Nine parental genotypes of okra were crossed in complete diallel design to study combining ability and nature of gene action for earliness and yield components. Mean squares of genotypes were found to be highly significant for all studied traits, providing evidence for presence of considerable amount of genetic variation among studied genotypes. The results showed that (P5) and (P9) were the best general combiners for earliness, while (P1), (P4), (P5) and (P7) were found to be good general combiners for total yield per plant. The crosses (P5xP6), (P5xP9), (P5xP8) ang (P9xP2) were the earliest crosses in comparison with the other crosses. Meanwhile, the cross (P1xP9) had the highest mean value for fruit diameter, plant height and fruit weight. In addition, the crosses, (P1xP6), (P1xP7), (P5xP9) and (P6xP4) had the highest mean values for No. of fruit/plant and total yield /plant. Therefore, these promising crosses among F1 hybrids and F1 reciprocal (F1r) combinations could be used for further breeding studies to improve the economic traits in okra. The results revealed that the general combining ability (GCA) and specific combining ability (SCA) mean squares were highly significant for all studied traits. Significant reciprocal effect mean squares were observed for all studied traits, indicating that these traits were controlled by extra-nuclear factors as well as nuclear factors. The results indicated that the magnitude of additive genetic variance (σ2A) were positive and  lower than those of non additive (σ2D) one for most of studied traits, indicating that non additive gene action played a major role in the inheritance of these traits. The broad sense heritability estimates (H2b %) were more than 75% and larger than their corresponding narrow sense heritability (H2n %) for all studied traits. However, estimates of  narrow sense heritability were 13.9%, 32.4, 40.5, 47.1, 76.8  for earliness, fruit length, fruit weight, plant height and fruit diameter, respectively. The estimates of narrow sense heritability ranged from 11.3 % to 17.34% for total fruit yield per plant and No. of fruit per plant, respectively. It could be concluded that the most studied traits were mainly controlled by non additive effects and cytoplasmic factors. Therefore, the genetic material used in this study could be used for hybridization for producing promising crosses to improve economic traits in okra.