Evidence exists that breast tumors differing by estrogen receptor (ER) and progesterone receptor (PR) status may be phenotypically distinct diseases resulting from dissimilar etiologic processes. Few studies have attempted to examine the association of physical activity with breast cancer subtype. Such research may prove instructive into the biological mechanisms of activity. Consequently, this investigation was designed to assess the relationship between physical activity and hormone receptor-defined breast cancers in a population of Asian women in which the distribution of receptor types differed from traditional Western populations. Participants, ages 25 to 64 years, were recruited into this population-based, case-control study of breast cancer conducted in Shanghai, China from August 1996 to March 1998. Histologically confirmed breast cancer cases with available receptor status information (n = 1001) and age frequency-matched controls (n = 1,556) completed in-person interviews. Polytomous logistic regression was used to model the association between measures of activity with each breast cancer subtype (ER+/PR+, ER-/PR-, ER+/PR-, and ER-/PR+) using the control population as the reference group. Exercise in both adolescence and the last 10 years was associated with a decreased risk of both receptor-positive (ER+/PR+) and receptor-negative (ER-/PR-) breast cancers in both premenopausal and postmenopausal women (odds ratios, 0.44 and 0.51 and 0.43 and 0.21, respectively). Sweating during exercise within the last 10 years was also associated with decreased risk for receptor-positive and receptor-negative breast cancers among postmenopausal women (odds ratios, 0.58 and 0.28, respectively). These findings suggest that physical activity may reduce breast cancer risk through both hormonal and nonhormonal pathways.