Sixty inbred lines of maize (Zea mays L) were screened in the green house at seedling stage both under normal and water deficit conditions. The layout of design was complete randomized desidn (CRD) with two replications. Six water deficit tolerant inbred lines were selected on the basis of dry root weight, root/shoot ratio and relative leaf water content. These inbred lines were sown in the field conditions to attempt all possible crosses in a diallel fashion. The layout of the design was randomized complete block design (RCBD). F1 crosses, their reciprocal crosses along with their parental lines were planted under both normal and water deficit conditions. Normal irrigations(8) were applied to normal set of experiment whereas 50% of yhe normal irrigations(4) were applied to water deficit condition. The data pertaining to various morpho-physiological parameters was recorded at physiological maturity of the crop. The data was subjected to statistical analysis. The coefficient correlation among seedling traits like fresh root weight, dry root weight, root/shoot ratio and relative leaf water contents showed strong association with each other in both normal and water deficit conditions. Mean squares for all the parameters were significant under both conditions, indicating the presence of sufficient genetic variability. Relative general combining ability, specific combining ability and reciprocal effects were calculated for all the characters (Griffing, 1956, Method I, Model I). Estimates of variance components for general combining ability (GCA), specific combining ability (SCA) and reciprocal effects for all the parameters i.e., plant height, leaf temperature, days taken to tasseling, days taken to silking, anthesis-silking interval, no kernels per ear,100 grain weight, grain yield per plant, no of kernel per row, ear height, relative leaf water content, no of ears per plant under normal and water deficit regimes. These components were computed in order to obtain an estimate of relative importance of additive and non-additive or dominance type of gene action. Estimation of components of variation indicate larger values of GCA variance as compared to SCA variance depicting the preponderance of additive genetic effects for all the traits except number of kernels per ear having greater SCA variance than GCA variance 15 showing non additive genetic effects under normal conditions. Heritability estimates for grain yield and yield related parameters shoed maximum ability to transfer the genes to the next generation. Inbred lines W-10× W64SP proved to be best grain yielders on mean basis under normal and water deficit conditions. Hence, these inbred lines might be exploited in future breeding programs.Under water deficit condition, best performing cross was N-18×W-64SP. Hence this cross can be utilized for future breeding program under water deficit conditions.