Direct renal nitric oxide (NO) measurements were infrequent and no simultaneous measurements of renal cortical and medullary NO and local perfusion. Large-surface NO electrodes were placed in renal cortex and medulla of anaesthetised rats; simultaneously, renal blood flow (RBF, index of cortical perfusion) and medullary laser-Doppler flux (MBF) were determined. NO synthase inhibitors: nonselective (L-NAME) or selective for neuronal NOS (nNOS) (S-methyl-thiocitrulline, SMTC), and NO donor (SNAP), were used to manipulate tissue NO. Baseline tissue NO was significantly higher in medulla (703+/-49 NM) than in cortex (231+/-17 nM). Minimal cortical and medullary NO current measured after maximal L-NAME dose (2.4 mg kg(-1) i.v.) was taken as tissue NO zero kevel. This dose decreased RBF and MBF significantly (-43%). SMTC, 1.2 mg kg(-1) h(-1) i.v., significantly decreased tissue NO by 105+/-32 nM in cortex and 546+/-64 nM in medulla, RBF and MBF decreased 30% and 20%, respectively. Renal artery infusion of SNAP, 0.24 mg kg(-1) min(-1) significantly increased tissue NO by 139+/-18 nM in cortex and 948+/-110 nM in medulla. Since inhibition of nNOS decreased medullary NO by 80% and MBF by 20% only, this isoform has probably minor role in the maintenance of medullary perfusion.