Abstract:

In order to systematically evaluate the comprehensive performance of the agronomic traits of maize varieties, this study took the maize varieties (in the high-density group) participating in the summer maize regional trials in Anhui Province from 2011 to 2023 as the research objects. Thirteen key agronomic traits, including plant height, ear height, ear length, ear diameter, bald tip length, 1000-grain weight, and yield, were measured, and a comprehensive evaluation system was constructed through multi-dimensional statistical analysis methods. Firstly, correlation analysis was used to reveal the correlations among traits. The results showed that the yield had an extremely significant positive correlation with the ear diameter and the number of grains per row (r=0.869, (r=0.836), a significant positive correlation with the ear length and the grain yield rate (r=0.626, (r=0.573), and a significant negative correlation with the bald tip length (r=-0.558). Furthermore, principal component analysis was adopted to extract five principal components (with a cumulative contribution rate of 87.96%), which respectively reflected the yield potential, morphological characteristics, and stress resistance, and a comprehensive evaluation model of principal components was constructed. Based on cluster analysis, the varieties were divided into high-yield and stable-yield type (Type I) and wide-adaptability type (Type II). Combined with the grey relational analysis, the relational degrees of various traits with the ideal variety were quantified (relational order: yield > number of rows per ear > ear length > ear diameter > growth period), and the varieties in the years with the best comprehensive performance, namely 2023, 2017, and 2019 (relational degree > 0.82), were screened out. Through the integration of multiple methods, this study established an evaluation model for the agronomic traits of maize varieties, providing a theoretical basis for variety breeding and production promotion.

Key words: corn, correlation analysis, principal component analysis, cluster analysis, grey correlation analysis