摘要
应用DPS统计软件,对25个蟹爪兰品种(系)资源的16个生物学性状进行了主成分分析和聚类分析,按照方差累计贡献率88.2%提取了7个主成分,通过计算各品种的主成分得分和综合得分值,分析出综合性状优良和主要性状突出的品种,另外通过16个性状的相关矩阵R分析得出各个性状间的相关性,以利于亲本和杂交后代的筛选。根据聚类分析结果,将25个供试品种(系)从株高、冠幅、茎节长等主要生物学性状的差异聚为3大类,并细分为7小类,为蟹爪兰的分类和良种选育与利用提供理论依据。
Principal component analysis and cluster analysis were conducted on 16 biological characters of 25 Zygocactus trurncatus by using the statistical software DPS.According to the accumulated contribution ratio to the variance over 88.2% from the total information reflected by original 16 indexes, 7 principal components were extracted, and based on the scores of the principal components and the general score , the varieties of good comprehensive and main characters were analyzed and the correlation among the various characters were obtained through the analysis of coefficient of correlation on 16 characters which benefits the selections of parents and filial generation.The differences among the main biology characters such as plant height , crown, cauline and so on were divided into 3 types and 7 subtypes from 25 cultivars.It provided the theoretical foundation for classifications , breeding of Zygocactus trurncatus.
出处
《浙江农业学报》
CSCD
北大核心
2014年第2期319-324,共6页
Acta Agriculturae Zhejiangensis
基金
宁波市科技攻关项目(2006C100035
2010C10006
2013C10006)
宁波市农科教结合项目(2008NK31
2012NK39)
宁波市农科院院长基金(200911
201208)
关键词
蟹爪兰
品种资源
主成分分析
聚类分析
生物学性状
Zygocactus trurncatus
variety resources
component analysis
cluster analysis
biological character