A critical appraisal was undertaken of the evidence that dietary fat intake is related to breast cancer risk by application of the criteria for causal inference proposed by Bradford Hill to the published evidence that relates dietary fat to breast cancer risk in humans. These criteria concern the consistency, strength, and temporal relationships of possible causative associations and also require the existence of a biologic gradient and examine the extent to which the proposed causal association is in keeping with other biological and epidemiological knowledge. The published reports were inconsistent in their ability to detect a significant association between dietary fat and breast cancer risk, correlation studies that examined the effect of fat over large ranges being largely positive, and studies with stronger designs (case-control, cohort) that examined fat intake over much smaller ranges being largely negative. It was postulated that methodologic limitations associated with the design of the latter studies, in particular the small ranges of fat intake examined and inaccuracies in the measurement of fat intake, may have obscured any relationships between dietary fat and breast cancer that did exist. The remaining criteria, with the exception of temporality and epidemiological coherence, were not satisfied. Insufficient evidence existed to conclude a causal association existed between dietary fat and breast cancer risk in humans. A need for further study was identified in several areas, and it was concluded that intervention studies that examined the effect of fat over large ranges were most likely to yield the information required to determine whether dietary fat intake was causally related to breast cancer risk.