Analyses of Variance of Diallel Crosses in Split-Plot Experiments

Abstract

The analysis and interpretation of four split. plot type designs involving diallel crosses are presented. In two designs all crosses in a given whole plot have one parent in common; the distinction between the two designs arises from the inclusion or non-inclusion of parents. In the other two designs (which are split-split.plot designs) reciprocal crosses appear together in split plots while whole plots are composed of all split plots with one common parent. Mean square expectations are given for all four designs for both random and mixed (all effects except genetic ones are considered random) models. The estimation and interpretation of genetic variance components under random models is discussed and a numerical example is presented.