应用计算机系统对87份油菜种质资源的14个性状进行了因子和聚类分析。结果表明，14个性状的变异系数值从6.02% (含油量)到96.87%(硫苷)含量，变异范围较大，表明供试材料的遗传多样性是显著的。通过相关分析表明，主花絮有效角果数是1个比较重要的性状，它与多数产量构成因子呈显著或极显著正相关，与株高、一次有效分枝部位、主花序有效长度等株型性状呈极显著正相关。采用主成分分析将14个性状归纳为6个主成分，它们占总遗传变量的80.43%。将87份材料的14个性状值标准化处理后进行聚类分析，结果将87份油菜种质资源分为5个类群，而根据4个品质性状指标进行聚类，将87份油菜种质资源分为6个类群。 Using computer systems, the factor and cluster analysis was done to 14 quality traits of 87 rapeseed germplasm resources. The results showed that the coefficient of variability of 14 agronomic traits ranged greatly from 6.02% (oil content) to 96.87% (glucosinolate content), which indicated the diversity of materials was significant. The correlation analysis shows that the effective silique of main inflorescence is an important trait. It is in significant or highly significant positive correlation with most factors of yield components, and it is in significant positive correlation with plant height, an effective branch height, effective length of main inflorescence and other plant traits. Adopting the principal component analysis, 14 traits were divided into 6 principal components, which accounted for 80.43% of total genetic variations. After 14 quality traits of 87 materials have been standardized, through cluster analysis, all materials were classified into 5 groups, but 6 groups according to 4 quality traits.
References
[1]
Anderberym, R. (1972) Cluster analysis for application. Academic Press, Waltham, 2-235.
