Among the methods currently used to demonstrate a sensitization to foods, the measurement of food specific IgE antibodies (sIgE) is the most practical but not the most accurate. The "sensitivity" of food sIgE determinations is, for example, suboptimal with unstable allergens in fruits and vegetables that are involved in the (birch) pollen-related immediate oral allergy syndromes. In this particular syndrome the history is often conclusive and can be substantiated by skin prick tests with fresh foods. The "sensitivity" of sIgE tests is much better when sIgE are directed to stable plant or animal food allergens which often cause non-immediate generalized reactions. Foods, usually, contain many different (glyco)proteinic allergens of which some are stable and others not. The "sensitivity" of the sIgE test with a particular food, therefore, varies according to the type of allergen that is recognized by the patient. The "specificity" of sIgE tests with foods is affected by the existence of homologous food allergens which induce cross-reactive IgE that may or may not be clinically relevant. While variable, clinical cross-reactivity is more common among botanically-related fruits, among different nuts, among mammalian foods and among seafood than among cereals, grains and legumes. The "specificity" of food sIgE tests is much better when sIgE are directed to unique non-cross-reactive food allergens. Unfortunately, neither the presence of food sIgE nor its level are predictive of clinical reactivity. The identification of individual allergens in foods and the characterization of the relevant IgE binding sites in these allergens might lead to the development of tests that only measure sIgE to clinical relevant food allergens.