论文部分内容阅读
以入编《中国稻种资源目录》的 1567份浙江粳稻地方品种为材料 ,采用变种类型分层 (丁颖分类法 ) ,聚类分组和直接随机取样 3种方法 ,构建了浙江粳稻地方品种核心样品 ,并探讨了适宜的样品规模和利用编目性状进行核心样品构建的可靠性。研究表明 :变种类型下聚类分组取样优于直接随机取样 ,编目性状可作为构建核心样品的重要性状 ,浙江粳稻地方品种核心样品可采用变种类型分层 /系统聚类分组 /组内随机取样方法构建 ,适宜规模为原有样品的 6.5%。
A total of 1567 Zhejiang japonica rice cultivars were selected from the “China Rice Germplasm Resources Inventory”. Three varieties of japonica rice varieties from Zhejiang Province, including Dingjing classification, cluster classification and direct random sampling, were constructed. Samples and discusses the appropriate sample size and reliability of core sample construction using cataloged traits. The results showed that cluster sampling was better than direct random sampling under variable genotypes, and cataloger traits could be used as important traits for constructing core samples. The core samples of local japonica rice varieties in Zhejiang Province could be classified by type of stratification / systematic clustering / random sampling Construction, suitable for the original scale of 6.5% of the sample.