SUMMARY Are overnight urine specimens adequate for characterizing the daily salt intake of individuals, i.e., can the overnight specimen replace the 24-bour specimen? Data from 142 male participants of an ongoing trial on the primary prevention of hypertension were used to examine this question with correlation analysis and quantile classification. Estimated correlation between the true mean 24-hour and the true mean overnight sodium excretion was 0.72. Furthermore, 67% of the individuals in the upper third of the distribution of true mean overnight urine sodium were also in the upper third of the distribution of true mean 24-hour sodium. Thus, these data are promising in regard to the use of overnight urine specimens for characterizing the salt intake of individuals. The number of overnight urine collections required to estimate accurately the correlation between an individual's true mean overnight urine sodium and a variable of interest (e.g., blood pressure) was calculated. Given the observed infraand inter-individual variation, the data indicate that 14 measurements are needed to limit the diminution of the correlation coefficient to 10%.