Abstract:

To reveal the effect of main agronomic characters on the yield per plant of Chinese cabbage, 16 new varieties were used as materials, and the correlation analysis, partial correlation analysis, multiple stepwise regression analysis and path analysis were conducted to analyze yield per plant and 8 agronomic characters. The results showed that the variation coefficient of the maximum leaf petiole length was the largest, followed by yield per plant. The maximum leaf width had an extremely significant and positive correlation with the yield per plant, while the plant width was significantly and positively correlated with the yield per plant. The partial correlation analysis showed that the yield per plant was positively correlated with all the 8 agronomic characters, among which the leaf number and the maximum leaf width were extremely and significantly correlated with the yield per plant, while the partial correlation coefficients between other characters and yield per plant were the maximum leaf petiole thickness and the plant width in order. The multiple stepwise regression analysis indicated that Y (yield per plant) = -1480.847 + 31.562X₅ (the maximum leaf width) + 26.028X₃ (the number of leaves) + 13.425X₂ (the plant width) + 429.535X₈ (the maximum leaf petiole thickness) was the optimal regression equation for yield per plant character, whose coefficient of determination R² was 0.930. The direct path coefficients of the 8 agronomic characters to the yield per plant were all positive, and the top 4 characters were in the order of the maximum leaf width > the number of leaves > the maximum leaf petiole thickness > the plant width. Therefore, in the high yield breeding practice of Chinese cabbage, it is necessary to focus on the selection of the characters of the maximum leaf width and leaf number, and the maximum leaf petiole thickness and plant width should also be considered.

Key words: Chinese cabbage, yield per plant, agronomic characters, correlation analysis, partial correlation analysis, regression analysis, path analysis