Csiro Publishing

Whether to develop inbred cultivars or F1 hybrids is a dilemma faced by many rice (Oryza sativa L.) breeders. This could be partially answered if one could select superior recombinant inbred lines with an equal yielding ability and good quality traits from commercial F1 hybrids. Thus, it was attempted in this study to select superior inbred lines from 2 commercial F1 hybrids after application of honeycomb selection and panicle-to-row selection. The 2 F2 populations were advanced to F6 generation by both methods and, finally, 5 F5:6 lines with high yield potential and good grain quality were selected from each population and selection method and were tested in a randomised complete block design for 2 years in Kalochori, Thessaloniki, Greece. In each case the respective F1 hybrid and the cultivar Strymonas were used as checks. Fourteen of the 20 lines selected by honeycomb selection and panicle-to-row selection from both populations exhibited a yielding ability that was not significantly different from the yield of the F1 hybrids in both years. Three of them, however, in 1 of the 2 years, had a significantly higher grain yield than the corresponding F1 hybrid. In addition, 6 of the above lines exhibited significantly higher values for more than 1 of the 4 quality traits (total milling yield, grain vitreosity, grain length, and grain length/width ratio) and they were not inferior for the remaining ones. It was concluded that application of combined selection for yield and quality could lead to the isolation of recombinant inbred lines with equal yielding ability and quality equal to or higher than the F1 hybrids. This, together with the higher cost associated with hybrid technology, indicates that the long-term goal of a rice-breeding program should be the production of superior inbred lines, unless hybrid production cost is low and quality is not critical for the particular market. Additional keywords: Oryza sativa, recombinant inbred lines.