The kinetic features of glucose transport in human erythrocytes have been the subject of many studies, but no model is consistent with both the kinetic observations and the characteristics of the purified transporter. In order to reevaluate some of the kinetic features, initial rate measurements were performed at 0 degree C. The following kinetic parameters were obtained for fresh blood: zero-trans efflux Km = 3.4 mM, Vmax = 5.5 mM/min; infinite-trans efflux Km = 8.7 mM, Vmax = 28 mM/min. For outdated blood, somewhat different parameters were obtained: zero-trans efflux Km = 2.7 mM, Vmax = 2.4 mM/min; infinite-trans efflux Km = 19 mM, Vmax = 23 mM/min. The Km values for fresh blood differ from the previously reported values of 16 mM and 3.4 mM for zero-trans and infinite-trans efflux, respectively (Baker, G.F. and Naftalin, R.J. (1979) Biochim. Biophys. Acta 550, 474-484). The use of 50 mM galactose rather than 100 mM glucose as the infinite-trans sugar produced no change in the infinite-trans efflux Km values but somewhat lower Vmax values. Simulations indicate that initial rates were closely approximated by the experimental conditions. The observed time courses of efflux are inconsistent with a model involving rate-limiting dissociation of glucose from hemoglobin (Naftalin, R.J., Smith, P.M. and Roselaar, S.E. (1985) Biochim. Biophys. Acta 820, 235-249). The results presented here support the adequacy of the carrier model to account for the kinetics.