Nixon (1) showed, using comparative data from different systems, that increased nitrogen load to shallow coastal estuaries increased production of phytoplankton. Furthermore, it has been well established that the growth of coastal producers is nitrogen limited (2). In Waquoit Bay, we have a complex of separate estuaries that are subject to different nitrogen loading rates (3). This variation in loading rate provides the opportunity to test, in one system, whether increased nitrogen loads result in increased production. The range of nitrogen loading to the estuaries extended from a high rate of 8.1 X 1 O3 kg N y-’ in Childs River to approximately 0.05 1 kg N y-’ in Sage Lot Pond. Because phytoplankton growth in shallow estuaries is nitrogen limited (2), increased loading rates are likely to affect activity and abundance of these primary producers. Salt marsh habitats are active sites of denitrification and nutrient uptake (2). A strip of salt marsh located between the watershed and the estuary could, therefore, intercept incoming nitrogen and significantly reduce estuarine nitrogen loading. The estuaries of Waquoit Bay are surrounded by different areas of salt marsh. We could, consequently, also evaluate the effects of salt marsh on interception of nitrogen by comparing phytoplankton abundance and activity in estuaries with different extents of fringing salt marsh. In this paper we ask, first, whether there is a relationship between nitrogen loading rate and phytoplankton abundance and productivity; and second, whether the presence of a salt marsh fringe decreases the nitrogen loading rate and, accordingly, lowers phytoplankton abundance and productivity. We measured gross primary production (GPP) and chlorophyll a concentration at two stations in each of five estuaries of Waquoit Bay (Childs River, Quashnet River, Jehu Pond, Hamblin Pond, and Sage Lot Pond). We used standard light/dark bottle technique with 5-h in situ incubation period, and the Winkler titration method to determine primary production of the estuaries. Chlorophyll a concentration was measured by the Lorenzen method (4). The nitrogen loading rate was calibrated based on total dissolved nitrogen (DIN) at shore edge, rate of water recharge, and total area of the estuary. GPP and chlorophyll a increased significantly with higher nitrogen loads (Fig. 1, top panels). For the regression of phytoplankton and loading, P < 0.003 for both GPP and Chl a. In Childs River, for example, the average chlorophyll a concentration and GPP levels were about three times as high as those in Sage Lot Pond. Both GPP rates and chlorophyll a concentration decreased in estuaries with larger areas of fringing salt marsh (Fig. 1, middle panels). The cause of this decrease is not well established. The salt marshes could be physically removing phytoplankton from the flooding estuarine water during high tide and, thus, lowering