以河南4个主要野生冬凌草〔Isodon rubescens (Hemsl.) H. Hara〕种群的77株冬凌草为研究对象,采用方差分析、变异分析、相关性分析、主成分分析以及聚类分析法对12个表型性状进行比较,并分析这些表型性状与环境因子的关系。结果表明:野生冬凌草12个表型性状的F值均达到极显著水平(P＜0.01),其中,叶柄长宽比的F值最大,叶面积和叶形指数的F值较大。4个种群12个表型性状的变异系数范围是0.62%～24.50%,均值为9.93%,其中,一级分枝数和叶干质量变异系数的均值较大,均达到18%以上。野生冬凌草叶片、茎部位的表型性状与环境因子间大多呈显著(P＜0.05)或极显著相关；经度、年均温、年降水量、最冷月最低温、土壤pH值和土壤类型与大多数表型性状呈显著或极显著负相关,而纬度、海拔和最热月最高温与大多数表型性状呈显著或极显著正相关。前4个主成分的累计贡献率为90.686%,其中,第1主成分中叶长、叶宽、叶面积、叶周长和叶鲜质量以及第2主成分中叶形指数的特征向量绝对值均大于0.8。聚类分析结果表明:在欧氏距离25处,4个种群的个体被分为2组,商城县和宜阳县种群个体聚为一组,卢氏县和济源市种群个体聚为另一组。综上所述,4个野生冬凌草种群存在较为丰富的表型变异,纬度、经度、年均温、年降水量、最冷月最低温和土壤类型对野生冬凌草的生长有显著影响,叶长、叶宽、叶面积、叶周长、叶形指数、叶鲜质量可作为评价种质资源的主要指标,卢氏县和济源市的野生冬凌草可作为优良种质资源的选育材料。

Taking 77 Isodon rubescens (Hemsl.) H. Hara individuals from four major wild I. rubescens populations in Henan Province as research objects, 12 phenotypic traits were compared by methods of variance analysis, variation analysis, correlation analysis, principal component analysis, and cluster analysis, and the relationships between these phenotypic traits and environmental factors were analyzed. The results show that the F values of 12 phenotypic traits of wild I. rubescens all reach extremely significant levels (P＜0.01), among which, the F value of petiole lengthwidth ratio is the largest, while those of leaf area and leaf shape index are relatively large. The range of coefficients of variation of 12 phenotypic traits in four populations is 0.62%-24.50%, with an average of 9.93%, among which, the averages of coefficients of variation of number of primary branches and leaf dry mass are the largest, both exceeding 18%. Phenotypic traits of leaves and stems of wild I. rubescens mostly show significant (P＜0.05) or extremely significant correlations with environmental factors; longitude, annual mean temperature, annual precipitation, min temperature of the coldest month, soil pH value, and soil type show significant or extremely significant negative correlations with most phenotypic traits, while latitude, altitude, and max temperature of the warmest month show significant or extremely significant positive correlations with most phenotypic traits. The cumulative contribution rate of the first four principal components reaches 90.686%, among which, the absolute values of eigenvectors of leaf length, leaf width, leaf area, leaf perimeter, and leaf fresh mass in the first principal component and leaf shape index in the second principal component are all greater than 0.8. The cluster analysis result shows that the individuals from four populations are divided into two groups at a Euclidean distance of 25, with individuals from Shangcheng County and Yiyang County populations are clustered into one group, and individuals from Lushi County and Jiyuan City populations are clustered into the other group. In conclusion, there are relatively rich phenotypic variations among the four wild I. rubescens populations, and latitude, longitude, annual mean temperature, annual precipitation, min temperature of the coldest month, and soil type have significant effects on the growth of wild I. rubescens, leaf length, leaf width, leaf area, leaf perimeter, leaf shape index, and leaf fresh mass could be used as main indexes for germplasm resource evaluation, and wild I. rubescens from Lushi County and Jiyuan City could be used as breeding materials for superior germplasm resources.