Various foodstuffs were prepared in silicone baking moulds and analyzed for siloxane migration using a previously developed and validated (1)H-NMR method. Meat loaf significantly exceeded the overall migration limit of 60 mg kg(-1) (10 mg sdm(-1)) in the first and third experiment. The highest siloxane migration found in a meat loaf after preparation in a commercial mould was 177 mg kg(-1). In contrast, milk-based food showed very low or non-detectable migration (<2.4 mg kg(-1)), even containing high fat levels. Similar results were achieved using 50% ethanol as the simulant for milk-based products, as defined in the Plastics Directive 2007/19/EEC. After solvent extraction of the moulds in simulating long-term usage, no further migration into the food was detectable, indicating that there is no significant formation of low molecular weight, potentially migrating siloxanes from the elastomer. During repeated usage, the moulds showed a high uptake of fat: up to 8.0 g fat per kg elastomer. Proper tempering of the moulds had a major influence on the migration properties of siloxanes into different foodstuffs. Non-tempered moulds with a high level of volatile organic compounds (1.1%) were shown to have considerably higher migration than the equivalent tempered moulds.