Abstract:

Correlation and path analysis for 16 traits of Kosteletzkya virginica was carried out. The results were as follows: (1) Seed yield had highly significant positive correlations with number of fruits (r = 0.945), collar diameter (r = 0.770), ratio of bearing branches to branches (r = 0.567), number of bearing branches (r = 0.497) and hundred-seed weight (r = 0.397), and significant positive correlations with crude fat content (r = 0.348); but significant negative correlations with height of bearing branches (r =－0.343); crude fat content exhibited significant positive correlations with hundred-seed weight (r = 0.389), bottom diameter of bearing branches (r = 0.383) and collar diameter (r = 0.358), but significant negative correlations with height of bearing branches (r =－0.397). (2) The number of fruits had the highest positive direct relationship with seed yield followed by leaf length, collar diameter, length of bearing branch, hundred-seed weight, the number of branches, angle of bearing branches, the height of bearing branches and plant height; five of 14 seed yield components exhibited negative direct relationships with seed yield, including leaf width, ratio of leaf length to width, the number of bearing branches, bottom diameter of bearing branches and ratio of bearing branches to branches. The analyses, combining correlation, direct effects and indirect effects, showed that direct selection for the number of fruits, collar diameter and hundred-seed weight would likely be effective in increasing seed yield; it is logical to select plants with lower height of bearing branches for the improvement of seed yield; the direct selection for the number of bearing branches and ratio of bearing branches to branches would be helpful for increasing seed yield due to their high indirect effects via the number of fruits and collar diameter to seed yield.

Key words: Kosteletzkya virginica, Quantitative trait, Correlation, Path analysis