Cell Wall Constituents of Reed Canarygrass: Genetic Variability and Relationship to Digestibility and Yield1

Objectives of this study were to obtain estimates of combining ability and gene action for cell wall constituents (CWC) and to ascertain the relationships among CWC, in vitro digestible dry matter (IVDDM), and forage yield in a diverse germplasm source of reed canarygrass (Phalaris arundinacea L.). Ramets of 12 randomly selected clones and an incomplete diallel cross consisting of 30 crosses obtained by mutual pollination of each clone to five of the remaining 11 were grown for 2 years at two locations with two replicates arranged in randomized complete blocks. Each 1.25 m2 plot consisted either of 15 F1 plants or ramets of the parents. General (GCA) and specific combining ability (SCA) regrowth forage were estimated by least squares analysis. Mean squares among parents for CWC and yield for IVDDM and GCA were highly significant (P < 0.01). Parent mean square for IVDDM and GCA mean square for CWC were significant (P < 0.05). Mean squares of SCA for traits were nonsignificant. Estimated variance components were interpreted assuming disomic inheritance. Estimates of total genetic variance (s2p), covariance of parent‐off‐spring ( ) and GCA variance (s2g) were obtained. Estimates of s2g were larger than estimates of 2 (and 4s2g for CWC) indicating that non‐additive genetic variance was important for CWC and yield. Broad and narrow sense heritabilities for CWC were 0.83 and 0.33, respectively. Phenotypic correlation (r = 0.87) for CWC between parent clone performance and GCA effect suggested that clone performance could be used in initial selection for CWC. Phenotypic and genotypic (in parentheses) correlations between CWC and IVDDM were r = 0.77 (−0.75) and r = −0.68 (−0.90) for parents and crosses, respectively. Phenotypic correlations between CWC and forage regrowth yield were nonsignificant. The multiple phenotypic correlation of forage on cell solubles (100 − CWC) and IVDDM was R = 0.47 and R = 0.50 for parents and crosses, respectively. The relatively small coefficients of determinations (22% and 25%) suggested that progress could be made in selecting for lower concentration of CWC and higher concentrations of IVDDM without adversely affecting forage yield.