陆地棉主要品种的遗传距离估测与聚类分析

ESTIMATION OF THE GENETIC DISTANCE AND CLUSTER ANALYSIS ON THE MAIN VARIETIES OF UPLAND COTTON

本文通过主成分分析,对40个常用陆地棉亲本进行了遗传距离分析,并采用最短、最长、类平均及重心法等对供试材料进行了系统聚类。主成分分析表明,以产量,重量与比强度、纤维细度、纤维成熟度及衣分为代表的5个主成分累计贡献率可达到91°。以上聚类分析证明,40个供试品种可分5类,最长距离聚类法比较适宜,亲本起源与血缘关系及类型划分无必然联系。亲本必须根据性状表现及主成分值综合选择后,再根据遗传距离进行选择,不宜过分强调遗传距离远(如本研究之Ⅰ/Ⅳ类间杂交)。根据主成分值选配的9个理想亲本为:鲁棉六号、鲁棉四号、邢台6871、岱字16号、PD0113、冀棉八号、冀合3016、徐州514及鄂荆92。

The genetic distance of 40 main varieties in upland cotton has been estimated, through the techniques of analysis on special components. Then, using the clusting analysis methods of the shortest, longest, mean of cluster and the gravity of cluster, clusted to the experiment material. Analysis of the special components demonstrated: devotion percent of the five components, which represented as fiber yield, weight and fiber strength, fiber fineness, fiber maturity and lint percent, can exceed 91%. Cluster analysis illustrated: 40 varieties used in the experiment can be divided into 5 clusters. The longest cluster method is better than others. There is not a closing connection between varietis origin and desent system with cluster. Selection of the parents must be according to the special components firstly, then to the genetic distance. Longer of the genetic distance can not be specially pursued(for example, crosses between Ⅰ/Ⅳ). According to the special components, 9 better parents have been selected. They are LM-6, LM-4, XT-6871, Dlt-16, PD0113, JM-8, JH-3016, XZ-514 and EJ-92.