1. In the smooth muscle of the guinea-pig taenia caeci, bradykinin produces a relaxation followed by a contraction. In the presence of hexamethonium and guanethidine, both these phases of the response were insensitive to tetrodotoxin (100 nM), omega-conotoxin GVIA (100 nM) and ibuprofen (1 microM), suggesting that they are due to a direct action on the smooth muscle. 2. The B1 receptor-selective agonist, [des-Arg9]-BK (1-100 microM), was inactive in the taenia caeci, and the B1 receptor-selective antagonist, [Leu8,des-Arg9]-BK (1-10 microM), did not inhibit either phase of the bradykinin-induced response. The B2 receptor-selective antagonist, D-Arg-[Hyp3,Thi5,D-Tic7,Oic8]-BK (Hoe 140) (30-300 nM), inhibited both the bradykinin-induced relaxation and contraction with a similar affinity (apparent pKB estimates of 8.5 +/- 0.1 and 8.4 +/- 0.1 respectively). 3. In a depolarizing high-K(+)-solution, bradykinin produced concentration-related contractions, though of diminished magnitude; but no relaxation was observed in such media. In Krebs solution, the Ca(2+)-activated K(+)-channel blocker, apamin (10 nM), abolished relaxant responses. These observations suggest that contraction results both from membrane potential-dependent, and membrane potential-independent, mechanisms; whereas relaxant responses result entirely from membrane potential-dependent mechanisms. Contractile responses obtained in the high K(+)-solution were inhibited by D-Arg-[Hyp3,Thi5,D-Tic7,Oic8]-BK with an apparent pKB value of 8.4 +/- 0.1. 4. In a Ca(2+)-free, EGTA-containing medium, relatively high concentrations of bradykinin (> 100 nM) produced transient contractions, suggesting that a component of the contractile response results from release of Ca2+ from an intracellular store. This intracellular Ca2+ store could be refilled in the presence of extracellular Ca2+. The B, receptor antagonist, [Leu8,des-Argj-BK (10 micro M), did not inhibit this bradykinin-induced contraction, whereas the B2 receptor antagonist, D-Arg-[Hyp3,Thi5,D-Tic7,Oic8]-BK(100 nM) markedly attenuated it (P<0.001; n = 6).5. Bradykinin (10 nM- 100 micro M) significantly elevated tissue levels of total [3H]-inositol phosphates in the presence of Li?, after incubation with myo-[3H]-inositol. The B, receptor-selective agonist, [des-Argl-BK(100IM) did not stimulate [3H]-inositol phosphate formation, and the B, receptor-selective antagonist,[Leu8,des-Argl-BK, did not inhibit the formation of [3H]-inositol phosphates in response to a submaximal concentration of bradykinin (1I0 1M; P> 0.05). Two B2 receptor antagonists, D-Arg-[Hyp3,DPhe7]-BK and D-Arg-[Hyp3,Thi5,D-Tic7,Oic8]-BK, inhibited bradykinin-induced accumulation of total[3H]-inositol phosphates with apparent pKB estimates of 5.4 +/0 0.3 and 8.4 +/- 0.1, respectively.6. These data suggest that in the guinea-pig taenia caeci, the five aspects of the action of bradykinin studied (the relaxant and the contractile elements of the biphasic mechanical response, the contractile response in a depolarizing high-K' solution medium and zero-Ca2+ media, and stimulation of phosphatidylinositol turnover), all result from activation of B2 receptors. A possible causal relationship is suggested between these B2 receptor-mediated membrane potential-dependent, and -independent events,and their roles in excitation contraction coupling.