Experiments were performed on adult rainbow trout (Oncorhynchus mykiss) in vivo using chronically cannulated fish and in situ using a perfused posterior cardinal vein preparation (i) to characterise the desensitisation of chromaffin cell nicotinic receptors and (ii) to assess the ability of fish to secrete catecholamines during acute hypoxia with or without functional nicotinic receptors. Intra-arterial injection of nicotine (6.0x10(-)(7 )mol kg(-)(1)) caused a rapid increase in plasma adrenaline and noradrenaline levels; the magnitude of this response was unaffected by an injection of nicotine given 60 min earlier. Evidence for nicotinic receptor desensitisation, however, was provided during continuous intravenous infusion of nicotine (1.3x10(-)(5 )mol kg(-)(1 )h(-)(1)) in which plasma catecholamine levels increased initially but then returned to baseline levels. To ensure that the decline in circulating catecholamine concentrations during continuous nicotine infusion was not related to changes in storage levels or altered rates of degradation/clearance, in situ posterior cardinal vein preparations were derived from fish previously experiencing 60 min of saline or nicotine infusion. Confirmation of nicotinic receptor desensitisation was provided by demonstrating that the preparations derived from nicotine-infused fish were unresponsive to nicotine (10(-)(5 )mol l(-)(1)), yet remained responsive to angiotensin II (500 pmol kg(-)(1)). The in situ experiments demonstrated that desensitisation of the nicotinic receptor occurred within 5 min of receptor stimulation and that resensitisation was established 40 min later. The ability to elevate plasma catecholamine levels during acute hypoxia (40-45 mmHg; 5.3-6.0 kPa) was not impaired in fish experiencing nicotinic receptor desensitisation. Indeed, peak plasma adrenaline levels were significantly higher in the desensitised fish during hypoxia than in controls (263+/-86 versus 69+/-26 nmol l(-)(1); means +/- s.e.m., N=6-9). Thus, the results of the present study demonstrate that activation of preganglionic sympathetic cholinergic nerve fibres and the resultant stimulation of nicotinic receptors is not the sole mechanism for eliciting catecholamine secretion during hypoxia.