The postnatal development of calmodulin-stimulated phosphorylation of endogenous proteins, in particular the autophosphorylated subunits of the calmodulin-stimulated protein kinase II, were investigated in subcellular fractions of rat cerebral cortex. The major subunit had a mol. wt. of 53,000 Da (designated 50 kDa) and the minor one a mol. wt. of 63,000 Da (designated 60 kDa). The 50-kDa subunit was found to be the only significant phosphoprotein in each fraction and throughout development at its molecular weight. However, the 60-kDa subunit was found to comigrate with other phosphoproteins that accounted for up to 15% of the radioactivity at this molecular weight and which differed between the fractions. 50-kDa autophosphorylation was found to be 3-fold greater in cytoplasmic fractions at day 10 and by adults was evenly distributed between membrane and cytoplasmic fractions. A similar pattern was also found for the total calmodulin-stimulated phosphorylation. Changes in autophosphorylation activity of the 50-kDa subunit were found to represent changes in kinase activity rather than alterations in phosphatase activity. In the membrane, this change was shown to be due to changes in the amount of enzyme. Although in the adult autophosphorylation activity is evenly distributed between membrane and soluble fractions, when differences in phosphatase activity and lack of autophosphorylation activity of the majority of post-synaptic density-associated kinase is taken into account, it is clear that the vast majority of the enzyme is membrane-bound. Phosphorylation of endogenous substrates paralleled the development of 50-kDa subunit autophosphorylation, most of which occurred between day 14 and day 30, a period which follows the most rapid phase of synaptogenesis. This pattern was different from that of the phosphorylation of myelin basic protein and two substrates of the calcium-phospholipid-dependent protein kinase. There was also a change in the ratio of autophosphorylation activity of the 50-kDa and 60-kDa subunits during development which appears to be due to a change in the amount of the subunits themselves. This ratio was the same in all fractions at any one age. We suggest that this change is due to the existence of at least two developmentally regulated isoenzymes in the cortex.