1. Insulin sensitivity was quantified using a modified euglycaemic technique after hepatic cholinergic blockade with atropine and compared with that after surgical denervation. 2. Intraportal administration of atropine produced dose-dependent inhibition of insulin sensitivity in glucose metabolism. ED50 of atropine was 0.99 mg kg-1 (1 mg = 1.5 microM) with maximum inhibition of 40.3 +/- 11.6%. 3. Atropine (3 mg kg-1) reduced insulin sensitivity by a similar amount (33.6 +/- 3.4%) to that produced by hepatic surgical denervation (37.8 +/- 9.8%). Doses greater than 3 mg kg-1 failed to further alter the insulin resistance produced by surgical denervation or atropine (3 mg kg-1) administration, suggesting that activation of hepatic parasympathetic nerves is necessary to fully express the insulin effect. 4. Atropine reduced insulin sensitivity without changes in plasma concentrations of glucagon or insulin. The temporal response to insulin in this euglycaemic study was not changed after atropine administration or after surgical hepatic denervation. 5. It is suggested that hepatic parasympathetic nerves show a synergistic effect with insulin. Disease states that result in hepatic parasympathetic neuropathy would be expected to produce an insulin resistant liver. 6. The modified euglycaemic clamp method for assessing insulin responses was shown to be reproducible up to four times in the same animal and was sufficiently sensitive and quantitative to be able to generate a dose-response curve in each animal for atropine-induced insulin resistance.