广宁红花油茶优树综合评价及指标筛选

Comprehensive evaluation and index screening of excellent plants of Camellia semiserrata

【目的】研究广宁红花油茶Camellia semiserrata经济性状遗传变异特征,筛选核心评价指标,建立优树评价预测模型,为广宁红花油茶种质资源创制利用提供科学依据。【方法】以广宁红花油茶核心产区初选的31株优树为研究对象,采用多元统计方法对其冠幅面积、单株产果量和单株产油量等20个经济性状指标进行测定分析。【结果】参试株优树的20个性状存在不同程度的变异,变异范围为3.25%~69.47%;各指标间共有38对达到极显著水平(P<0.01),15对达到显著水平(P<0.05)。主成分分析将11个评价指标简化为4个较为独立的综合指标,累计贡献率达80.23%,保留了供试材料各性状的大部分信息。根据4个综合指标的权重以及隶属函数值,优树综合得分范围为0.151~0.674,其中HY25、HY4、HY20、HY17和HY1综合评价表现较好。采用逐步回归分析建立广宁红花油茶优树评价预测模型,筛选出冠幅面积、单株产果效率、单株产油量、鲜出籽率、干籽出仁率和鲜果含油率等6个核心评价指标。基于6个核心指标采用聚类分析可将31株样株分为3类,第Ⅰ类属于优质型;第Ⅱ类丰产性和优质性较好;第Ⅲ类属于丰产型。【结论】建立了广宁红花油茶优树评价模型,筛选出6个核心评价指标,获得了5株综合表现良好的单株。

[Objective]This study,with an investigation of the genetic variation characteristics of economic traits of Camellia semiserrata,is aimed to establish a prediction model based on the core evaluation indexes screened so as to provide a scientific basis for the formulation and utilization of C.semiserrata germplasm resources.[Method]With 31 excellent trees selected from the core production area of C.semiserrata as research subjects,a determination and analysis was conducted of 20 economic traits such as crown area,fruit yield per plant and oil yield per plant using multivariate statistical methods.[Result](1)There were different degrees of variation in the 20 indicators of the participating strains of excellent trees,with a range of 3.25%−69.47%and between these indicators a total of 38 pairs of indicators reached the highly significant level(P<0.01)whereas 15 pairs reached the significant level(P<0.05).(2)With the principal component analysis,the 11 evaluation indicators were simplified into four independent composite indicators,with a cumulative contribution of 80.23%,retaining most of the information of each trait of the test material.(3)Based on the weights of the four integrated indicators and the values of the affiliation functions,the integrated scores of the excellent trees ranged from 0.151−0.674,with HY25,HY4,HY20,HY17 and HY1 displaying better comprehensive evaluation performance.(4)The prediction model was built for the evaluation of Camellia semiserrata excellent trees using a stepwise regression analysis with canopy area,fresh seed yield,dry seed yield,fresh fruit oil content,oil yield per plant and fruit yield efficiency per plant being the six core evaluation indicators.(5)The six core evaluation indicators can be classified into three categories using cluster analysis with categoryⅠbeing the high quality type,categoryⅡdisplaying better fecundity and quality whereas categoryⅢbeing the fecundity type.[Conclusion]Single plant fruit yield,single plant fruit yield efficiency,canopy area,fruit yield efficiency per plant,oil yield per plant,fresh seed yield,dry seed yield and fresh fruit oil content can be used as important indicators to identify excellent trees of C.semiserrata and in accordance with this model,HY25,HY4,HY20,HY17 and HY1 were selected with favorable comprehensive evaluation.