1. The aim of the study was to elucidate the mechanism of the previously reported close correlation observed between noradrenaline and cAMP in a lymphocyte preparation (LP) isolated from peripheral venous blood in healthy subjects. A close correlation was also obtained in the present study between lymphocyte noradrenaline and adrenaline and cAMP both in the basal state and after stimulation with isoproterenol (P<0.05 to 0.007).2. Although 99% of the thrombocytes were removed from the LP during the washing procedure, LP contained approximately one thrombocyte per lymphocyte. The noradrenaline concentration in LP which could be ascribed to residual thrombocytes, calculated from the average noradrenaline concentration in thrombocytes and the number of thrombocytes in LP, correlated closely to noradrenaline in LP (P<0.007).3. To test the hypothesis that noradrenaline in LP was primarily derived from plasma, we studied three patients with phaeochromocytoma, who had high levels of noradrenaline and adrenaline both in plasma and in LP. 4. Further studies showed that the addition of thrombocytes to LP increased cAMP. The response was inhibited by indomethacin, whereas the addition of cimetidine and propranolol had no effect on basal or thrombocyte-stimulated cAMP.5.We conclude that noradrenaline in LP is a marker of the number of residual thrombocytes. The addition of thrombocytes to LP increased cAMP in lymphocytes. This response was not mediated by catecholamines but possibly by prostaglandins.