Experiments were designed to determine whether band 3, the anion transport protein of the red cell membrane, contains a mobile element that acts as a carrier to move the anions across a permeability barrier. The transport site-specific, nonpenetrating irreversible inhibitor 4,4'-diisothiocyano-2,2'-stilbene disulfonate (DIDS) was found to be effective only when applied extracellularly. It was used to sequester transport sites on the extracellular side of the membrane in intact cells. The membranes were then coverted into inside-out vesicles. The number of anion transport sites available on the cytoplasmic side of the vesicle membranes was then estimated by measuring the binding of N-(-4-azido-2-nitrophenyl)-2-aminoethyl-sulfonate (NAP-taurine), a photoreactive probe. Pretreatment with DIDS from the extracullular side substantially reduced the binding of NAP-taurine at the cytoplasmic side. Since NAP-taurine does not appear to penetrate into the intravesicular (normally extracellular) space, a transmembrane effect is apparently involved. About 70% of the DIDS-sensitive NAP-taurine binding sites are located in band 3, with the remainder largely in a lower molecular weight (band 4) region. A similar pattern of reduction in NAP-taurine binding is produced by high concentrations of Cl-, but this anion has little or no effect in vesicles from cells pretreated with DIDS. Thus the DIDS-modulated sites seem to be capable of binding either NAP-taurine or Cl. It is suggested that band 3 contains a mobile transport element that can be recruited to the extracellular surface by DIDS, thus becoming unavailable to NAP-taurine at the cytoplasmic face of the membrane. The results are consistent with a model of carrier-mediated transport in which the movement of the transport site is associated with a local conformational change in band 3 protein.