To evaluate the accuracy ofisoelectric point determination by capillary isoelectric focusing, the pI values of nine proteins and a peptide, the pI values of which had been determined by other methods and ranging pI 3.55-9.60, were determined by capillary isoelectric focusing by cofocusing of recently developed peptide pI markers ranging 3.38-10.17, and the consistency of the pI values was examined. Isoelectric focusing was carried out in neutral polymer-coated capillaries, and the pH gradient was mobilized by pressure toward the cathode, to detect samples with absorption at 280 nm at a fixed detection point. Carrier ampholytes from two different suppliers and in different pH ranges were used. The sharp peaks of the highly pure peptide pI markers greatly facilitated the unambiguous identification of the peaks. When a carrier ampholyte ranging over the acidic side was used, the detection of acidic pI samples was anomalously delayed. This could be partly mitigated by reducing the viscosity of the anode solution in comparison with the pH gradient formed in the capillary. Since the detection times vs the pH relationships were not linear in most cases, the use of a linear calibration line over an entire pH gradient would be erroneous. Instead, the pI values of samples were calculated by assuming a linear relation for pH against detection time between two flanking marker peptides. Close agreement between the pI values, determined by capillary isoelectric focusing, and the reference values of the samples was observed within an average difference range of 0.04-0.08 pH unit with a sample consumption of 10-100 ng within 30-60 min. Some carrier ampholytes were preferentially more effective at either the acidic or the basic side of the pH gradient. For confirmation of the completion of focusing, the use of two different focusing times is recommended.