In this thesis, I investigated the how the life history characteristics of the clam Mya arenaria determine the population response to chronic contaminant exposure. To predict the potential responses of a broadcast-spawning life history such as that of M. arenaria, I surveyed the literature on a variety of bivalve species. By incorporating information on growth, survival, and reproduction into matrix population models I could evaluate the relative contributions of these factors to fitness. For broadcastspawners, long life is an important factor enabling them to gamble on rare, large recruitment events. Another important aspect of the broadcast spawning strategy is the possibility of high variation in larval settlement from year to year. I evaluated the role that this variability plays using a stochastic matrix model, and showed that it tends to increase population growth because of the larger size of rarer, successful recruitment events.