SR52 is the first single cross maize hybrid to be commercialised in the world and forms the basis of maize breeding in Zimbabwe and Eastern and Southern Africa region. Although SR52 is such an exceptional hybrid, the genetic basis of its outstanding yield and heterosis has never been explained. Therefore genetic components of yield and associated traits in SR52 were studied using a generation mean analysis. Parents (P1 and P2), F1 and segregating F2 and backcross generations (BCP1 and BCP2) were evaluated. The experiment was laid out as a randomised complete block design with two replications across two sites in South Africa, during the 2012/2013 season. The study confirms high levels of heterosis exceeding 300 % for grain yield potential and revealed a significant lack of fit of the additive–dominance model for yield and most secondary traits. Consequently, the role of epistasis was investigated by partitioning it into additive × additive , additive × dominance and dominance × dominance gene effects. The full model confirmed presence of epistasis of highly significant (P ≤ 0.01) additive × additive nature for grain yield though negligible in magnitude (<10 %). Both dominance and additive gene effects were highly significant for yield and associated traits. However, dominance gene action contributed over 80 % to grain yield potential. Additive and additive × additive gene effects played a significant but minor role for grain yield potential (less than 20 %). We therefore concluded that dominance gene action is the basis for exceptional heterosis displayed by SR52.