Analysis of gene action in diallel crosses among some Faba bean (Vicia faba L.) genotypes under Maryout conditions

Seven genotypes of faba bean (Vicia faba L.) were used in carrying out half diallel cross, 21 F1 hybrids and 21 F2 progenies evaluated under well watered and rainfed conditions at the Maryout Agriculture Experimental Station of Desert Research Center. Mean squares of genotypes in F1 and F2 generations showed that the differences due to genotypes were significant for all characters studied under well watered and rainfed conditions. The four parental genotypes; P2, P4, P5 and P6 were the earliest in days to 50% flowering and recorded values raging from 30.88 to 47.98 days under well watered and rainfed conditions and the two crosses; P1 × P3 and P2 × P7 in both generations under the two treatments. The parental genotype P2 recorded the highest number of branches per plant (7.85 and 6.94 branches) under well watered and rainfed treatments, respectively. While, the two crosses; P2 × P4 and P2 × P6 recorded the highest number of pods per plant in both generations under well watered and rainfed treatments. For 100-seed weight the parent Aquadulce (P4) recorded the highest values under well watered and rainfed conditions (95.62 and 71.72 g, respectively). As well as the two crosses; P2 × P5 and P2 × P6 recorded the highest values for seed yield per plant. Significant positive heterosis and heterobeltiosis were detected for different traits; With respect to seed yield per plant, the seven crosses; P1 × P7, P2 × P5, P2 × P6, P4 × P6, P5 × P6, P5 × P7 and P6 × P7 had significant positive heterotic effects relative to mid and better parents under the two irrigation treatments. Mean squares of both GCA and SCA estimates were highly significant or significant in both generations for all the studied traits under well watered and rainfed conditions. Variances due to GCA were larger than those for SCA and exceeded the unity were detected for days to 50% flowering in F1 generation under rainfed conditions, No. of branches per plant, No. of pods per plant in both treatments and generations except for F1 generation under rainfed condition, 100-seed weight in both treatments and generations except for the F1 generation under well watered conditions and seed yield per plant in both treatments and generations except for the F2 generation under well watered and rainfed conditions, revealing that the largest part of the total genetic variance associated with different traits being the result of additives types on gene action. General combining ability results showed that the three parental genotypes (P1 (G.461), (P2 (NBL2) and P4 (Aquadulce)) were good combiners for improving most studied traits. Such combinations might have desirable transgressive segregations, provided that the additive genetic system is present in different crosses for increasing plant yield and its components under targeted well watered and rainfed conditions. For SCA, the desirable inter-and intra-allelic interactions were presented in the cross P6 × P7 in the two generations under both treatments along with P1 × P2 in the F2 generation under both treatments, P1 × P5 in F1 and F2 generations under well watered and rainfed conditions respectively, P1 × P6, P2 × P5 and P4 × P6 in F1 under both treatments and the three crosses; P1 × P7, P3 × P4 and P3 × P5 in F1 generation under well watered conditions showed significant positive effects for 100-seed weight. Moreover, seven F1′ P1 × P7 in both generations and under the two treatments, P1 × P4 in the F1 generation under well watered conditions, P2 × P5, P2 × P6, P4 × P6 and P5 × P7 under F1 generation under both treatments and P6 × P7 in the F2 generation under rainfed conditions possessed significant positive effects for seed yield per plant. These crosses might be of interest in breeding programs to produce pure lines while most of them involve at least one good combiner for the trait in view.