Selection of lima bean (Phaseolus lunatus L.) genotypes based on morpho-agronomic descriptors in univariate and multivariate analyses

Abstract

Abstract Lima bean (Phaseolus lunatus L.) is a legume of great agricultural, nutritional, and socioeconomic importance in tropical and subtropical regions. Although it shows high adaptation and elevated protein content, its productive and genetic potential remains underexplored due to the lack of regional studies and limited information on genetic variability. Thus, the characterization and evaluation of genetic variability are essential to support conservation and the selection of superior materials. This study aimed to morpho-agronomically characterize lima bean genotypes and analyze genetic variability using univariate and multivariate techniques to identify promising genotypes. Ten landraces of P. lunatus were cultivated of the Paraíba. The evaluated descriptors followed the recommendations of the International Plant Genetic Resources Institute and included qualitative traits seed [background color, primary color, and secondary color] and quantitative traits [number of seeds per pod, leaflet length and width, plant height, pod width, and pod length]. Data were subjected to descriptive analysis, ANOVA with Tukey’s test (1%), and estimates of genetic parameters. Selection was performed using Tocher’s clustering method and three selection indices: Williams, Mulamba & Mock, and the genotype–ideotype distance index (IDGI). Significant effects were observed for all variables at 1% and 5% probability levels, except for leaflet width, indicating high genetic variability. Heritability estimates were high, particularly for leaflet length (89.68%). Genetic variability was observed among the materials, and Branca Pequena, Cara Larga, Roxinha, and Moita stood out due to their superior performance across key agronomic traits, being recommended for selection and use in breeding programs.