To obtain reference levels for subsequent investigations, we analysed the IgG1 and IgG4 antibody levels to common foods in the sera of 213 unselected children (age 3 months to 14 years). The children were clustered into five age groups and tested on a broad screening panel of common foods. We used the IgG1 and IgG4 RAST with Sepharose-coupled antigens: cows' milk, hens' egg white, banana, legumes (a mixture of soybean and peanut), grains (a mixture of wheat and rice), potato, orange and pork. In all age groups and all antigens, a considerable variability in the antibody response was found. As for some assays more than half of the sera were negative or borderline, statistics based on interval or ordinal scaling were considered inappropriate and we resorted to nominal classification. We decided to use, for each of the assays, the 75-percentile of the age group as a cut-off level. Each antibody titre was thus converted into positive (more than the 75-percentile of that age group) or negative; the number of positive tests was used as the score. This resulted in a sigma G1-score and a sigma G4 score (summed scores for IgG1 and IgG4 antibodies, respectively). The results of the present study indicate that children with a high response to one food tend to have elevated responses to other non-related foods, possibly explained by a defective mucosal barrier and/or a hyperactive immune system. This suggests that a high-food responder phenotype may exist.