1. We have used anaesthetized, acutely instrumented non-pregnant (NP) and late pregnant (P) New Zealand white rabbits to examine the possible role of nitric oxide (NO) in the pregnancy-induced fall of vascular tone and arterial pressure. Systemic, renal and pulmonary vascular resistance, as well as plasma concentrations of cyclic GMP (PcGMP) were compared before and after the inhibition of NO synthesis by N(G)-nitro-L-arginine methyl ester (L-NAME). 2. P rabbits had lower baseline total peripheral resistance (TPR), mean arterial pressure (MAP) and higher PcGMP than NP controls (all P < 0.05 or less). L-NAME (1, 10, 50 mg kg1, i.v.) resulted in dose-dependent elevation of TPR in both groups. However, the absolute, as well as percentage increases in TPR were greater (P < 0.05) in NP than in P rabbits. 3. Cardiac output (CO) was reduced more (P < 0.01) by NO inhibition in NP than P rabbits. Therefore, despite the smaller increase in TPR, the elevation of MAP was greater (P < 0.001) in P than NP rabbits. After L-NAME, NP rabbits developed more severe bradycardia and a greater increase of pulmonary vascular resistance which might have contributed to the more pronounced reduction of CO. 4. PcGMP increased in both groups following L-NAME, but more (P < 0.01) in NP than P rabbits. 5. Infusion of acetylcholine (ACh, 0.02 micromol l-1 kg-1) reduced MAP and TPR more (both P < 0.05) in NP than P rabbits and L-NAME reduced the ACh-induced depressor response only in NP rabbits. 6. These results suggest that the low vascular tone and arterial pressure in pregnant rabbits is not mediated by NO.