Abstract
Population improvement through recurrent selection is a traditional breeding method that has been used in maize for over 60 years. Objectives of the research were to: a) evaluate effect of selfed progeny recurrent selection on earliness and yield traits, b) compare responses of cycle-1 (S1-line) and cycle-2 (S2-line) populations, and c) determine better strategy for improvement of maize source population "PSEV3" . The experiments were carried out in partially balanced lattice square design with two replications. In cycle-1 and cycle-2 populations, the differences were highly significant for all studied traits. Selfing in both cycles of selection, resulted increase in days to tasseling while reduction in population means for yield traits. In selected progenies, an increase was seen in mean values of yield traits; however, not in days to tasseling and grain moisture in both cycles of selection. Moderate to high heritability values were observed for almost all the traits in both cycles. Selection differential values were positive and high for grain yield, ear height, prolificacy, ear length, and 100-grain weight in cycle-1 and cycle-2. However, negative values of selection differential were seen for days to tasseling and grain moisture in cycle-1 and 2 populations. The expected responses for days to tasseling and grain moisture were negative in first and second selection cycles. Comparatively, larger and positive responses were noted in cycle-2 than cycle-1 for grain yield and its components. Selfed progeny recurrent selection method was found more effective in improving the maize source population "PSEV3" for earliness and yield traits.