Many foods contact polymeric packaging materials which contain residues of the polymerization process or additives employed to facilitate processing. The extent of migration of such materials from the packaging to foods is the focus of the present article. A major experimental program using eight polymer-migrant systems is described. Migration was measured to food-simulating liquids (FSL) and to foods. Accelerated tests were conducted with FSL under FDA guidelines conditions so as to develop correlations between such data and those found using foods under normal storage temperatures and shelf lives. In the majority of tests, the migration was found to be approximately proportional to the square root of time, to increase significantly with a rise in temperature, and to be proportional to the initial concentration of migrant in the polymer. Stirring in the FSL or food phase was generally not important except for the system involving dioctyl adipate migrating from polyvinyl chloride film. In some instances, after a period of time, migration rates became very low, and this effect was attributed to saturating the FSL or food phase with migrant. The foods comprised a variety of types, including liquid, semisolid, solid, and dry; both oily and aqueous foods were included. The physical steps involved in migration include the diffusion of the migrant from the interior of the film to the surface, where it can dissolve in the external FSL or food phase. The nature of the FSL or food is shown to be very important in that components can penetrate the polymer and dramatically increase migration rates. Consistent with the FDA guidelines in effect at the time of this study, testing was performed with five FSL (water, 3% acetic acid, 8% and 50% ethanol, and n-heptane) at 49 degrees C. Detailed comparisons were made between the migrations to foods and to FSL; following are the more relevant conclusions. (1) Three percent acetic acid showed no advantage over water as a food simulant even in those cases where the food could be considered acidic in nature. (2) Water, when used as an FSL at 49 degrees C for 5 days, overestimated migration in aqueous foods in about 75% of the cases. In some instances, however, the water phase became saturated with migrant. In other situations, this test protocol underpredicted migration--especially in those cases where there were components in the food that were able to penetrate into the polymer and enhance migration (such as orange juice).(ABSTRACT TRUNCATED AT 400 WORDS)