A
Genetic model is proposed for analyzing generation means of endosperm quantitative traits in cereal crops. The total genetic
effect (G) of endosperm quantitative traits is partitioned into endosperm direct genetic effect(Go),cytoplasm genetic effect (C) ,and maternal plant genetic effect (Gm). Endosperm direct genetic effect can be further partitioned into direct additive(A) and direct dominance (D) genetic
components. Maternal plant genetic effect (Gm)can also be partitioned into maternal additive (Am) and maternal dominance (Dm) genetic components. MINQUE (0/1) method can be used for estimating endosperm direct genetic variance components (σ2A and σ2D ), cytoplasm genetic variance (σ2C) and maternal plant genetic variance components (σ2Am and σ2Dm). Genetic covariance components between endosperm genes and maternal plant genes(σA.Am and σD. Dm) are estimable. Unbiased estimation for covariance components between two traits can also be obtained by the MINQUE(0/1)method. The random genetic effects in the model are able to be predicted by the Adjusted Unbiased Prediction (AUP) approach with MINQUE (0/1) method. Since genetic variation is analyzed for different generations in several crosses,only sam-pled seeds (neither individual seed nor seeds of individual plant) from the whole generation are measured for endosperm traits.