This study was undertaken to determine whether lactating women consuming a low-magnesium diet compensate for magnesium secreted into breast milk by decreasing urinary magnesium losses. Six lactating (L) women, six nonlactating (NL) women (75 +/- 5 and 61 +/- 5 d postpartum, respectively), and seven never-pregnant (NP) women were studied while consuming a constant magnesium intake of 8.97 +/- 0.01 mmol/d for 20 d. After a 5-d stabilization period on the controlled diet, 24-h urine and fecal samples were collected for the next 15 d. The L women excreted significantly less (P < 0.01) urinary magnesium (2.10 +/- 0.35 mmol/d) than the NP women (3.45 +/- 0.37 mmol/d). No significant differences were detected in mean apparent magnesium absorption among the three groups of women, because of large individual variations of fecal magnesium and small sample size. L women apparently compensated for magnesium losses in breast milk (1.04 +/- 0.06 mmol/d) by reducing urinary magnesium losses (1.38 mmol/d) when consuming 8.97 mmol Mg/d.