The Real Yield Expectancy (RYE) database is used by North Carolina growers for nutrient management and conservation. There is an apparent lack of reference yield data for forage grasses grown on varying soil types, so it was our objective to determine the yield of forages grown on key soil management groups. Stands of tall fescue and bermudagrass were harvested and maintained for a three year period at six locations in the Coastal Plain and Piedmont regions of NC. Bermudagrass was also overseeded with either cereal rye or prairiegrass ‘Matua' on three locations in the Coastal Plain. Test sites were selected to represent eight prominent soil management groups in North Carolina, ranging from poorly drained mineral-organic ultisols to well-drained alfisols. Ammonium nitrate was applied at 0%, 75%, 100%, 125%, and 200% of the RYE recommended N rate based on soil type and forage species. Dry yield was measured for the four harvest events per year, and summed to provide yearly estimates. The RYE database model overpredicted dry yield for tall fescue grown on well-drained residual soils in the Piedmont by a minimum of 3089 and 1833 lb/acre. The over-prediction of yields may be a result of low infiltration, and therefore greater runoff and nitrate loss. The RYE model also overpredicted dry yields for bermudagrass overseeded with prairiegrass on arenic and quartzipsamment soils by a minimum of 2958, 4850, and 3667 lb/acre for Bladen, Sampson, and Pitt counties, respectively. Further evaluation is needed, although it appears at this time that the RYE should be modified to account for lower yields for bermuda/prairiegrass rotations and tall fescue grown on well-drained Piedmont Ultisols and Alfisols.