论文部分内容阅读
对中国收集编目的４２５１份芝麻种质资源材料，依据来源、品种类别、生态类型３层分成１４个组。应用计算机系统对随机小样本的１４个性状数据进行聚类分析，选择确定了离差平方和法是芝麻资源聚类的适宜方法。通过组内聚类分析并结合随机选取和预选比例２０％的原则，预选出核心收集品材料８８４份，对其１４个性状的６个特征值的初步研究结果表明，基本代表了总收集品的遗传变异范围。对预选核心收集品在中国３个生态试验点两年的农艺性状鉴定表明：６个质量性状的５３０４性状次与原评价资料之间完全一致的为９７．８５％，３个数量性状呈良好的显著正相关。对决选的核心收集品代表性测验表明，１４个性状的平均值符合率达９７％以上，标准差符合率在９０％以上，表明核心收集品良好地代表了预选核心收集品。
4251 copies of sesame germplasm resources collected from China were classified into 14 groups according to source, species and ecotype. The cluster analysis of 14 traits data of random small samples was made by computer system. The method of determining the sum of squares of variance was the suitable method for the clustering of sesame resources. According to the clustering analysis within the group and the principle of random selection and pre-selection of 20%, 884 core collection materials were pre-selected. The preliminary results of 6 eigenvalues of 14 traits showed that the total collection Range of genetic variation. The identification of the agronomic traits of two preselected core collections at three ecologic experiment sites in China showed that the traits of 5304 traits of the six quality traits were 97.85% identical to the original evaluation data, and the three quantitative traits were good Significantly positive correlation. Representative tests of the core collection of the selected candidates showed that the average of the 14 traits met the criteria of over 97% and the standard deviation of more than 90%, indicating that the core collection represents a good pre-selected core collection.