关键词: 油菜（Brassica napus L.）, 产量, 农艺性状, 主成分分析, 多重线性回归分析, 相关性分析, 通径分析

Abstract: In order to find out the influence of various agronomic traits on the yield of rape（Brassica napus L.） in Hubei province, the statistical analysis of yield and agronomic traits of tested varieties in the rape regional trials in Hubei province from 2016 to 2017. The results show that the rape varieties （lines） in Hubei province are limited by plant height but still have a large selection space for the selection of ideal plant types. Principal component analysis was used to classify the traits of rape, and the number of horns per leaf was determined as the main factor affecting yield. Multiple linear regression analysis was used to construct a prediction model of rape yield in Hubei province, a linear relationship between the three components of production and production was determined. Correlation analysis and path analysis showed that agronomic traits except plant height and growth period have a direct positive impact on rape yield. The branch site promotes the number of effective pods, and the number of branches promotes the number of effective pods and the number of pods per pod, but inhibits thousand-grain weight. There is a significant mutual inhibition between the three factors of yield composition, but the largest contribution to yield is the number of grains per horn.

Key words: rape（Brassica napus L.）, yield, agronomic traits, principal component analysis, multiple linear regression analysis, correlation analysis, path analysis