The electrogenic Na+/HCO3 cotransporter (NBCe1) regulates acid/base movements in many epithelia (e.g., renal proximal tubule, pancreatic duct) and regulates intracellular pH in numerous cell types (e.g. neurons, glia). In glia, NBCe1 plays an important role during neuronal activity. As neurons fire action potentials, they release K+ into the extracellular microenvironment, depolarizing glia and causing NBCe1 to import HCO3-, thus acidifying the extracellular microenvironment. In 1994, Grichtchenko and Chesler observed that this depolarization-induced extracellular acidification is accentuated by inhibition of extracellular carbonic anhydrase (CA), and suggested that NBCe1 carries one carbonate (CO32-) rather than two HCO3-. The theory that NBCe1 transports CO32- is supported by the results of experiments performed by Lee and Boron in our own laboratory in which acetazolamide (ACZ), a CA blocker, accentuates surface pH (pHS) changes in oocytes co-expressing NBCe1 and CAIV. Here we use a three-dimensional mathematical model of a Xenopus oocyte to investigate the consequences on pHS of the influx of two bicarbonate ions (HCO3- model) versus the influx of one carbonate ion (CO32- model) when CAIV is active or not. The CO32- model predicts a change in pHS that is markedly enhanced by ACZ, in agreement with the experimental data. Thus, we conclude that NBCe1 cannot be a pure bicarbonate transporter. We performed additional simulations in which we implemented simultaneous transmembrane fluxes of carbonate and bicarbonate, with different carbonate/bicarbonate molar ratios but assuming the same total net flux. These simulations lead us to conclude that carbonate is the dominant substrate of NBCe1.