台灣南部第二期作水稻育種之研究（2）　水稻第一、二期作選拔效應之分析

Studies on the Breeding of the High Yielding Variety for Second Crop Rice in Southern Taiwan (2) Study on the Effect of Planting Season on the Selection Efficiency

林富雄(Fu-Hsiung Lin) ； 邱運全(Yun-Chian Chiu)

《花蓮區農業改良場研究彙報》 2期 (1986/03) Pp. 21-22

摘要

本試驗利用本省及引進之16個水稻品種來探討不同栽培期作下對產量之選拔效率，試驗結果第二期作各性狀之平均值均小於第一期作，且其變異性較大。抽穗日數、株高及千粒重之遺傳變方佔表現型變方之大部份，故其遺傳率估值高。除一株總乾物重，株高及收穫指數外，其餘性狀之遺傳率估值均以第二期作低於第一期作。遺傳相關分析結果谷產量與一穗結實粒數及總乾物重之遺傳相關均高，但在第二期作抽穗日數與谷產量遺傳相關亦高而穗數則與谷產量有負相關。選拔效率結果顯示對各性狀之直接選拔效果第一、二期作均相似，而直接選拔對產量之相關反應則以抽穗日數，株高、一穗結實粒數及一株總乾物重較大。第二期作時選拔較晚熟，粒數多及一株總乾物重大之個體對谷產量有利。各性狀在第一及第二期作種植時之遺傳相關除收穫指數較小外其餘均大。行第一期作之間接選拔以改進第二期作產量效果不大，宜進行不同期作之直接選拔。

並列摘要

Sixteen rice varieties, which randomly chosen from local and introduced varieties, were used to evaluate the effect of planting season on the relative selection efficiency for yield and various major agronomic traits. The mean values of various traits were smaller in the second crop than the first crop and most of them showed the greater coefficient of variation in the second crop. Genetic variances were much larger than environmental variances in the traits of days to heading, plant height and 1000-grain weight which resulted in higher heritabilities. Generally, the heritability estimates were higher in the first crop season. Genetic correlations between yield and filled grains per panicle, and between yield and total dry weight were high in the both crop seasons. The high positive and negative genetic correlations were also observed between grain yield and days to heading, and between yield and panicle number per plant for the seond crop, respectively. Direct responses to selection for various traits were similar in the both growing seasons. Correlated responses to selection for yield were large in the selections of days to heading, filled grains per panicle, plant height and total dry weight. Indirect selections for yield based on days to heading, filled grains per panicle and total dry weight in the second crop were larger than direct selection. Genetic correlations between two cultural environments for various traits were close to unity except for harvest index and the relative selection efficiencies for the second crop season based on various traits in the first crop were small, suggesting that direct selection in the first or second crop was necessary to develop high yield genotype for each crop season.