Reserpine-pretreated or normal rabbit aortic strips (whose noradrenaline-metabolizing enzymes were inhibited by treatment with 0.5 mM pargyline and by the presence of 0.1 mM U-0521) were incubated with 1.18 muM labelled noradrenaline for 30 min. Subsequently, strips were either analysed or washed out with amine-free solution for 240 to 250 min. From the efflux curves the original distribution of radioactivity was estimated by compartmental analysis. 1. The accumulation of radioactivity in the strips was reduced by 30 muM cocaine to that observed for nerve-free strips. In reserpine-pretreated strips this was due to a partial reduction of the filling of comartment IV (characterized by a half time of efflux of 10 to 20 min; earlier evidence had already shown that part of this compartment is of neuronal origin; Hensling et al., 1976) and a pronounced decrease of the filling of compartment V (characterized by the longest half time). In normal strips the same changes were observed as well as pronounced reduction of the "bound fraction". Thus, cocaine reduced the filing of those compartments which had been identified as neuronal ones (Henseling et al., 1976). 2. When 30 muM cocaine was added to the wash out solution only, the neuronal efflux of radioactivity from reserpine-pretreated strips was accelerated, while the efflux from extraneuronal or extracellular compartments remained unaffected. This effect of cocaine was the same for (-)- and (+)noradrenaline, and its was not significaltly decreased when paired strips exposed to an inhibitor of extraneuronal uptake (86 muM corticosterone) throughout the experiment. In normal strips, cocaine had very little or no effect on the efflux of radioactivity. 3. The results are consistent with the view that cocaine impairs the influx of amine into the neurone, while block of re-uptake of unchanged amine is one of the determinants of its effect on the rate of efflux from the axoplasm. Comparison of efflux curves with corresponding relaxation curves (determined in the absence and presence of cocaine) indicates that, when both noradrenaline-metabolizing enzymes are inhibited, relaxation of the strips is accounted for by a) the efflux of unchanged amine from neuronal (and extraneuronal) stores and b) the sensitivity of the preparation to noradrenaline.