The total free radical fluxes generated by illumination of surface water samples with a solar simulator were measured in a variety of coastal surface waters along the east coast of the United States in summer 1987. Photochemically generated free radicals were scavenged with a known excess of added nitric oxide, and the nitric oxide losses were measured by difference using a semiautomated, computer-controlled system coupled to the ship's clean surface water intake system. This first large free radical flux data set shows that radical production occcurs at significant rates that are relatively uniform over large regions. However, at time of this cruise a strong photoreactivity front diveded coastal waters north of Cape Cod (3.85¿0.82 nmol L-1 min-1 sun-1, n=8 sites) from coastal waters south to northern Florida (1.67¿0.32 nmol L-1 min-1 sun-1, n=20). The mean flux on an areal basis is estimated to be ~0.7 mmol m-2 d-1, or about 50 mmol m-2 for the summer season. These values are sufficiently large to produce interesting but unknown geochemical consequences. The regional uniformity and wavelength dependence of the rates are consistent with the concept that the principal mechanism of radical production is photolysis of long-lived dissolved chromophores of unknown structure, such as fulvic and humic acids. A vertical profile shows strong near-surface depletion of reactive materials under calm conditions. The dependence of total radical production rates on irradiation time, light intensity, concentration of added NO, and wave-length range of polychromatic radiation is also reported. The results support the conclusion that the method produces an operationally consistent, well-behaved estimate of photochemical free radical production potential that is not strongly affected by biological processes. ¿American Geophysical Union 1991