高梁基礎族群之遺傳研究

Abstract

為了建立一套可供本省蜀黍育種改良的種源，本試驗以9個優良細胞質雄不稔品系為母本及11個經過評估的優良F1雜種為花粉親，進行父母本雜交。然後，每一雜交組合分別取等量種子混合，並在臺中、臺北兩地進行初期族群之逢機交配。經逢機交配後之繼代逢機交配族群，種植於臺北以進行研究探討，並再進行一次逢機交配。最後，成功地建立一套具廣泛遺傳變異、適合本省栽植的基礎族群。由基礎族群各逢機取出50個half-sib(HS)及S1 families，估算各農藝性狀的遺傳變異。各農藝性狀皆無自交弱勢存在，表示這些性狀為遺傳可累加性基因的作用。而GCV則以株高最大。所有農藝性狀其累加性遺傳變方皆大於非累加性變方，因此族群有利於以輪迴選種進行改良。以親裔迴歸法及標準化值法，所估算出HS的遺傳率（狹義）皆小於S1之遺傳率（廣義）。開花期與葉數、綠葉數，莖徑與葉數，穗重與株高•綠葉數皆呈顯著的正相關；而株高與莖徑則呈顯著的負相關。在5×5公分密度之生育環境下，其存活率及SSD成功率為3.5×3.5公分的兩倍，差異極為顯著。以不同農藝性狀選出的9個次族群，在以5×5公分密植未施肥的環境下，以依株高選出的高株及中株族群，其存活率及SSD成功率較高。以在無前作環境的區集下，其存活率及SSD成功率較高，表示蜀黍以SSD法行密植栽培增進世代，若能控制養分的供應則有助於提高其成功率。 To develop a set of germplasm in order to improve plants of sorghum, nine elite cytoplasmic male sterile lines as female parents and eleven superior F1 hybrids as male parents were made cross in this experiment. Equal amount of seeds from each cross were bulked and carried out random- mating in Taipei (TP) and Taichung (TC). The filial random-mating populations derived from the random-mating were planted in Taipei for further experimental research and made random-mating once more. Finally, the base population was developed sucessfully and had broad genetic variation and suitable growth in Taiwan. Fifty half-sib (progeny of open-pollinated male -sterile plants) and S1 (progeny of unbaggedmale- fertile plants) families were randomly chosen from the base population. Genetic variability of agronomic characters were estimated. All of the agronomic characters did not exit inbreeding depression in this experiment, indicating they were all due to additive genetic action. Plant height had the highest estimation of coefficient of variation value. The estimation of additive variance was greater than non-additive variance for all traits in the experiment, therefore it is better to use recurrent selection for population improvement. Heritabilities for S families (broad sense) were slightly higher than for half-sib families (narrow sense) by using parent- offspring regression and standard unit methods. Positive correlation were found between days to flower and leaf number, stalk diameter and leaf or green leaf number, panicle weight and plant height or green leaf number, while that of between plant height and stalk diameter, was negative. Survival rate and SSD succeed rate for 5 X 5 cm density were two-fold larger than that of 3.5 X 3.5 cm. Nine sub-populations selected for different agronomic characters were planted under environment of 5 X 5 cm density without fertilizer application. Survival rate and SSD succeed rate were higher for high-plant and medium-highplant sub-populations selected for plant height. Survival rate and SSD succeed rate in the block without previous crops were higher than that in the block with previous crops of sorghum or corn. This experiment indicated that the SSD succeedrate could be higher if one can control the supplement of nutrition.