We examined the temporal and vertical distribution of total ammonia nitrogen (TAN) and un-ionized ammonia nitrogen (NH₃-N) in sediment pore water and compared the temporal patterns of TAN and NH₃-N concentrations in overlying surface water with those in pore water. Pore water was obtained by core extraction and subsequent centrifugation. We measured TAN concentrations and calculated NH₃-N concentrations from February through October 1993 at four sites in Pool 8, upper Mississippi River, at depths of 0 to 4, 4 to 8, and 8 to 12 cm below the sediment-water interface. Total ammonia nitrogen and NH₃-N concentrations were significantly different among sampling dates (p = 0.0001) and sediment depths (p = 0.0001). Concentrations of TAN and NH₃-N in surface water were significantly less than those in pore water from all sediment depths (p < 0.05). Concentrations in pore water ranged from 0.07 to 4.0 mg TAN/L and less than 1 to 20 μg NH₃-N/L in winter, and from 0.07 to 10.0 mg TAN/L and 1 to 175 μg NH₃-N/L in summer; greatest concentrations were usually found in sediments 8 to 12 cm deep. Annual mean TAN concentrations were positively correlated with silt and volatile solids content and were negatively correlated with sand content. Because of the high variability of TAN and NH₃-N concentrations in pore water, sediment toxicity studies should take into account the season and the depth at which sediments are obtained. The annual mean NH₃-N concentration in pore water at one site (55 μg/L) exceeded the concentration (30 μg/L) demonstrated to inhibit growth of fingernail clams in laboratory studies. However, these concentrations apparently were not lethal, as evidenced by the presence of fingernail clams at this site.