Hg and porphyrin levels in single void urine specimens (spot samples) were compared with calculated 24-hr urine levels in 35 (20 [correction of 25] male and 15 female) practicing dentists who had been occupationally exposed to low levels of elemental Hg. The study aimed to: 1) determine the individual variability for Hg and porphyrin concentrations in spot samples over a 24-hr period; 2) test for the presence of diurnal variation in urinary Hg and porphyrin concentrations; and 3) determine the time of day at which a spot sample would give a Hg concentration closest to the 24-hr average concentration. Results confirmed previous reports of a first-order diurnal pattern with a mid-morning peak for Hg concentration (P < .001). A second-order model best described creatinine excretion (P = .0089), with peaks at about 5:00 and 19:00. The use of creatinine adjustment for Hg concentration significantly reduced the intraindividual variation around the diurnal curve. No diurnal patterns were found for any of the porphyrins examined. We recommend that, for small clinical studies using urinary Hg concentration, 24-hr sampling would be ideal, but that for mass screenings and cross-sectional studies, spot samples may be useful because they correlate fairly well with 24-hr averages (creatinine adjusted, r = 0.61; unadjusted, r = 0.74). Because of the existence of diurnal variation, for all cases using serial sampling attention should be paid to time of day.