(1) Seawater-adapted Mallard ducks maintained in the laboratory will freely consume food that has been contaminated with either any one of a variety of crude oils or a petroleum derivative such as No. 2 fuel oil. (2) During a 100-day experimental period total masses of petroleum equivalent to 50% of the mean body weight were consumed by some birds and many showed no apparent symptoms of distress. (3) The consumption of petroleum-contaminated food was frequently accompanied by a persistent hyperphagia but no clear patterns of change in body weight were associated with this condition. (4) Among those birds that survived the 100-day experimental period only small changes in mean body weight were observed between successive weighings and in most instances these represented less than 10% of the previously recorded weight. (5) In all groups, including those maintained on uncontaminated food, most of the mortality occurred following exposure to continuous mild cold stress. The total number of deaths in the groups given petroleum-contaminated food, however, was always higher than that among birds given uncontaminated food. (6) The spate of mortality that occurred in groups given petroleum-contaminated food usually occurred earlier, lasted longer, and involved more birds than it did among groups fed uncontaminated food. (7) The pattern of each episode of mortality was sometimes quantitatively related to the concentration of petroleum in the food and a striking range of relative toxicities were observed among the crude oils from different geographic regions. (8) Throughout the experiment, the mean body weight of the birds that died was always significantly less than that of the survivors in the same group; in all instances most of the loss in weight occurred during the 2 weeks preceding death. (9) Autopsy revealed that adrenal hypertrophy and lymphoepithelial involution were characteristic in all of the birds that died, suggesting that a high level of adrenocortical stimulation preceded death. Such high levels of adrenocortical stimulation, therefore, probably occurred sooner in birds consuming petroleum-contaminated food than in birds given uncontaminated food. (10) The consumption of petroleum-contaminated food seemed to constitute a nonspecific stressor and among birds already exposed to stresses, such as hyperosmotic drinking water (seawater) and persistent cold, the ingestion of petroleum seemed to render them more vulnerable to adrenocortical exhaustion, and death frequently ensued.