The controversial hypothesis that capillary pressure (Pc) is autoregulated in response to arterial pressure (PA) alterations was tested in sympathectomized cat skeletal muscle by studying the relation between Pc and PA under conditions of well preserved vascular tone and reactivity, during papaverine-induced maximal vasodilatation (passive vascular bed), and during impaired vascular reactivity caused by preparatory surgery, or by low dose isoproterenol administration. The latter states resembled such under which Pc autoregulation unintentionally seems to have been studied previously. Capillary pressure was assessed with the Pcvenule method for continuous direct pressure recordings from capillaries/postcapillary venules (Mellander et al. 1987) and simultaneously derived from observed net transvascular fluid flux divided by CFC. Resistances in the whole vascular bed and in its pre- and postcapillary segments (Ra and Rv) were determined from recordings of blood flow, PA, Pc, and PV. During preserved vascular reactivity, Pc was found to be virtually constant, that is, almost perfectly autoregulated, over the PA range from 50 to 180 mmHg, whereas in the passive vascular bed there was a direct linear relation between Pc and PA (y = 0.137x + 11.69; r = 0.97). The delta Pc/delta PA ratio was about 1/70 in the normal reactive, and 1/7 in the passive, vascular bed, implying an increase in Pc by 1 mmHg for every 70 mmHg and every 7 mmHg increase in PA, respectively. Capillary pressure autoregulation was explained by precise adjustments of Ra/Rv in relation to PA elicited by myogenic and metabolic regulatory mechanisms. This protective reaction against plasma loss during increased PA was abolished during maximal vasodilation, and was much impaired by surgical trauma, partly via a beta-adrenergic inhibitory effect, and by isoproterenol, in turn causing gross transcapillary fluid fluxes. The latter findings might explain failing Pc autoregulation in some previous studies undertaken under seemingly similar conditions.