The method of Kidwai et al. (1971. Biochem. Biophys. Res. Commun. 45:901) offers a rapid and simple technique for the preparation of membrane fractions from rat heart, using sucrose density centrifugation of a 100,000x g pellet. We have investigated the distribution of adenylate cyclase, cyclic AMP phosphodiesterase, and norepinephrine binding activity in these fractions. Specific activity of adenylate cyclase was high in the plasma membrane (PM) and sarcoplasmic reticulum, as well as the nuclear (N) fractions, but 80-90 percent of the total activity was found in the N fraction. Epinephrine stimulation of adenylate cyclase was present in all fractions but did not exceed 25 percent of basal activity. The activity in the mitochondrial fraction was low and insensitive to epinephrine. About 10 percent of phosphodiesterase activity was found in the 100,000 x g pellet. After density gradient centrifugation, 70 percent of this portion was recovered in the N fraction, with the highest specific activity present in the PM fraction. Binding of [3H] norepinephrine was measured by a membrane filtration technique. Binding activity was found in all fractions, and it paralleled roughly the distribution of adenylate cyclase. However, only about 25-30 percent of the binding was blocked by propranolol, except in the PM fraction where binding was not prevented by this drug. Further, a comparison of adenylate cyclase activities with norepinephrine binding yields a turnover number for the enzyme of the order of 10(-2) sec-1, assuming a 1:1 relationship between beta-adrenergic recptor and adenylate cyclase. Since this value seems unrealistic, we suspect that either the binding activity measured is unrelated to the beta-adrenergic receptor or the receptor to cyclase ratio is considerably larger than unity.