[Recurrent selection for prolificacity in maize at two sowing densities]

The main purpose of the research was to determine the environmental influence (through the use of 60,000 and 30,000 plants per hectare) on the selection efficacy in the synthetic variety AS-Z through the direct response of prolificacy and the indirect response of yield and other agronomic characters. Three cycles of S1 recurrent selection were used. The evaluation was done in three environments in Pontevedra province (Spain) during 1991 and 1994. The crop densities evaluated were similar to the selected ones. The split plot design was used in two of the three environments and the triple lattice on the third. In the last one testcrosses and testers were included with the objective to evaluate the combining ability of the populations. Prolificacy had a favourable selection response in both densities and was better at the high one (60,000 plants/ha) where the linear and quadratic regression coefficients from cycle three to cycle zero were highly significant and positive. The number of ears per plant increased 0.062 and 0.092 at an actual average density of 50,000 and 30,000 plants per hectare, respectively. However, the populations selected at low density only increased 0.013 and 0.039 with a decrease in one case. Yield did not show a correlated response to selection for prolificacy because only populations selected at high density presented a non significant increase through the cycles (2.52 per cent). Genetic drift and the reduced number of selection cycles may be the reasons of the lack of a correlated response. High density selection may be the best environment to improve the yield components and the combining ability of AS-Z and their derivates. It is expected that further cycles of selection will improve yield. It is confirmed the importance in obtaining varieties with more root and stem quality in prolific genotypes with the purpose to select them at high densities