Inheritance of quantitative traits in opium poppy (Papaver somniferum L).

Generation mean analysis was carried out using five parameter model on five cross combinations with five generations i.e. parents, F1s, F2s, and F3s randomly selected from partial diallel breeding experiment. The aim of study was to investigate the mode of gene actions involved in the inheritance of quantitative traits viz. days to 50% flowering, plant height, leaves/plant, capsules/plant, capsule size, capsule weight/plant, seed yield/plant and opium yield/plant. C and D scaling test showed the presence of non allelic interaction in the inheritance for all the traits except for plant height, seed yield/plant (ND1001xIS13) and capsule size (NBR5xND1002) which showed non interacting mode of inheritance. In general, the interaction effect together i.e. additive x additive [i] and dominance x dominance [l] found in higher magnitude than the combined main effects of additive [d] and dominance [h] effects for all the traits in all the five crosses. Dominance effect [h] was found pronounced for most of the traits except days to 50% flowering where additive effect [d] was found prevalent. Among the interaction effects dominance x dominance [l] was predominant over additive x additive [i] for all traits in all the five crosses except capsules/plant and capsule size in cross ND1001xNBRI11 and leaves/plant and opium yield/plant in cross NBRI5xND1002. As per sign of dominance (h) and dominance x dominance (l) duplicate epistasis were noticed for all the traits except plant height and leaves/plant in cross ND1001xUO1285. Potence ratio indicated presence of over dominance for almost all the traits. Substantial amount of realized heterosis, residual heterosis in F2 and F3 progenies and high heritability with moderate to high genetic advance in F2 progeny and significant correlation among important traits in desirable direction were observed. A breeding strategy of diallel selective mating or biparental mating in early segregating generation followed by recurrent selection may be used for genetic improvement.