Abstract:

To gain a deeper understanding of the effects of major agronomic traits on yield per plant in maize，and to provide a scientific basis for screening and breeding high-yield varieties，this study conducted correlation analysis，multiple linear stepwise regression analysis，and path analysis on yield per plant and major agronomic traits of 96 self-bred maize hybrid combinations. The results showed that the genetic variation coefficients of yield per plant and the 10 agronomic traits across the 96 maize hybrid combinations ranged from 6.16% to 79.22%，among which the variation coefficient of bare tip length was the largest，while that of ear diameter was the smallest. Yield per plant was highly significantly （p<0.01）positively correlated with ear diameter，ear height，kernel row number，kernels per row，tassel branch number，and ear length；significantly（p<0.01）positively correlated with plant height；significantly negatively correlated with bare tip length. Through the establishment of a multiple regression equation for yield per plant of maize and agronomic traits，the optimal regression equation was obtained as Y=-135.672+3.580X?+0.379X?+1.489X?，in which X?，X?，X? represented ear diameter，ear height，kernels per row，respecitvely. The agronomic trait with the largest direct path coefficient to yield per plant of maize was ear diameter（0.430），followed by ear height（0.250）and kernels per row（0.240）. In the breeding of high-yield maize varieties in Zhumadian City，more attention should be paid to the three traits of ear diameter，ear height，and kernels per row，while also taking the interrelationships among the traits into account.

Key words: maize, yield per plant, agronomic trait, correlation analysis, regression analysis, path analysis