The consistent association seen between family size and childhood allergy has led to the 'hygiene hypothesis', namely that a lower frequency of infections in early childhood is associated with an increased risk of asthma and hay fever. Maternal atopy, however, is a strong predictor of childhood asthma and hay fever. If maternal atopy is inversely related to the number of siblings then the role of siblings in the development of childhood atopy, the basic tenet of the 'hygiene hypothesis', is challenged. We evaluated the association between number of pregnancies and number of live births with lifetime occurrence of maternal wheeze, asthma, allergic rhinitis, and allergic conjunctivitis in a cross-sectional study in four areas in Italy. A total of 1755 (35-74 year old) nonsmoking women filled a questionnaire on reproductive history as well as on lifetime occurrence of symptoms/diseases. The number of live births was inversely related to lifetime allergic rhinitis (P-value for trend=0.031) and allergic conjunctivitis (P-value for trend=0.011). The odds ratios for those with 4+ children (in comparison with those having 0-1) were: 0.53 (95% CI: 0.27-1.04) and 0.42 (95% CI: 0.22-0.81), respectively. A similar trend was seen for number of pregnancies, although not statistically significant. No association was found between number of pregnancies and number of live births with wheeze or asthma. The results may be interpreted as an indication that maternal atopy influences pregnancy outcomes or that pregnancy itself has an effect on maternal atopy.