SummaryThis study was conducted to estimate the requirements of sodium (Na) and potassium (K) in Japanese young adults. From 1986 to 2000,109 volunteers (23 males, 86 females), ranging from 18 to 28y old, took part in 11 mineral balance studies after written informed consent had been obtained. The duration of the study periods ranged from 5 to 12d, with a 2-4d adaptation period. Foodstuffs used in each study were selected from those commercially available. The Na and K content of the diet, feces, urine and sweat were mea sured by atomic absorption spectrophotometer. The results of a study in which Na intake was 6.87g/d (ca. 300mmol/d), the highest of all the studies, showed apparent positive Na balances. In contrast, another study in which Na intake was 2.21g/d (ca. 100mmol/d), the lowest of all the studies, showed apparent negative Na balances. These two studies seemed to differ from the other studies, as shown by regression equations calculated from either data of all the studies (n=109) or data that did not include the two studies (n=90). The dietary intakes of Na and K ranged between 38.56-142.23 and 26.77-74.42mg/kg body weight (BW)/d, or 2.21-6.87 and 1.83-3.61g/d, respectively in the complete data, and 43.71 -96.40 and 26.77-63.70mg/kg BW/d, or 3.06-4.06 and 1.83-2.68g/d, respectively in the data that did not include the two studies. The intakes of the two minerals were positively cor related. Na intake (Intake) was correlated positively with apparent absorption (AA) of Na, which was also correlated with Na urinary output (Urine). In the data that did not include the two studies, Na balance (Balance) was not correlated significantly with either Na Intake (r2=0.005) or AA of Na (r2=0.006). However, analysis of all the data showed a significant correlation between Na Balance and both Na Intake (r2=0.361) and AA of Na (r2=0.360). In the complete data, the mean value and upper and lower limits of the 95% confidence interval for the regression equation between Intake and Balance for Na, when balance was equal to zero (i.e mean, upper and lower limits), were 55.824, 60.787 and 50.862mg/ kg BW/d, respectively. K Intake was correlated positively with AA of K, which was also cor related with both Urine K and K Balance. There was a significant correlation between K Intake and K Balance in both the complete data (r2=0.213) and the data that did not include the two studies (r2=0.116). In all the cases, mean, upper and lower limits for K were 39.161, 41.782 and 36.540mg/kg BW/d, respectively. Intakes of Na and K did not corre late with their respective AA rates (%). Within the ranges of K Intake in this study, K Bal ance was affected markedly by K Intake itself as well as by Na Intake. However, in the case of Na, when the data of the highest and lowest Na intake studies were excluded from the analysis, Na Balance did not correlate with Na Intake, whereas the data of all the studies showed Na Balance was affected strongly by Na Intake. The data of this study allowed the estimated average requirements (EARs) for both mi...