(-)Isoprenaline was continuously administered to rats at a rate of 0.4 mg/kg/h for 7 days via subcutaneously (s.c.) implanted osmotic minipumps. This treatment induced cardiac hypertrophy and a marked decrease in basal as well as catecholamine-stimulated adenylate cyclase activity in a ventricular plasma membrane fraction. The total number of beta-adrenoceptors was downregulated by one-half the amount of the receptor sites obtained in a control group. However, in the isoprenaline-treated group, the beta 2-adrenoceptors constituted a significantly smaller proportion of the total beta-adrenoceptor population (28%) than in the control group (50%). Transformation of these relative into absolute values indicates that prolonged isoprenaline treatment induced a significantly higher downregulation of beta 2- than of beta 1-adrenoceptors. The fact of a different beta-adrenoceptor desensitization pattern in response to in vivo administration of nonselective beta-adrenergic agonists therefore must be taken into consideration when desensitization is used as a method for determination of subtype selectivity of an agonist per se. However, we were unable to detect the "lost" beta-adrenoceptors in a light vesicular fraction. In our study, this fraction was not separable from plasma membranes, as substantiated by levels of plasma membrane markers as high as in the plasma membrane fraction and by a guanine nucleotide-dependent adenylate cyclase activity.