Photoaffinity labeling of β 1- and β 2-adrenergic receptors in plasma membranes from various mammalian tissues has been been performed utilizing the recently developed β-adrenergic antagonist probe [ 125I]para-azidobenzylcarazolol. Tissues studied and their proportions of β 1 and /gb 2 receptors were: rat lung (18% β 1, 82% β 2), rabbit lung (72% β 1, 28% β 2), guinea pig lung (15% β 1, 85% β 2), dog lung (20% β 1, 80% β 2) and rabbit skeletal muscle (10% β 1, 90% β 2). As assessed by autoradiograms of sodium dodecyi sulfate-polyacrylamide gel electrophoresis, two to three specifically protected bands of M r 62,000–65,000, 50,000–55,000 and 38,000–42,000 were observed in each tissue system. In each case, β-adrenergic agonists and antagonists protected against photolabeling with appropriate β 1 and β 2 selectivity. Thus, in rat lung the β 2 selective antagonist ICI-118,551 was more potent in blocking incorporation than the β 1 selective antagonist betaxolol, whereas in rat, dog and guinea pig lung and rabbit skeletal muscle epinephrine was more potent than norepinephrine in blocking labeling, indicating a β 2 specificity in these tissues. Conversely, in rabbit lung membranes, norepinephrine was approximately equipotent with epinephrine in blocking photoincorporation, indicating a β 1 selectivity. In some systems protease inhibitors, especially those specific for metalloproteases (EDTA, EGTA), markedly diminished the amount of the smaller M r peptides. For example, in rat lung the ratio of M r 62,000 :47,000:36,000 peptides changed from 30:40:30 to 60:35:5 in the presence of inhibitors. These results demonstrate the applicability of using [ 125I]para-azidobenzylcarazolol to covalently label mammalian β-adrenergic receptors and suggest that mammalian β 1 and β 2 receptor binding sites primarily reside on peptides of M r 62,000–65,000 and that smaller ligand binding fragments may arise by proteolysis.