Technical Abstract: Yield gains in orchardgrass (Dactylis glomerata L.), like most cool-season grasses, have lagged compared to other crop species. One strategy for selection gain could use breeding methods that capitalize on combining ability rather than traditional synthetic breeding methods. An initial step in utilizing heterosis in breeding is the identification of combining ability in germplasm sources of interest for breeding. This study evaluated nine orchardgrass germplasm sources utilizing a diallel crossing design to determine general and specific combining abilities among population hybrids. Three forage harvests were taken each year for three years from population hybrids, parental populations, and commercial checks. Differences (P<0.05) were observed among the entries for dry matter yield within harvests and on a total yearly basis. Population hybrids were identified with comparable or higher yields than the check varieties. Significant general and specific combining ability effects for dry matter yield were observed. Two breeding populations, derived from collections from the Altai region of Russia and from Japan, respectively, possessed high overall general combining ability. A germplasm population, collected from Xinjiang Province in China, possessed high specific combining ability when hybridized with the germplasm from Japan and with the variety Latar, although this Chinese germplasm possessed low GCA. Genetic similarity among populations was not correlated with SCA, but genetic similarity within populations was correlated with GCA. These results highlight the presence of combining ability and its potential use in breeding, and suggest that increased homogeneity within populations prior to hybridization may increase the likelihood of GCA.