Effect of Maturity and Alfalfa Competition on Expected Selection Response for Smooth Bromegrass Forage Quality Traits1

Abstract

Genetic variation for forage quality traits of early headed smooth bromegrass (bromus inermis Leyss.) forage is well documented. There were two objectives of this study. One was to measure genetic variability in the smooth bromegrass experimental synthetic B8HD at seven growth stages of two stand types (pure stands and alfalfa, Medicago sativa L., mixtures) for neutral detergent fiber (NDF), acid detergent fiber (ADF), acid tergent ilgnin (ADL), cellulose, hemicellulose, and insoluble ash. The second objective was to evaluate the effects of growth stage and stand type on heritability esti~aates and expected selection response. Nine polycross families of B8HD were established in solid seeded plots, both with and without an alfalfa overseeding, at Hancock and Lancaster, WI in late August 1981. In 1982 and 1983, smooth bromegrass forage was sampled every seven days for seven weeks, beginning in early May. Differences among the nine B8HD polycross families were more frequently significant in alfalfa mixtures for NDF, ADF, and cellulose, while differences were more frequently significant in pure stands for ADL. Excluding ash, for which no significant differences among families were found, family differences were significant in 23 of 50 growth stage‐stand type combinations during reproductive growth, but in only 2 of 20 during vegetative growth. Expected direct responses to selection at heading for NDF, ADF, ADL, and cellulose were 3.8, 3.2, 0.6, and 2.4 g kg−1 year−1, respectively, in pure stands, and 2.5, 2.2, 0,5, and 1.6 g kg−1 year−1 in mixed stands. Expectedirect responses in pure stands for hemicellulose and ash were 0.4 and 0.1 g kg−1 year−1, respectively. for phenotypic recurrent selection based on uniparental control of pollination, expected selection gains generally were larger for direct selection than for indirect selection at other growth stages. With uniparental control, selection for high forage quality in pure stands would not be expected to identify genotypes with high forage quality in mixed stands. Using near‐infrared reflectance (NIR) analysis for forage quality evaluation and biparental control of pollination to double expected selection gains, selection at vegetative growth stages in pure stands is expected to be morefficient than direct. selection at the target growth stage in alfalfa mixtures. Selection at vegetative growth stages in pure stands using NIR would be expected to increase efficiency 18 and 68% for NDF, and 30 and 88% for ADF, and 92 and 66% for cellulose, as compared to direct selection in alfalfa mixtures.