Small forage fish are a conduit for the transfer of metals from lower trophic levels to their predators and potentially human consumers of contaminated seafood. A metal bioaccumulation model has been developed for aquatic organisms, including fish, to determine: (1) the predicted body burden of metals at steady-state based upon the rate of metal uptake and loss from aqueous and dietary sources, and (2) the percentage of accumulated metal attributable to dietary exposure. However, this model assumes that rates of metal uptake and loss following aqueous and dietary exposure are constant and that fish only consume one type of prey. As a result, the kinetic parameters entered into the model may be over- or underestimated. Furthermore, the importance of ingestion rate and growth rate on the predicted body burden of metals has not been adequately researched in bioaccumulation studies. I performed laboratory experiments using a radiotracer technique to investigate the influence of water chemistry (salinity, dissolved organic matter concentration), prey type, and fish physiology (ingestion rate, growth rate) on the uptake, loss, and resulting tissue distribution of five metals (Am, Cd, Cr(III), Hg (as Hg(II) and methylmercury (MeHg)), and Zn) and two metalloids (As(V) and Se(VI)) in killifish (Fundulus heteroclitus) and Atlantic silversides (Menidia menidia). The influence of salinity on metal uptake from the aqueous phase in killifish varied by metal; Cd showed an inverse relationship with salinity; As, Hg(II), and MeHg uptake increased with salinity; and Cr showed no relationship with salinity. Estuarine and marine fish drink to osmoregulate, and Cd dissection data at the end of uptake showed that drinking may be an important uptake mechanism for Cd in these fish. Humic acids had a varying effect on metal uptake by killifish from the aqueous phase; Cd uptake showed no relationship with humic acid concentration, whereas As uptake increased with increasing humic acid concentration, and Cr, Hg(II), and MeHg uptake decreased with increasing humic acid concentration. Following dietary exposure to radiolabeled amphipods and worms, assimilation efficiencies (AE) in killifish ranged from 0.2 to 4% for Cr and 92% for MeHg. Except for MeHg, AEs varied between prey type, and Hg(II) showed the greatest difference in AEs (14% amphipods, 24% worms). Calculated trophic transfer factors (TTF) indicated MeHg will biomagnify at this trophic step, whereas As, Cd, Cr, and Hg(II) will not. Killifish were intubated with radiolabeled sediment from three contaminated field sites; the AEs ranged from 0.01 to 0.03% for Cr and 10 to 14% for MeHg, and did not vary among field sites for each metal. Killifish do not actively consume sediment, but ingest it accidently while consuming benthic prey. The low AEs for all metals suggest that while sediment is a sink for metals in the estuarine environment, these metals are not readily bioavailable to killifish. The metal bioaccumulation model was used to predict steady-state metal concentrations in killifish in three contaminated estuaries (Baltimore Harbor, MD; Elizabeth River, VA; and Mare Island, CA) with varying salinity and concentration of dissolved organic matter. Uptake and loss from the aqueous phase was monitored in water from each site and the calculated kinetic parameters were combined with the calculated metal assimilation and retention kinetic parameters from diet (amphipods). The predicted values matched independent field data from contaminated estuaries, indicating that the model can account for the major processes governing metal concentration in killifish. The model indicated that diet accounts for >99% of Cd and MeHg accumulation in killifish on a site-specific basis and that diet was also the major source of As, Cd, and Cr. Finally, uptake and loss parameters were calculated for two populations of Atlantic silverside; Nova Scotia, at the northern end of the species range, with a higher rate of ingestion and growth, and South Carolina at the southern end of the species range, with slower physiological rates. Calculated body burdens of Am, Cd, and Zn were higher in South Carolina silversides and were attributed to higher AEs in South Carolina silversides and a higher growth rate in Nova Scotia silversides, resulting in somatic growth dilution of metals.