A half diallel cross among seven white maize inbred lines was made in 2012 growing season. The resulted 21 F₁ crosses and the commercial check hybrid SC10 were evaluated under two different nitrogen levels, i.e. 80 and 120 kg N fad^-1 at the Experimental Farm, Faculty of Agriculture, Kafrelsheikh University in 2013 growing season, to estimate general and specific combining ability effects (GCA and SCA) and their interactions with nitrogen levels as well as identify the superior inbred lines and crosses. Data were collected for number of days to 50% silking, ear length, ear diameter, number of rows ear^-1, number of kernels row^-1 and grain yield plant^-1 and were analyzed according to Griffing (1956) method-4 model-1(fixed model). The results revealed that, the mean squares due to nitrogen levels (N),  genotypes (G), crosses (Cr.), G × N interaction and Cr. × N interaction were significant for all the studied traits. General and specific combining ability (GCA and SCA) mean squares were significant for all the studied traits under the two nitrogen levels and their combined data. Both GCA and SCA effects were significantly interacted with nitrogen levels for most of the studied traits. The non-additive gene action played an important role in the inheritance of all the studied traits, except days to 50% silking and grain yield plant^-1under the two nitrogen levels and their combined data. The inbred lines P₅, P₆ and P₇ showed the best desirable GCA effects for earliness, whereas P₁, P₂ and P₄ were the best general combiners for grain yield plant^-1. The best crosses showed desirable SCA effects were P₃×P_5, P₃×P₇, P₅×P₆ and P₆×P₇ for earliness and P₁×P_4, P₂×P₄, P₃×P₆ and P₅×P₇ for grain yield plant ^-1 under the two nitrogen levels and the combined data. Two crosses P₁×P₄ and P₂×P₄ gave significantly positive superiority in grain yield over the check hybrid SC10 under the two nitrogen levels and the combined data. The genetic diversity (GD) among the seven parental inbred lines was investigated using Random Amplified Polymorphic DNA (RAPD) markers. Seven random primers were used to give a total 70 reproducible RAPD fragments, of them 56 (77.88%) being polymorphic. The GD among the inbred lines differed from 0.333 to 0.655 with an average of 0.503. The estimate value of correlation coefficient between GD and mean performance of the F₁ hybrids for grain yield plant^-1 was low (r = 0.335) or not high enough to be of predictive value. Therefore, the RAPD marker could not be predicted about the mean performance of the grain yield plant^-1  in this study.