Genotype‐Environment Interactions for Forage Yield of Reed Canarygrass Clones1

Abstract

Limited information is available on genotype‐environment (GE) interactions of dry matter yields in reed canarygrass (Phalaris arundinacea L.). We measured forage yield on 60 clones of reed canarygrass for 3 years in Minnesota, Iowa, Wisconsin, and North Dakota, 2 years in Missouri, and 1 year in South Dakota. Objectives were to determine the magnitude of GE interactions for forage yield testing, to characterize testing environments, to determine an efficient allocation of testing sites, and to ascertain effectiveness of stability parameters in selecting clones for breeding programs. Significant genetic differences were found for harvest 1, regrowth, and season total dry matter yields. Genotype ✕ location and genotype ✕ location ✕ year variances were significant, whereas genotype ✕ year interactions were not. Location and genotype ✕ location effects were inconsistent among years. Year ✕ location interactions were very large in relation to all other sources of variation. Correlations of clonal mean total dry matter yields among locations and resultant cluster analyses indicated that la. and Mo. should represent the minimum number of yield testing sites. Iowa had a high value as a testing site whereas Mo. was intermediate. North Dakota and/or S. Dak. could be added if additional information is desired on broader adaptability and to identify superior genotypes for these locations. Consistent conclusions regarding clonal performance could not be drawn from the stability parameters b, r2, and sy.x. Mean yields per se appeared to be the most useful in determining genetic yield potential.