6个藜麦品种(系)农艺性状的主成分分析  

Principal component analysis of agronomic traits of six quinoa varieties (or lines)

作　　者：宋娇 姚有华 刘洋[1] 迟德钊[1] 王越 

机构地区：[1]青海大学,青海省青稞遗传育种重点实验室,省部共建三江源生态与高原农牧业国家重点实验室,青海西宁810016

出　　处：《青海大学学报（自然科学版）》2017年第6期6-10,共5页Journal of Qinghai University(Natural Science)

基　　金：青海省科学技术厅项目(2015-NK-504)

摘　　要：为了解与藜麦产量相关的表型性状,进而为高产藜麦的选育提供理论依据,以HX1,HX2,NKY-S4,青藜2号、陇藜1号和青藜1号6个藜麦品种(系)作为供试材料,每个品种(系)挑选100个单株进行农艺性状考察,并运用SPSS进行相关性分析和主成分分析。结果表明:相关性分析发现单株产量与株高、茎秆长、有效穗数和千粒重有正相关关系,与主穗直径、穗色、杆色和穗型有负相关关系;主成分分析提取了可反映原变量信息82.793%的2个主成分,并据此给6个藜麦品种(系)做了综合评价。本研究筛选出青藜2号在参与分析的7个农艺性状上综合表现较好。In order to understand the phenotypic traits related to the yield of quinoa, and to provide theoretical basis for the breeding of high yield quinoa, 6 quinoa（ Chenopodium quinoa Willd. ） varie-ties （or lines） ,HX1 ,HX2 ,Qingli 1, Qingli 2, Longli 1 and NKY - S4 were used as materials. 100 individual plants of each variety were selected to investigate phenotypic traits. The correlation analy-sis show that the yield per plant has a positive correlation with plant height,the stemlength of main panicle, effective panicle number and 1000 - grain weight, and a negative correlation with stem di-ameter ,spike color, rod color and spike type. Two components which could reflect 82. 7 9 3 % infor-mation of the original variables were extracted using principal component analysis,and accordingly 6 varieties （or lines） were synthetically evaluated. The seven agronomic traits of Qingli 2 are better than others, and the variety is a good variety.

关 键 词：藜麦 相关性分析 主成分分析 

分 类 号：S519[农业科学—作物学]