Calcium balance performance was evaluated in 130 normal perimenopausal women ages 35 to 50, studied on their usual, self-selected dietary calcium intakes. Two distinct balance methods were used, one based on customary intake and output measurements, and the other based on absorption and clearance of two simultaneously administered calcium isotopes. Both methods yielded essentially identical results. Under study conditions as closely approximating normal living as possible, these women averaged -24 and -31 mg Ca/day by the two balance methods (exclusive of unmeasured dermal losses). More significantly, there was a positive correlation between intake and balance, with women on higher self-selected intakes exhibiting more positive balance than women selecting lower intakes. The calcium intake predicted from our data which would be required to produce an average balance of zero was 1.241 g/day. This is significantly in excess of the current recommended dietary allowance for calcium. The implications and significance of this discrepancy are explored in detail.