| In line with the breeding objective of improving rice grain yield, selection criteria for the indirect selection of yield were studied through path analyses of yield-related traits, the partitioning of total nonstructural carbohydrates (TNC) between plant structures, and the nitrogen use efficiency (NUE) of diverse rice genotypes. Fifteen rice genotypes which represented the combinations of low and high levels of four important yield determinants—maximum number of tillers, grain weight, panicle node number, and particle size were used in this study. These genotypes included Lemont, Teqing and 13 inbred lines obtained from a Lemont × Teqing cross. Field experiments were conducted during the 1994 and 1995 cropping seasons at the Texas A&M University Agricultural Research and Extension Center, Beaumont, Texas.;The interrelationships among 14 yield-related traits and grain yield were studied using path analyses. The analyses of the combined 1994 and 1995 data revealed that panicle weight, numbers of spikelets and filled grains per particle, panicle density, maximum tiller density, and 100-grain weight had positive direct effects on grain yield while panicle node number had a negative direct effect on grain yield.;Path analysis revealed that during the early- to late-heading period and the late-heading to grain hardening periods, the changes in stem TNC content had significant direct effects on the changes in panicle TNC content. Significant genotype x developmental stage x plant structure interaction indicated the potential for selection of rice lines with high stem TNC concentration at heading. Linear contrasts indicated that low-grain weight genotypes had higher stem TNC concentration at harvest, which in turn suggested for the selection of high-grain weight genotypes.;Rice genotypes with relatively higher NUE would be beneficial to both rice breeders and farmers. Significant variation in NUE due to genotypes was observed, with the NUE values ranging from 25.3 to 63.9 g GW g N absorbed –1. NUE and grain yield were found to be correlated. The simultaneous improvement of NUE and grain yield cannot be achieved by indirectly selecting for any one of the four traits studied. It would be necessary to select for both NUE and grain yield. |
