Adaptabilidade e estabilidade de variedades crioulas de feijão

Abstract

In Brazil, the bean culture is subjected to different edaphoclimatic conditions, resulting in different genotype performances, influenced by environmental variations, resulting in a reduction in productive potential. In addition, due to the genetic improvement process, many commercial varieties of beans have different responses to different edaphoclimatic conditions, a condition less likely to occur in creole varieties. Thus, the objective of this work was to evaluate the genotype x environment iteration and the adaptability and stability of 30 common bean land races of in the municipality of Dois Vizinhos-PR. The study was carried out by analyzing the productivity data of tests conducted in the experimental area of the Federal Technological University of Paraná, Câmpus Dois Vizinhos (UTFPR-DV). 30 common bean land races were evaluated over eight growing cycles, 1) 2015 dry season, 2)2015/2016 water season, 3) 2016 dry season, 4) 2016/2017 water season, 5) 2017 dry season, 6) 2018 dry season, 7) 2018/2019 water season and 8) dry season 2019, where each cycle represented one environment, totalling eight environments. The experimental design used was completely randomized blocks, with three replications per treatment. At the end of the cycle of each variety in the corresponding season, manual harvesting was done, the total yield was obtained by measuring the total grain mass of the plot extrapolating to kg ha-1. The results of productivity were carried out in the analysis of variance, considering the genotypes, environment and the G x A interaction. Complementary analyzes were made for the decomposition of the G x A interaction through the realization of environmental stratification and evaluation of adaptability and phenotypic stability by Lins and Binns methods modified by Carneiro, Eberhart and Russel, Cruz, Torres and Vencovsky, and Analysis of the main additive effects and the multiplicativeAMMI interaction. The Cruz, Torres and Vencovsky method classified environments 1, 4, 6 and 7 as favorable, with two water season environments, environments 4 and 7, and two dry season environments, environments 1 and 6. Lin's method and Binns classified environments 1, 6 and 7, and the AMMI analysis method considered only environment 8 as favorable. Environments 6 and 4 were the ones with the highest yeld averages, and environment 8 showed the lowest yeld average. The land race Carioca Rosa showed greater phenotypic stability followed by Pardinho Mineiro, Carioca Siriri and Pombinho by the method of Cruz, Torres and Vencovsky. Serrana Vagem Branca was the one with the highest yeld average among the evaluated environments, as well as the largest number of environments with the highest yeld average. Serrana Vagem Branca and Gralha MST were classified as widely adapted, and more stable by the Lin and Binns method. The methods of analysis did not show similar results, the methodologies of Eberhart and Russel and Cruz, Torres and Vencovsky were not very consistent in relation to the indication of behavior of the genotypes.