牡丹栽培品种数量分类的初步研究

以22个牡丹栽培品种为分类运算单位(OTU),选取5个花粉性状和23个形态性状进行聚类分析和主成分分析,结果表明:数量分类基本可以反映牡丹品种间的亲缘关系,与传统分类法基本吻合;Q型聚类分析能比较理想地将供试品种在Lq1=4.516水平上分为2个大群,在Lq2=3.765水平上分为3个品种群。主成分分析显示:28个性状可综合为7个主成分,其累积贡献率达75.9%,进而选出了影响力比较大的一些性状。

Preliminary Study on Numerical Classification of Peony Cultivars

Taking 22 peony cultivars as OTU,this study performed cluster analysis and principal component analysis on the 5 pollen characters and 23 morphological characters.The results showed that numerical classification could basically reflect the genetic relationship among peony cultivars,which was nearly consistent with the traditional classification.Q-mode cluster analysis divided the observed peony cultivars into two types when the coefficient of genetic distance Lq1 was 4.516,and into three types when the coefficient of genetic distance Lq2 was 3.765.Principal component analysis showed that the 28 characters could be synthesized as 7 principal components,and their accumulative contribution rate reached 75.9%,then some characters with greater force were selected.