1. The gigaseal voltage-clamp technique was used to record responses of hamster taste receptor cells to synthetic sweeteners and cyclic nucleotides. Voltage-dependent currents and steady-state currents were monitored during bath exchanges of saccharin, two high-potency sweeteners, 8-chlorophenylthio-adenosine 3',5'-cyclic monophosphate (8cpt-cAMP), and dibutyryl-guanosine 3',5'-cyclic monophosphate (db-cGMP). 2. Of the 237 fungiform taste cells studied, only one in eight was sweet responsive. Outward currents, both voltage-dependent and resting, were reduced by all of the sweeteners tested in sweet-responsive taste cells, whereas these currents were unaffected by sweeteners in sweet-unresponsive taste cells. 3. In every sweet-responsive cell tested, 8cpt-cAMP and db-cGMP mimicked the response to the sweeteners, but neither nucleotide elicited responses in sweet-unresponsive cells. Thus there was a one-to-one correlation between sweet responsivity and cyclic nucleotide responsivity. 4. Sweet responses showed cross adaptation with cyclic nucleotide responses. This indicates that the same ion channel is modulated by sweeteners and cyclic nucleotides. 5. The sweetener- and cyclic nucleotide-blocked current had an apparent reversal potential of -50 mV, which was close to the potassium reversal potential in these experiments. In addition, there was no effect of sweeteners and cyclic nucleotides in the presence of the K+ channel blocker tetraethylammonium bromide (TEA). These data suggest that block of a resting, TEA-sensitive K+ current is the final common step leading to taste cell depolarization during sweet transduction. 6. These data, together with data from a previous study (Cummings et al. 1993), suggest that both synthetic sweeteners and sucrose utilize second-messenger pathways that block a resting K+ conductance to depolarize the taste cell membrane.