We evaluated the performance of the lithium ion-selective electrode (ISE) in the Du Pont Na/K/Li analyzer. Lithium concentrations in 106 serum samples from patients being treated with lithium were measured in duplicate with the ISE and by flame photometry. The slope of the regression line for the two methods was 1.004 with a standard error of the estimate of 0.049 mmol/L (x = flame photometry, y = ISE). Lithium measurements by the ISE method in serum or aqueous standards were linear to greater than 2.0 mmol/L. Within-run CVs for low (0.31 mmol/L) and high (1.15 mmol/L) lithium controls were 5.9% and 1.7%, respectively (n = 20). Day-to-day CVs for the same controls were 9.8% and 3.3%, respectively (n = 20). There was no significant interference when the concentrations of sodium, potassium, calcium, or magnesium were varied, nor did intervening urinary lithium analyses affect the measurement of serum lithium. Results for lithium measurement in four serum-based survey materials compared well with results by isotope dilution/mass spectrometry.