Objective: To determine the relationships of dietary iron sources, other dietary factors, and lifestyle to iron status among premenopausal and recently postmenopausal Chinese women with widely varying regional dietary patterns. Design: Cross-sectional. Subjects were interviewed, blood samples were drawn, and dietary intakes were measured by a 3-day dietary survey for subjects in the ®ve survey counties. Setting: Rural China Subjects: About 80 randomly selected subjects per county among women aged 32 ± 66 y. Main outcome measures: Blood hemoglobin, plasma ferritin, and plasma iron. Results: Total iron intake was relatively high (15 ± 29 mgad) compared to developed counties. Heme iron intake was negligible in two of the study counties. Overall levels of iron de®ciency anemia were relatively low in these generally iron-stressed women. There was no clear statistical relationship between iron intake and physiological iron status. Although several measures of dietary intake (heme iron, dietary calcium, animal protein) were correlated with several measures of iron status before adjusting for survey county, only dietary animal protein was signi®cantly positively correlated with plasma ferritin after adjusting for the possibly confounding factor of the survey county (r 0.15, P 0.009). Intakes of potential inhibitors of iron absorption, such as tea, even in very high amounts, were not correlated to iron status. Plasma ferritin was positively correlated with plasma retinol (P 0.024) and cholesterol (P 0.007). Systemic in¯ammatory response, as indicated by high plasma Creactive protein levels, was shown to be raised in a group of subjects with apparently contradictory high levels of ferritin and low levels of hemoglobin (P 0.03). Conclusions: Iron nutriture in these areas of rural China seemed more related to physiological factors such as in¯ammatory response, menses, plasma vitamin A and cholesterol, than to dietary factors.