1. The profile of a range of alpha 1 adrenoceptor antagonists was determined in vitro against cloned human alpha 1A, alpha 1B and alpha 1D adrenoceptors and against noradrenaline-mediated contractions of rat aorta and human prostate. The in vivo profile of compounds was determined in an anaesthetized dog model which allowed the simultaneous assessment of antagonist potency against phenylephrine-mediated increases in blood pressure and prostatic pressure. 2. The quinazoline antagonists, prazosin, doxazosin and alfuzosin displayed high affinity but were non selective for the three cloned human alpha 1 adrenoceptors. Indoramin and SNAP 1069 showed selectivity for alpha 1A and alpha 1B adrenoceptors relative to the alpha 1D subtype. Rec 15/2739, WB 4101, SL 89,0591, (+)- and (-)- tamsulosin showed selectivity for alpha 1A and alpha 1D adrenoceptors relative to the alpha 1B subtype. RS 17053 showed high affinity and selectivity for alpha 1A adrenoceptors (pKi 8.6) relative to alpha 1B (pKi = 7.3) and alpha 1D (pKi = 7.1) subtypes. 3. (+)-Tamsulosin, (-)-tamsulosin, SL 89,0591, Rec 15/2739, SNAP 1069 and RS 17053 appeared to act as competitive antagonists of noradrenaline-mediated contractions of rat aorta yielding pA2 affinity estimates which were similar to binding affinities at cloned human alpha 1D adrenoceptors. The following rank order was obtained: prazosin = (-)-tamsulosin > doxazosin > SL 89,0591 = (+)-tamsulosin > Rec 15/2739 > RS 17053 = SNAP 1069. 4. (-)-Tamsulosin was a very potent, insurmountable antagonist of noradrenaline-mediated contractions of human prostate, yielding an approximate pA2 estimate of 9.8 at 1 nM. The corresponding (+)-enantiomer was 30 fold weaker. SL 89,0591, SNAP 1069 and Rec 15/2739 yielded pA2 estimates which compared well with their alpha 1A binding affinities. The affinity estimate for prazosin on human prostate was lower than the corresponding binding affinity determined at alpha 1A adrenoceptors and RS 17053 was a very weak antagonist on human prostate (pA2 = 6.0) relative to the high affinity (pKi = 8.6) determined at cloned human alpha 1A adrenoceptors. 5. In the anaesthetized dog, in vivo pseudo "pA2' values showed that doxazosin, (+)- and (-)-tamsulosin inhibited phenylephrine-induced increases in prostatic and blood pressure with similar affinity, implying that these agents show little or no selectivity for prostatic responses in this model. SL 89,0591 and SNAP 1069 were moderately selective (3 and 6 fold respectively) for prostatic pressure relative to blood pressure. Rec 15/2739 was a more potent antagonist of phenylephrine-mediated increases in prostatic pressure ("pA2' = 8.74) compared to blood pressure ("pA2' = 7.51). 6. Data in this study suggest that the alpha 1 adrenoceptor mediating noradrenaline-induced contractions of human prostate, whilst having some of the characteristics of an alpha 1A adrenoceptor, cannot be satisfactorily aligned with cloned alpha 1A, alpha 1B or alpha 1D adrenoceptors. In addition, studies in the anaesthetized dog have shown that agents having high affinity and selectivity for prostatic alpha 1 adrenoceptors, particularly over the alpha 1D subtype, appear to inhibit phenylephrine-induced increases in prostatic pressure selectively compared to blood pressure.