| Abstract |
The research work 'Breeding for shattering resistance in rice (Oryza sativa L.)' was conducted in the Department of Plant Breeding and Genetics, College of Horticulture (COH), Vellanikkara, during the academic year 2017-2018. The research work was mainly divided into three experiments. Experiment 1 consisted of screening of rice genotypes for shattering resistance which was conducted at Agricultural Research Station (ARS), Mannuthy. After screening, four selected shattering resistant rice genotypes were crossed in Line x Tester pattern with three shattering prone high yielding varieties in experiment 2. In experiment 3 progenies were evaluated along with parents for shattering resistance. Shattering was measured based on Induced Random Impact method using a force gauge apparatus. Wide variability was found to exist among twenty five genotypes for yield and most yield attributes studied indicating ample scope for improvement through selection. High heritability coupled with high genetic advance as per cent of mean indicating the influence of additive gene action in the expression of trait were observed for days to fifty per cent flowering, flag leaf width, flag leaf length, panicle per plant, seed yield per plant and shattering per cent. Substantial improvement in the expression of these characters over base population can be expected through simple selection. Tillers per plant showed moderate heritability along with high genetic gain implying influence of both additive and non additive gene action in the expression of these characters. Improvement of these traits could be attained by following recurrent or reciprocal recurrent selection to exploit both additive and non-additive genetic components. Seed yield per plant recorded high significant correlation with number of tillers per plant, flag leaf width, number of panicles per plant and seeds per panicle and significant correlation with test weight. Shattering per cent recorded a high significant inter correlation with panicle length and significant inter correlation with kernel length. High estimates of general combining ability (OCA) over specific combining ability (SCA) is indicative of preponderance of additive gene action and it was evident in the case of days to maturity, flag leaf width, seeds per panicle and kernel width. Prevalence of non-additive gene action i.e., dominance and epistatic gene action in trait expression indicated by high SCA over OCA was registered for the remaining characters. When parents were evaluated on the basis of mean performance, it was evident that for seed yield per plant L2, L3 and T1 recorded a high response. The response recorded for panicle length and shattering per cent were exactly similar. When all the fourteen characters for the seven parents considered together, L2, L3 and T2 represented as best parents for yield and yield attributes and T1 recorded to be a moderate response. Evaluation of hybrids based on mean performance revealed that out of the twelve hybrids, H1, H2, H3, H4, H6 and H11 recorded a high response towards reduced shattering and H5 reported moderate response. Among the twelve hybrids H1 (L1 x T1), H2 (L1 x T2), H3 (L1 x T3), H5 (L2 x T1), H6 (L2 x T2) and H11 (L3 x T3) showed a high total response compared to the rest. Results from gca effects of parents indicated that L1 and T3 to be better combiner for reduced seed shattering indicating a scope for further utilization of these lines in plant breeding programmes for reduced shattering. When all the fourteen characters were considered, L2, T1, T2 and T3 recorded as better combiner. All the seven parents recorded to be moderate combiner for tillers per plant and flag leaf width. Scoring based on mean performance and combining ability effects for seed yield per plant revealed that, L2, L3 and T2 proved to be promising. T1, T2 and T3 proved to be promising parents for reduced seed shattering and T4 and Ll with a moderate response. When all the fourteen characters considered together L2, T1 and T2 were recorded as most promising parents.Specific combining ability studied among the cross combinations indicated that out of the twelve hybrids, H4, H5, H6 and H11 recorded high response and HI exhibited a moderate response for seed shattering. When the fourteen characters considered together, H3, H4, H5, H6, H8, H10 and H11 recorded to be better cross combination among the twelve characters. When mean performance and sea effect for all the fourteen characters for hybrids considered, the hybrids H1, H2, H3, H5, H6, and H11 reported high scores and H8 recorded moderate response. All the twelve hybrids recorded high significant heterobeltiosis for plant height, flag leaf width, panicles per plant, panicle length, test weight, kernel width, kernel length and shattering per cent. Based on the scoring for both mean performance and heterotic effect revealed that hybrids H5 recorded highest score followed by H3, H9, H11, and H1. Evaluation of hybrids based on mean performance, sea effects and heterosis revealed that hybrids H5, H6 and H3 recorded the first three positions among the twelve hybrids evaluated. Among this H5 and H6 have both the parents as better combiners. The remaining cross combination involving different combinations of parents viz., good x good, average x average, average x poor, poor x good, poor x average and poor x poor etc., can be used for transgressive breeding since there will be better recombinations in segregating generation. |
