have been incurred by farmers over the years. Breeding work on especially upland rice for drought tolerance has been met with challenges arising mainly from the complexity of mechanisms involved in inheritance of drought resistance. Nevertheless, several attempts have been made to improve rice for drought tolerance and varying levels of success have been made. In the case of Uganda, rice improvement is a new focus, although limited improvement of rice cultivars for various traits has been done. To broaden the genetic base by the National Crops Resources Research Institute (NaCRRI), interspecific rice lines have been introduced from the International Centre for Tropical Agriculture (CIAT) and the Africa Rice Centre formerly known as West Africa Rice Development Association (WARDA). However, the performance of these lines under Ugandan conditions is not clearly known. The overall objective of this study was to determine the mode of gene action for the various traits used in breeding for drought tolerance and specifically to: i) determine the reaction of parents to water stress; ii) determine the combining ability and heritability for drought tolerance in upland rice progeny derived from selected parental crosses; and iii) evaluate previously identified rice quantitative trait loci (QTL) for drought tolerance in F2 segregating population. The study was carried out at NaCRRI located at 0o32l N and 32o37l E. Ten inter-specific upland parent genotypes selected from several rice lines acquired from Africa Rice center and CIAT were used. These parents were planted in soil boxes following a randomised complete block design to screen them for tolerance to water stress. The parents were also planted in pots and crossed following the North Carolina mating design II to generate F2 segregating progeny. Five tolerant parents were used as female and five susceptible parents used as males in the forward cross. Reciprocals too were included in the study. The F2 progeny along with parents were planted following an alpha lattice design with two replications in soil boxes in the screen house. Water stress was imposed by restricting watering at maximum tillering and data were collected on various traits (leaf rolling, leaf drying, plant height, number of tillers and leaf recovery following rehydration) during the vegetative and reproductive stages (days to flowering, panicle exertion, panicle length, number of spikelets per panicle, panicle production per plant, number of effective tillers and grain yield) when the soil moisture content was 30 %. Re-watering was done at the end of the stress period until plant harvesting. Data were used to determine the reaction of genotypes to water stress, the general and specific combining ability as well as heritability of the different traits in GENSTAT. Genotype 188 (WAB 881-10-37-18-3-P1-HB) was most tolerant to water stress among parents while genotypes 46 (CT16353-CA-11-M) and 41 (CT163339(1)-CA-15-M) were most susceptible. The progenies that resulted from crosses involving genotypes 26x46 and 26x41 were most tolerant to drought among the F2 populations manifested by their ability to form grain under severe water stress compared to other progenies. Based on combining ability and heritability analyses, days to flowering were found to have a high broad sense heritability and additive genetic control in action. Both additive and non additive gene action have been found important in the regulation of the number of spikelets per panicle produced. Parent genotypes 188 (WAB 881-10-37-18-3-P1-HB) and 138 (WAB450-1-BL1-136-HB) were found to have good general combining ability for the number of spikelets per panicle and could be used to improve rice yields under drought conditions Parental screening using 27 SSR markers associated with drought tolerance traits showed that marker RM339 and RM224 were polymorphic in parents 188 and 46, and thus were used to screen the F2 segregating progeny. Previous work had located RM224 and RM339 on linkage group 11 and 8 respectively. The marker RM339 was shown to be associated with leaf drying, while the marker RM224 was shown to be associated with all the traits studied based on the chi square test of independence, although only tillering had significant association from the regression analysis. The results showed that RM224 was linked to tillering and the analysis showed that 34% of the variation observed in tillering of the genotypes was associated with the SSR marker RM224 segregation. | en_US |
