The estimation of nonadditive genetic effects plays a signi cant role in the accuracy of estimated breeding values of growth traits. Data for genetic estimation of individual and maternal heterosis of crossbreed cattle in Ethiopia are limited to research institutions and universities. This paper aims to estimate the crossbreeding effects on the growth performances of Holstein FriesianxHorro and JerseyxHorro crosses in Ethiopia. The data used in the study included pedigree and weight data of animals born between 1980 and 2008. Heritability of growth traits were estimated using restricted maximum likelihood (ASREML). However, the additive, maternal, and heterosis effects of the crosses were estimated using the crossbreeding effects (CBE3) package and tting Kinghorn's Model one for growth traits. The direct and maternal heritability estimates for one-year weight, for the best model, which included the direct maternal covariance were 0.77±0.12 and 0.26±0.09, respectively. Individual heteroses estimated for Jersey and Holstein Friesian were signi cantly high and positive for one year weight (21.6±6.7 and 26.0±3.9 kg), pre-weaning average daily gain (27.4±26 and 28.9±15 g), and post-weaning average daily gain (68.8±16.6 and 61.8±9.9 g), respectively. The maternal additive effect for growth traits was mostly positive; therefore, it may be preferable to use crossbred cows resulting from purebred dams instead of using crossbred cows from crossbred dams. Signi cantly higher heterosis and additive parameters in Holstein Friesian x Horro crosses showed that crossbreeding of Horro with Holstein Friesian may be advisable in the areas of Horro cattle breeds occupied.