In 1981, an oil spill field experiment was done in Maine to assess the effects to the benthos of dispersant used in nearshore oil spills. Three test plots, each 60 by 100 m, were set up, each with an upper and a lower intertidal sampling area. There were also five subtidal sampling stations in water depths from 5 to 20 m. One plot was exposed to 945 L (250 gal) of Murban crude oil released on an ebbing tide within containment booms and cleaned up by conventional mechanical methods 24 h later. A second plot was exposed to 945 L of Murban crude oil premixed with 94 L (25 gal) of a widely available self-mix nonionic dispersant. The dispersant-treated oil was discharged over a 2-h period around high water slack tide. During discharge, mixing gates augmented natural energy to provide a worst-case scenario for exposure of the benthos to the complete dispersal of a nearshore oil spill. During and after discharge, dispersed oil in water was monitored fluorimetrically. Total integrated exposure of dispersed oil to the bottom at both upper and lower sampling areas was 30 to 40 ppm · h. Discrete water samples were also taken for other analyses. Dispersed oil in water reaching the bottom had lost most of the hydrocarbons more volatile than n-C17 compared with dispersed oil in water sampled at the same time near the surface. Petroleum retention by intertidal sediments and bivalves measured one week postspill was less in areas exposed to dispersed oil than in areas exposed to untreated oil.