摘要
利用因子分析法,把70个四川绿豆地方品种的13个主要农艺性状分成受7个独立公因子制约的7个性状群。这7个公因子是;f_1,荚粒数因子,它是由单株英数和单株粒数构成的;f_2,粒重因子,由百粒重、英长和英宽构成;f_3,节、枝数因子,由主茎节数,单株分枝数构成;f_4、f_5和f_6均由一个性状组成,分别称之为单荚粒数因子、株高因子和生育期因子;f_7,产量因子,由单株粒重,播种至开花日数和小区产量构成。受同一因子制约的所有性状称为因子群,文中建议在育种过程中按因子群进行性状选择。试验结果的分析表明,因子分析法是处理多种性状,确定数目较少的起主导作用的因子的一种有用方法,文中介绍了因子分析法的统计学原理和具体计算步骤。
13 agronomic traits of 70 mungbean local variety in Sichuan were divided into 7 independent groups (called factor complexes)by means of Principle Factor Analysis which were controlled by 7 independent factors. The 7 factors which were determined and named were: f_1, pod and grain number factor, consisted of pod/plant and grain/plant; f_2, grain weight factor, constituded by 100-seed weight, pod-length and pod-width; f_3, stem- aode and branch number factor, consisted of stem-nodes and branches/ plant; f_4, f_5 and f_6, formed by one trait, Called grain per pod factor, plant height factor and growth period factor respectively; f_7, yield factor, constituded by yield per plant, sowing-flowering day and yield per plot. The comparative importance of these factor and relationship among tbe traits of the same factor complex were determined. A trait selection procedure which takes a whole factor comp1ex as an independent unit was suggested. Results showed that factor analysis was an effective metbod to clar- ify the relationship among a large number of correlated traits and to reduce them to a smaller number of main factors.
