Evaluation of rice genotypes for grain yield and resistance to bacterial leaf blight (Xanthomonas oryzae pv.oryzae) disease.

Abstract

Rice (Oryza sativa L.) is a staple food crop in many African countries including Tanzania. However, both regional and national rice production have failed to meet demand due to several constraints, among which is the bacterial leaf blight (BLB) disease caused by Xanthomonas oryzae pv. oryzae. Moreover, attempts to increase rice production through the introduction of modern cultivars has motivated farmers to leave local land races for high yielding, but often susceptible varieties. The overall goal of this study was to increase and strengthen rice production in Tanzania through development of high yielding and BLB resistant varieties. The specific objectives were: to i) analyse genotype x environment interaction (GEI) effects for reaction to bacterial leaf blight under natural infection and rice grain yield performance across different environments in Tanzania ;ii) assess the heritability, variability and efficiency of indirect selection using secondary traits for grain yield improvement among rice genotypes; and iii) assess relationship among traits using correlation, path coefficients and genotype-by-trait associations in rice. The study was conducted at three sites namely Katrin, Igurusi and Kyela, all in Tanzania. Thirty rice genotypes, which include two checks, Txd 306 (susceptible check) and IR- 24 (resistant check), were evaluated. The experimental design was a 6 x 5 alpha lattice design with three replications. Data was collected on early vigour, days to early flowering, plant height (cm), panicle length(cm), number of tillers per hill, dead heart, bacterial leaf blight scoring, lodging percent, days to maturity, dry straw weight (kg), spikelets per panicle, grain length (mm), grain width (mm), 1000-grain weight (g), harvest index (%) and yield per plot (kg). Data were analysed using SAS version 9.4 and GenStat 17th edition. ANOVA was used to detect the significance of GEI. The Additive Main Effect and Multiplicative Interaction (AMMI) and the Genotype plus Genotype by Environment Interaction (GGE) biplot models were used for further analysis of GEI and stability. From the results, genotypes NERICA 4 followed by IR-24 were the most resistant to BLB while Supa India was the most susceptible. Dakawa 83 was the most resistant at Katrin while NERICA 4 was the most resistant at Igurusi and Kyela. Genotypes NERICA 2 and LOWLAND NERICA 6 were the most stable across environments for BLB resistance, while IR54 and Txd 306 were the most unstable. Based on the GGE biplot analysis, the three environments fell into two mega environments where as at Kyela, NERICA 4 and IR-24 were identified as the most resistant genotypes while at Katrin Dakawa 83 and NERICA 1 were identified as the most resistant genotypes. Genotype by Environment Interaction effect for grain yield was not significant and as a result, genotype comparison for the same trait was based solely on mean performance across all the environments. The best three genotypes for grain yield were Txd 306, Txd 88 and WITA 10, but in contrast, NERICA 4, Supa India and Mwanza were the worst performers for the same trait. As for broad sense heritability estimates, days to early flowering had the highest estimate of 99.67%, indicating less influence of the environment, while lodging% had 0.00% heritability indicating high influence of the environment. For variability, the phenotypic coefficient of variation (PCV %) values were higher than the genotypic coefficient of variation (GCV %) for all the traits. The highest PCV(%) was for lodging percent (5325.463) followed by number of spikelets per panicles (1005.352)and the lowest was for grain width (1.197) followed by grain length(2.406).The GCV (%) was highest for number of spikelets per panicle (419.902) followed by plant height (97.843) and the lowest was for lodging percent (0.000) followed by grain yield (0.314), genetic advance (GA) was highest for spikelets per panicles (66.79) and lowest for lodging percent (0.000), while for genetic advance as a percentage of mean (GAM %) the highest was for yield per plot (104.13) followed by dry straw weight (92.11) and the lowest was for lodging percent (0.00) followed by panicle length (8.89).Not all the traits under consideration could be used for indirect selection for yield per plot since none of them had a relative selection efficiency equal to or greater than unity. Regarding diversity assessment, cluster analysis based on Euclidian distance indices revealed that Txd 88 and SATO IX were the most similar pair, followed by IR-56 and IR54, which were also similar to each other, and the most divergent genotypes were Txd 306 and Wahiwahi followed by Wahiwahi and Txd 85. Diverse genotypes can be targeted for hybridization since progenies of diverse parents are often more heterotic than those of related parents. The assessment of relationship among traits using correlation and path analysis the traits which were positive and highly significantly correlated to grain yield were harvest index (0.77***) followed by dry straw weight (0.46***), while negative significant correlations were observed for early vigour (-0.22*). Direct effects on grain yield were positive for harvest index (0.80) and dry straw weight (0.51), while indirect effects were highest for days to maturity through harvest index (0.25) followed by number of tillers per hill through harvest index (0.23). For genotype-by-trait associations, genotypes NERICA 1, NERICA 2, NERICA 4, WAB 450-12-12-BL1- and IR-24 were associated with BLB resistance; on the other hand Txd 306, WITA 10, Txd 88, Txd 85, and SATO I were associated with high yield, although Txd 306 was also associated with susceptibility to BLB, whereby WITA 10 was high yield and resistance to bacterial leaf blight. Moreover this study provided information on the presence of genotype by environment interaction in Tanzanian rice growing environment, valuable blight resistance and high yielding genotypes such as WITA 10 and moderate BLB resistance with high yield for genotypes such as Kalalu, Txd (88) and Txd (85), which could be used in rice breeding improvement and conservation efforts of rice.