Efficient Control of Spatial Variation in Yield Trials Using Neighbour Plot Residuals

Abstract

SUMMARYThe effect of spatial variation on experimental error variance and the significance of differences between genetic entries was evaluated in five performance trials. A significant portion of spatial variation could be detected in all the experiments investigated and various neighbour plot residuals were applied to adjust for local field trends. Neighbour plot adjustment was clearly more efficient than the randomized complete block design in reducing error variance and in detecting significant differences between entries. It was also more efficient than lattice designs in trial fields exhibiting short distance trends, which could not be covered efficiently by incomplete blocks. In most experiments with long and narrow plots, longitudinal adjustment using two neighbours at each side of a test plot was superior to adjustment by only one ‘nearest’ neighbour.