蜀葵品种的数量分类研究

为了探究在蜀葵品种分类中起到关键作用的指标以及建立全面的蜀葵品种分类系统，以75个蜀葵品种为研究材料，对各品种的性状进行观测，选定26个性状指标进行R型聚类分析和主成分分析，去掉相关性较大且累积贡献率较小的性状指标，然后对75个蜀葵品种进行Q型聚类分析。R 型聚类分析结果显示：除了少数指标相关性较大外，其他指标较为分散，指标选取合理。主成分分析的结果显示： 26个指标可综合为8个主成分，累计贡献率达82.818%；Q型聚类分析的结果显示：75个品种在不同等级结合线处可以被分为2类、3类、4类和5类，其中叶型、裂叶长/叶长、花型、雌雄蕊瓣化情况、花期早晚、花瓣数、花色等指标在分类中有较大的影响力。叶型和花型是蜀葵品种分类最主要的标准和依据。因此，可以将蜀葵品种划分为圆叶品种群、裂叶品种群及全裂叶品种群三大品种群。

Study on numerical classification of Alcea rosea varieties

In order to explore the indicators that play a key role in the classification of Alcea rosea varieties and establish a comprehensive classification system of Alcea rosea varieties, 75 Alcea rosea varieties were used as research materials, the traits of each variety were observed, and 26 trait indexes were selected for R-type cluster analysis and principal component analysis. The trait indexes with large correlation and small cumulative contribution rate were removed, and then Q-type cluster analysis was carried out on 75 Alcea rosea varieties. The results of R-type cluster analysis showed that except for a few indexes with high correlation, other indexes were scattered and selected reasonably. The results of principal component analysis showed that the 26 indicators could be integrated into 8 principal components, with a cumulative contribution rate of 82.818%. The results of Q-type cluster analysis showed that 75 varieties could be divided into 2, 3, 4 and 5 categories at different levels of binding lines. Among them, the leaf type, leaf length/leaf length, flower type, pistil and stamen petalization, flowering time, petal number, flower color and other indicators had great influences in the classification. Flower type, leaf type and flower color were the most important criteria and basis for the classification of Alcea rosea varieties. Therefore, this experimental results indicated that the varieties of Alcea rosea can be divided into three groups: round leaf group, split leaf group and palmatisect leaf group.