Nutrient concentrations in Narragansett Bay change in a regular way through the seasons, so that characteristic and well defined cycles are observed, but are different for each nutrient. The cyclic changes are not explainable by processes in the water column alone, nor by advection, even though the replacement time of water in the bay is only 10–40 days. It appears possible to incorporate much of the cycling activity in 13-m3 microcosms, so these must include the dominating features of the complex biogeochemical processes involved. Observations in the microcosms suggest that the processes maintaining the annual cycles are sufficiently strong that, in the absence of deliberate experimental manipulation, the cycles might continue not significantly altered through at least one year. Therefore, the nutrient cycles in the bay can be driven largely by activities internal to the bay, especially sediment-water exchanges.