Since ancient times, controversy has centered on whether breast cancer is a systemic or local disease at inception. In recent years, support for the systemic hypothesis has come from randomized clinical trials showing that variations in locoregional therapy have no impact on breast cancer mortality. Yet at first glance, the results of the breast cancer screening trials seem to refute this hypothesis. These trials indicate that screening reduces breast cancer mortality by 30% in postmenopausal women. This could be interpreted to mean that 30% of postmenopausal breast cancers metastasize relatively late in their natural history and are therefore amenable to cure with early diagnosis and timely extirpation. Closer scrutiny of the results of the screening trials, however, shows that screening changes the slope of the breast cancer survival curves (so that survival is prolonged) but does not cause the curves to plateau. Thus, screening may simply increase the time to recurrence and death, rather than eliminate the risk of death for a fraction of postmenopausal women with breast cancer. One may speculate that the timely extirpation of the primary tumor reduces the burden of micrometastatic disease, allowing the host's own defense to exert an effect. Alternatively, the benefit of screening is perhaps the result of early administration of systemic therapy, when the burden of micrometastatic disease is low. Regardless, the results of the screening trials are not necessarily inconsistent with the systemic hypothesis. Longer follow-up of women enrolled in the screening trials may prove useful in better understanding the natural history of breast cancer.