Bra cup size and handedness were studied as possible risk factors for breast cancer. Data for 3918 cases and 11,712 controls from 7 centres were used to examine the association of handedness with laterality of breast cancer; data for 2325 cases and 7008 controls from 4 centres were used to assess the relation of bra cup size to breast cancer risk. There was a suggestive (P about 0.10) association of handedness with breast cancer laterality: odds ratio of a left-handed (or ambidextrous) woman having a left-sided cancer 1.22 (95% CI 0.96-1.56). Handedness may affect the lateral occurrence of breast cancer, although this tumour is in general more common in the left breast, possibly because this breast is usually slightly larger. Premenopausal women who do not wear bras had half the risk of breast cancer compared with bra users (P about 0.09), possibly because they are thinner and likely to have smaller breasts. Among bra users, larger cup size was associated with an increased risk of breast cancer (P about 0.026), although the association was found only among postmenopausal women and was accounted for, in part, by obesity. These data suggest that bra cup size (and conceivably mammary gland size) may be a risk factor for breast cancer.