1 The hypothesis of negative feedback regulation of noradrenaline release was studied in guineapig left atrial halves mounted in vitro. 2 Tissues were transmurally stimulated with 30, 100 or 300 pulses at 2 Hz with pulse durations ranging from 50 lts to 2,000 yts, and the efflux of 3H-transmitter determined.3 The efflux of tritium increased with increasing pulse duration as was anticipated, but the effects of supposed presynaptic antagonism by yohimbine were opposite to expectations for a negative feedback system. The magnification of efflux by yohimbine, compared to untreated controls was less rather than more as stimulation-induced transmitter efflux climbed with increases in pulse duration, and with all other parameters of stimulation held constant. 4 It is concluded that the neuronal effect of yohimbine is not linked to negative feedback or to any other system sensing the perineuronal concentration of previously released transmitter. 5 Analysis of the effects on tritium efflux of yohimbine and of prolongation of the stimulation pulse duration, reveals a similarity in the way that they promote transmitter release. Yohimbine increased efflux to approximately the same value at all pulse durations between 50 and 1,000 lis and the value reached was equivalent to that obtained in untreated atria during stimulation with very long pulses (2,000 pjs duration). It is suggested that yohimbine prolongs the outward current attributable to the efflux of potassium from axon terminals, and by this means prolongs depolarization and the period of transmitter release.6 Tetraethylammonium (TEA), a quaternary ion known to plug potassium efflux channels, had an effect on transmitter efflux that was, in some ways, similar to that of yohimbine but of greater magnitude. The present findings provide, for the first time, an alternative to the hypothesis of negative feedback, that might explain the presynaptic effects of adrenoceptor antagonists and possibly other compounds.