This paper summarizes a workshop on temporal trends in levels and effects of persistent toxic contaminants in the North American Great Lakes. Information on trends in contaminant levels is reasonably good for sediments, fish, and birds, but is scanty or absent for other ecosystem components. Information on trends in effects has been reported for birds, but is scanty or absent for other groups of organisms. In principle, information on differential trends in effects of contaminants could be used to validate or improve hypotheses about cause-effect relationships and to verify the effectiveness of management actions. However, little or no useful information on differential trends appears to be available. Use of trend data for these purposes will require collection of more detailed information and greater attention to conceptual formulation of hypotheses.