ABSTRACT: The American lobster Homarus americanus fishery is heavily dependent on the use of fish as bait to entice lobsters into traps. There is concern that this food supplementation is nutritionally insufficient for lobsters, but previous experiments reported conflicting results. We conducted a long-term feeding experiment in which 1 yr old American lobsters were fed one of 7 diets for a period of 352 d, a time that allowed the lobsters to molt thrice. The diets consisted of fresh frozen herring, a ‘wild’ diet (rock crab, mussel, and Spirulina algae), a formulated artificial diet for shrimp, paired combinations of these 3 diets or a diet formulated at the New England Aquarium (Artemia, fish and krill meal, Spirulina algae, soy lecithin, vitamins and minerals). The lobsters fed the diet of 100% fish had higher initial molting rates, but within the period of this experiment all either contracted shell disease or died. Mixed diets resulted in higher survival and a lower probability of mortality. This research demonstrated a critical time component to diet studies in lobsters. Short- and long-term impacts of diet differ. In the long term, continual high consumption rates of fish by the lobsters promote poor health in all lobsters, not just those of market size. The use of fish as bait may make lobsters more susceptible to the stress associated with environmental fluctuation, thereby leading to increased disease and mortality. This nutritional stress can be used to develop a laboratory model of shell disease in American lobsters.