Produtividade, adaptabilidade e estabilidade de genótipos de milho (Zea mays L.), de baixo custo de sementes

Abstract

The identification of genotypes with broad adaptability and phenotypic stability is one of the main objectives of maize (Zea mays L.) breeding programs, aimed at increasing production to meet the growing demand for this crop in Brazilian agribusiness. This increase in maize demand has led to higher input prices, especially hybrid seeds, making it challenging for small-scale farmers to afford them, often resulting in farmers opting to save seeds. The use of varieties with good performance can be a viable alternative for small farmers, both in terms of market cost and production yield, compared to saving seeds, which significantly reduces productivity. This study aimed to evaluate the adaptability and stability of 25 maize genotypes with low seed costs to recommend them to small farmers. Trials were conducted in 11 environments across northern, northeastern, central-western, and southeastern regions of Brazil. The studies were carried out during the 2022/2023 growing season. Data analysis was performed using the REML/BLUP (Restricted Maximum Likelihood/Best Linear Unbiased Prediction) procedure, with a deviance analysis (ANADEV) based on a mixed linear model under heterogeneity within environments, and means were compared using the LRT (Likelihood Ratio Test). Experimental efficiency was assessed through selective accuracy. To verify adaptability and stability, as well as environmental stratification, the traditional method was used, evaluating the significant interaction between environments and environmental dissimilarity by the method proposed by Annicchiarico (1992), integrated with GGE Biplot (Genotype plus Genotype-by-Environment). Using the Annicchiarico method and GGE biplot graphical analysis, the genotypes Synthetic 91801601 (G20), HTC 697 (G9), Synthetic 91801592 (G16), and Synthetic 91801596 (G17) were considered high-performing based on the confidence index and the distance from the ideal genotype, respectively. However, the genotypes Synthetic 91801601 (G20), HTC 697 (G9), and Synthetic 91801596 (G17) combined high adaptability, stability, and better average productivity, with an average yield of 5.99 tons/ha. Two mega-environments were identified, indicating significant environmental variations.