Extracts of fasted rat diaphragms, previously treated with or without insulin were assayed for glycogen synthase, protein kinase and cyclic [3H]-AMP binding. Treatment with insulin produced an elevation in the % of glycogen synthase I and a concurrent decrease in cyclic AMP-dependent protein kinase activity and cyclic [3H]-AMP binding. Analysis of extracts by disc gel electrophoresis demonstrated the inhibition of cyclic [3H]-AMP binding to involve the Type I protein kinase holoenzyme. Inhibition of protein kinase activity was most apparent in the presence of 0.2 micrometer cyclic AMP, with enzymatic activity of the insulin-treated extracts typically 60--65% of control. Higher assay concentrations diminished the difference between control and insulin-treated extracts and concentrations greater than 20 micrometer abolished it. The inhibition of cyclic AMP-dependent protein kinase activity after insulin was a transient and labile phenomenon. The effect was independent of ATP concentration in the assay, but was sensitive to the pH of tissue extraction, requiring a pH of 7.0 to 8.4 to be observed. Insulin-mediated inhibition of protein kinase activity was reversed upon preincubation of extracts at 0--2 degrees. Relatively concentrated homogenates (less than 4 microliter buffer/mg tissue) yielded extracts which exhibited little or no inhibition of protein kinase activity compared to extracts prepared from more dilute (6--10 microliter/mg) homogenates. A model for the inhibition of the cyclic-AMP dependent protein kinase by an insulin-generated inhibitor which becomes directly associated with the Type 1 holoenzyme is proposed.