Productive performance of elephant grass clones from full-sib families in an advanced stage of breeding

Abstract

The objectives of this study were to estimate genetic parameters and the repeatability coefficient and to select clones of elephant grass from full-sib families. Fifteen clones and two commercial controls were evaluated in a randomized block design with three replications. Three measurements were taken at regular 10-month intervals. Genotypic variance was significant. The observed heritability based on the mean of the genotypes (0.58), coupled with the results of repeatability (0.53) and selection accuracy (0.76), demonstrates that the genotypes performance is constant between measurements. Thus, only three measurements are necessary to reliably select genotypes and obtain real genetic gains with selection. Among the 15 clones evaluated, six showed higher new means than those of the BRS Capiaçu and BRS Canará controls and can be used as parents and/or to compose value for cultivation and use experiments aimed at the release of cultivars for bioenergy production.