Measurements of benthic fluxes have been made on four occasions between February 1980 and February 1981 at a channel station and a shoal station in South San Francisco Bay, using in situ flux chambers. On each occasion replicate measurements of easily measured substances such as radon, oxygen, ammonia, and silica showed a variability (??1??) of 30% or more over distances of a few meters to tens of meters, presumably due to spatial heterogeneity in the benthic community. Fluxes of radon were greater at the shoal station than at the channel station because of greater macrofaunal irrigation at the former, but showed little seasonal variability at either station. At both stations fluxes of oxygen, carbon dioxide, ammonia, and silica were largest following the spring bloom. Fluxes measured during different seasons ranged over factors of 2-3, 3, 4-5, and 3-10 (respectively), due to variations in phytoplankton productivity and temperature. Fluxes of oxygen and carbon dioxide were greater at the shoal station than at the channel station because the net phytoplankton productivity is greater there and the organic matter produced must be rapidly incorporated in the sediment column. Fluxes of silica were greater at the shoal station, probably because of the greater irrigation rates there. N + N (nitrate + nitrite) fluxes were variable in magnitude and in sign. Phosphate fluxes were too small to measure accurately. Alkalinity fluxes were similar at the two stations and are attributed primarily to carbonate dissolution at the shoal station and to sulfate reduction at the channel station. The estimated average fluxes into South Bay, based on results from these two stations over the course of a year, are (in mmol m-2 d-1): O2 = -27 ?? 6; TCO2 = 23 ?? 6; Alkalinity = 9 ?? 2; N + N = -0.3 ?? 0.5; NH3 = 1.4 ?? 0.2; PO4 = 0.1 ?? 0.4; Si = 5.6 ?? 1.1. These fluxes are comparable in magnitude to those in other temperate estuaries with similar productivity, although the seasonal variability is smaller, probably because the annual temperature range in San Francisco Bay is smaller. Budgets constructed for South San Francisco Bay show that large fractions of the net annual productivity of carbon (about 90%) and silica (about 65%) are recycled by the benthos. Substantial rates of simultaneous nitrification and denitrification must occur in shoal areas, apparently resulting in conversion to N2 of 55% of the particulate nitrogen reaching the sediments. In shoal areas, benthic fluxes can replace the water column standing stocks of ammonia in 2-6 days and silica in 17-34 days, indicating the importance of benthic fluxes in the maintenance of productivity. Pore water profiles of nutrients and Rn-222 show that macrofaunal irrigation is extremely important in transport of silica, ammonia, and alkalinity. Calculations of benthic fluxes from these profiles are less accurate, but yield results consistent with chamber measurements and indicate that most of the NH3, SiO2, and alkalinity fluxes are sustained by reactions occurring throughout the upper 20-40 cm of the sediment column. In contrast, O2, CO2, and N + N fluxes must be dominated by reactions occurring within the upper one cm of the sediment-water interface. While most data support the statements made above, a few flux measurements are contradictory and demonstrate the complexity of benthic exchange. ?? 1985 Dr W. Junk Publishers.