1. The whole-cell voltage-clamp technique was used to study the effects of extracellular ATP on smooth muscle cells isolated from the rat vas deferens. 2. ATP (1-200 microM) elicited an inward-rectifying current that was rapid in onset (less than or equal to 100 ms), reached a peak value that depended on [ATP], and desensitized in the continued presence of ATP (half-time approximately 2 s). 3. Cells recovered from desensitization when incubated in the absence of ATP (resensitization half-time approximately 2 min). 4. A comparison was made of the ability of ATP and several of its structural analogues to stimulate inward current at a negative holding potential. ATP was by far the most effective compound among the series ATP, ADP, AMP, adenosine, GTP, UTP and ITP. ADP elicited a current that was 20-25% as large as that produced by ATP, while the other compounds were ineffective at a concentration which produced a maximal ATP response. 5. AMP-CPP (alpha, beta-methylene ATP), AMP-PCP (beta, gamma-methylene ATP), and AMP-PNP (beta, gamma-imido ATP), which are relatively resistant to hydrolysis, were similarly compared to ATP. While none of these analogues elicited a current resembling the ATP-induced current, AMP-CPP and AMP-PNP each produced a small, relatively sustained inward current; AMP-PCP had little or no effect. 6. The ATP-sensitive conductance is cation selective, but does not appear to discriminate strongly between Na+, K+ and Mg2+. 7. Analysis of the fluctuations which accompany the ATP-induced current suggests that ATP controls a population of channels with a unitary current greater than 0.5 pA at -130 mV. 8. The ATP-evoked current discussed in this report may be responsible for the depolarizing effect of ATP previously described in multicellular preparations of the vas deferens.