Abstract Isolated frog skins were voltage clamped at transepithelial potentials ( V t) ranging from −60 mV to 60 mV to measure transepithelial 36Cl − fluxes from the apical to the basolateral bathing solution ( J 13) and in the opposite direction ( J 31). The potential dependence of fluxes obtained in Na +-free choline Ringer's indicates the presence of conductive components that probably correspond to fluxes through paracellular and cellular pathways, respectively. Rectification of fluxes with reversal of the potential reflects a structural asymmetry, presumably in surface charge density. The data are consistent with a charge density of one negative charge per 280 Å 2 on the apical side. A new model for passive Cl − transport was developed that includes surface charge asymmetry and specifically accounts for the observed variation of conductance with potential. In normal frog Ringer's, J 13 was larger than J 31 at zero potential (active Cl − transport), J 13 rose exponentially with increasing positive potential to reach a maximum at 40 mV (approximately open-circuit), and the predicted partial Cl − conductance exceeded the measured conductance leading to the conclusion that when J 13 is largely driven by Na + transport, much of the coupling occurs via nonconductive pathways. Theophylline stimulates Cl − transport that also occurs via nonconductive pathways as V t becomes more positive.