Pentachlorobenzenethiol (PCBT) has been considered an anomalous uncoupler. It was reported as active in mitochondria, but not effective in inducing electrical conductivity in lipid bilayer membranes. We have overcome the experimental difficulties associated with accurate determination of the induced conductivity. The main contributing factors to the difficulties, we discovered, are the photolability and the low solubility of the compound in aqueous medium. We have conclusively demonstrated that PCBT does induce conductivity in lipid bilayers and compared this conductance with its uncoupling activity reported by other investigators in the literature. We present the results of steady-state current-voltage measurements: conductance dependence on applied voltage for various values of pH, buffer strength and PCBT concentration, as well as the dependence of the conductance on pH, buffer strength and PCBT concentration in the limit of zero applied voltage. We have also compared the above results with those obtained previously with pentachlorophenol. Our experimental results on PCBT-induced membrane conductance suggest that PCBT belongs to class II uncouplers and that "disulfide dimer" of PCBT is membrane inactive. Thus the replacement of oxygen in molecular structure of pentachlorophenol (R-OH) by sulfur (R-SH) does not change the protonophoretic activity of the compound. The conductivity of a membrane is due to PCBT-induced hydrogen ion transfer and it was found to be limited by the kinetics of reactions coupled to transmembrane charge transfer.(ABSTRACT TRUNCATED AT 250 WORDS)