The thermal effects of black versus white membranes on an insulated low-slope roof were studied over an 18-month period as one of a series of tests conducted on the Roof Thermal Research Apparatus (RTRA) at a national laboratory.Half of the insulated roof tests panel’s 4 ft x 8 ft surface~was covered with a black polyisobutylene (PIB) membrane. A white PIB membrane was installed over the other half. Both sections were identically insulated with fiberglass board to provide a nominal thermal resistance of 7.5 hr-ft2-°F/Btu. A heat flux transducer was installed in the center of each section with thermocouples placed below the membrane, at the midplane and beneath the insulation. Local weather data together with specimen heat flux and temperature data were recorded hourly.Seasonal distinctions in measured data between the black and white membranes are reported. Included are cumulative and instantaneous heat fluxes and hourly surface temperature variations. Peak membrane temperatures were observed to differ by up to 50 F during the day. Nighttime differences in membrane surface temperatures were negligible.Changes due to dirt accumulation and local environmental factors were observed in surface reflectance values calculated from the energy balance at the roof membrane and from reflectometerm easurements.The experimental results reported here are currently being used to validate numerical modeling of surface effects upon roof system performance and building energy requirements. The results of the modeling and analysis will be reported in a subsequent paper.Units: I-P