Corrosion phenomena in metal containers differ from those encountered in, for example, pipeline or bridge construction, in that they can be apparent at very low corrosion rates. Thus, contamination of the contents of the package can occur with relatively low metal uptakes, rendering them unsuitable for use. Dissolution of tin or iron from tinplate and of aluminium from body or end stock are examples of this contamination. Sulphide stains may be unsightly, but are not harmful. Perforation of the container may allow the product to escape, leading to secondary corrosion or contamination, and associated losses; or to the ingress of contaminants, leading to flavor changes or microbial spoilage. In practice, the incidence of problems is very low. Corrosion testing frequently uses the actual products to be packed; this is the ultimate test. Synthetic systems have been used to simulate complex natural food in particular, but also to clarify the behavior of formulated products. Test methods involve electrochemical studies, coupon testing, and test packing. The behavior of coatings for cans may also be studied in this way. Performance is assessed by considering metal pick-up and incidence of swelling or perforation. Controlled-environment chambers are used in accelerated testing. Characterization of materials also involves using synthetic media in laboratory testing, e.g., measurement of tin oxide levels in tinplate. The properties of containers and components may be checked using synthetic or, occasionally, natural materials. While such procedures are helpful in clarifying corrosion processes and studying components or materials, they have limitations. Ideally, the test procedure should correspond to the products to be packed under anticipated conditions of use. Nevertheless, the use of synthetic systems contributes to the optimization of the final container/product combination.