The questions addressed in this study were: 1) whether insulin added to the incubation medium can down-regulate (125)I insulin binding to isolated hepatocytes of Atlantic salmon (Salmo salar) and rainbow trout (Oncorhynchus mykiss); 2) whether quantitative assessment of insulin processing can be made on isolated fish liver cells; 3) how ambient temperatures can affect insulin binding, and down-regulation of insulin receptors.After isolation and a short (up to 4h) "metabolic recovery period", liver cells were used either directly in (125)I insulin binding assay or first preincubated for 18h at 4°C or for 3h at 15°C, with or without mammalian or salmon insulin in concentrations ranging from 1 to 1000 nM.Preincubation at 15°C, decreased binding capacity (number of binding sites per liver cell) in all five independent hepatocyte preparations treated with 1000 nM insulin and in four out of five preparations treated with 100 nM insulin. At 4°C insulin binding sites were down-regulated in less than 50% of all hepatocyte preparations and only in the presence of 1000 nM insulin.Differential quantitive assessment was made of a) intact free insulin; b) insulin degraded; c) intact insulin bound to the cell membrane; d) internalized but degraded insulin, and e) intact insulin internalized by liver cells. Hepatocytes preincubated with 100 - 1000 nM insulin at 15°C bound and internalized less (125)I insulin.We hypothesize that in vivo, at water temperatures of 15°C and higher, extreme physiological levels of plasma insulin may regulate the numbers of insulin receptors in the salmonid liver. In contrast, in fish inhabiting cold waters the regulation of insulin receptors by circulating plasma insulin seems to be of little physiological importance.