Identification of repetitive bean samples using multivariate analysis

Abstract

Fifty bean (Phaseolus vulgaris L.) samples from the Active Germplasm Bank of Embrapa - Centro Nacional de Pesquisa de Arroz e Feijão (CNPAF), Goiânia, GO, Brazil, were studied using multivariable techniques to screen descriptors for characterization, to measure genetic diversity and to group samples, looking for repetitive access identification. Germplasm evaluation was carried out in June 1993, using the randomized block design with two replications. Ten quantitative and phenologic descriptors were used for characterization and analysed by canonic variables and Mahalanobis distance techniques. There were no redundant characters and all descriptors were important for sample description. The grouping technique using the diversity genetic measures through the Mahalanobis generalized distance showed to be efficient and viable for identification of repetitive samples, being an appropriate procedure to be used in genebanks.