Experimental design in corn

Abstract

It is important to know the number of replications to evaluate characters of corn cultivars with desired accuracy. The objective of this work was to determine how many evaluations (replications) necessary to evaluate characters of corn (Zea mays L.) cultivars. Data were collected from 16 cultivar competition trials of corn, carried out in the Rio Grande do Sul State, Brazil, in the crop years of 2002/2003, 2003/2004 and 2004/2005. In each experiment, it was used a randomized blocks design with three blocks or replications. The following characters were measured: number of days from sowing to 50% flowering male, plant and ear heights at harvest and the number of ears at harvest. Assumptions of the mathematical model were examined, analysis of variance was performed, the repeatability coefficient was estimated and the number of repetitions was calculated. Experiments with four replications are adequate to evaluate the characters number of days from sowing to 50% flowering male, plant height, ear height and the number of ears, with respectively 88.7; 75.1; 74.5 and 69.4% of accuracy.