Heterosis has been reported for grain yield in soybeans, and despite the fact that hybrid cultivars have not been used yet, the knowledge of heterosis magnitude is very important for a previous selection of crosses, since heterosis is related to parental divergence. However, the obtention of enough F1 seeds for experimental evaluation in plots is a time-consuming task, and thus, other indicators of the occurrence of heterosis could be very useful. The objective of this work was to evaluate heterosis and its relationship with AFLP molecular genetic distance (DG). Six F1 hybrids, derived from parents with different levels of genetic distances (DG) and their respective parents, were evaluated in completely randomized block designs, with four replications. Heterosis estimates were very different among different crosses, varying from 6.29 to 56.50% for mid-parent heterosis (h mg) and from -0.34 to 51.30% for high-parent heterosis (h gs). Besides, the correlation between heterosis and genetic distances (DG) were very high (0.83 and 0.60,respectively, for h mg and h gs), which indicates that DG can be used as indicative of more divergent crosses, and thus, as one criterion for selection of more divergent parents.
Glycine max; heterosis; genetic distance; AFLP marker
