Four pure lines of pepper developing locally namely (S-1004, L-1008, K-1017 and E-1001) were used in a full diallel crossing program to evaluate the performance of those parents and their crosses, to estimate combining ability and genetic parameters for agronomic, fruit yield and its components traits. Crosses among parents were carried out in plastic house in fall season of 2012 using full diallel cross mating to produce 16 hybrids the parents, F1s and reciprocals hybrids plus hybrid check were sown in plastic house in fall season of 2013 using a R C B D design with four replicates. Data were recorded for plant height, branches number per plant, fruit number per plant, fruit diameter, fruit weight, fruit length and fruit yield per plant. The obtained results showed significant differences among the different genotypes for all traits under investigation. The hybrid (1×4) and the reciprocal hybrid (4×2) produced higher fruit number (73.90, 63.40) and fruit yield per plant (4.06, 3.28 Kg) while, the hybrid (1×3) and reciprocal hybrid (3×2) gave higher fruit length 8.02 cm. Positive and negative heterosis were found among the F1,s and reciprocals. The hybrid (1×4) and reciprocal (4×2) produced higher heterosis in fruit number (88.85, 62.02%) and fruit yield per plant (79.64, 53.99%) respectively. Significant differences were revealed in general and specific combining abilities for all studied traits. Suggesting the presence of both additive and Non-additive gene effects in the inheritance of the various studied traits. The ratio of σ²gca to σ²sca was less than unity for all studied traits, expect for fruit weight, which showed pre dominant role of non-additive gene action in the inheritance. Estimates of GCA effects showed that the best combiner parents were found to these of P1 for plant height, fruit diameter and fruit weight, while P4 for branches number, fruit number, and for fruit yield per plant, and P2 for plant height and P3 for fruit length. Estimates of SCA effects showed that the hybrid 1×2 reflected the highest value in fruit diameter (0.56) and fruit weight (1.35) and the hybrid 1×4 in fruit number (10.02) and fruit yield per plant (0.54) and 2×3 in branches number per plant (0.81) and fruit weight (1.21), and 3×4 in plant height (19.30) and fruit length (0.38). Estimates of RCA effects showed that the reciprocal hybrid 3×1 was the best for fruit weight (8.09), while the reciprocal hybrid 4×1 was the best for branches number per plant (1), fruit number (16.13), fruit length (0.55) and for fruit yield per plant (0.91) whereas 3×2 was the best for plant height branches number  per plants (5.37) and fruit diameter, and 4×3 was the best for fruit number (6.03) and for fruit diameter (0.30). The values of σ² D and σ² D_r were more than that (σ² A) for all studied traits expect for fruit weight and fruit yield per plant, which were less than those for fruit diameter, fruit weight and fruit yield per plant, and this reflects the exceeded one for the value of average dominance degree for all studied traits in both diallel and reciprocal crosses expect for fruit diameter in reciprocal crosses which was less than one. As for broad sense heritability it was high for all studied traits in both diallel and reciprocal crosses, while, the values narrow of sense heritability were low in diallel and reciprocal crosses for all studied traits except for fruit weight, and fruit diameter were high value in reciprocal crosses (0.81) and (0.52) respectively. It was concluded that two inbred (p1 and p4) could be used in a breeding program to develop new versions of high fruit yield per plant and high SCA to produce better fruit yield hybrids and most studied traits were under over dominance gene action. This showed that developing elite hybrids, and most studied traits were under over dominance gene action. This showed that developing elite hybrids were the best method for improving pepper fruit yield per plant in pepper plant.