Abstract:

【Objective】 The experiment was conducted to provide a theoretical basis and reference opinions for maize (Zea mays L.) breeding by researching yield, heterosis and main agronomic characters of maize hybrids in different periods in China.【Method】 Twenty maize single cross hybrids and thirty-two parents in different periods utilized widely in China were chosen as materials in the study. The changing trend in yield, heterosis and agronomic characters, correlation analysis between yield and agronomic characters were made by split plot design in three regions. 【Result】 The results showed that the yield of hybrids and mid-parents increased gradually with the varieties replacement. The yield of the earlier varieties showed a decreasing trend in high plant density, while the yield and plant density increased accordantly in present varieties. The heterosis of hybrids presented a trend of up and down in low and high plant density, respectively, and no change was found in middle and average density. Investigation on the changes in plant traits showed that the tassel branch number and the angle between leaf and the uppermost ear reduced gradually, and rows per ear, ear diameter, kernel length, and 100-kernel weight increased gradually, but the kernels per row, shelling percentage and bulk density not changed. The relationship between yield and heterosis was not significant. Rows per ear, ear diameter, kernel length, 100-kernel weight and yield of mid-parents had a significant positive relationship with yield at 0.01 level. The tassel branch number and the angle between leaf and the uppermost ear had a significant negative relationship with yield at 0.01 level. 【Conclusion】 It was found that there was a conflicting evidence to support the heterosis that is the principal contribution to yield increase, the contribution of maize yield increase should owe to the increase of mid-parents yield and plant density, optimization of plant type and ear traits, decrease of tassel branch number.

Key words: maize, yield, heterosis, agronomic traits, correlation analysis