A sample of 930 children, between 9 years and 15 years old from six Italian towns, were studied. Their family history of allergy, personal allergy and medical history, current allergy symptoms, exposure to environmental allergens, feeding as infants, and their parents’ occupations were recorded. They were skin-prick tested (SPT) with a range of common allergens. A subject was defined as atopic if at least one SPT caused a weal greater than 3 mm diameter. The association between the recorded information and atopy was investigated by logistic multiple regression. Atopy was positively associated with: high density housing; medium or high exposure to environmental allergens; a history of rhinitis, asthma or atopic eczema; male gender; and a history of breast-feeding. It was independent of infectious diseases, vaccinations and operations, social class and family history. Thus, there was no evidence of a genetic factor in atopy, other than sex.