The interactions of divalent cations, Mn 2+, Mg 2+, and Ca 2+, with the cytosolic guanylate cyclase (GTP pyrophosphate-lyase (cyclizing, EC 4.6.1.2) from different tissues were studied. Guanylate cyclase activities of the kidney, liver, and lung were strongly dependent on Mn 2+. In contrast, the enzyme in smooth muscle of the colon, aorta, and vas diferens was active with Mg 2+ as well as with Mn 2+. Ca 2+ was ineffective in all tissues. Preincubation, at 30°C, of colon extracts, but not those of kidney and liver, increased guanylate cyclase activity. The Mg 2+-dependent activity was preferentially enhanced by this treatment. These results suggest that when the enzyme was autoactivated by endogenus factors it became more Mg 2+ dependent. Dithiothreitol strongly inhibited the Mg 2+-dependent colon enzyme, whereas activities in kidney and liver were not affected and the response of the enzyme in lung was intermediate. This suggests that autoactivation involved an oxidative-reductive alteration of the enzyme. Ca 2+ markedly inhibited the Mg 2+-dependent activity in smooth muscles but Mg 2+-dependent activities in lung, liver, and kidney were not influenced appreciably. Exogenous activators, dehydroascorbate and NaN 3, increased guanylate cyclase, assayed with either Mg 2+ or Mn 2+. However, the relative stimulation of the enzyme assayed with Mg 2+ was greater than with Mn 2+. When activated by these exogenous agents, guanylate cyclase in all tissues became inhibitable by Ca 2+. These findings suggest that guanylate cyclase in smooth muscle, as prepared, was in a partially activated form. The endogenous activating factors in colon smooth muscle were heat-stable, largely extractable with chloroform/methanol, and cochromatographed with authentic fatty acids. Arachidonic acid stimulated colon guanylate cyclase and enhancement of the Mg 2+-dependent activity was blocked by Ca 2+. This strongly infers that a significant part of the endogenous activating factors in the colon was fatty acids or their derivatives. The colon activators had only minimal affects on the enzyme in the kidney. Similarly prepared activator extracts from the kidney increased colon guanylate cyclase but did not stimulate the renal enzyme. Thus, the ability of the enzyme to be stimulated by endogenous activators was dependent on the tissue from which the enzyme was derived. The possibility that the interactions of Mg 2+ and Ca 2+ on guanylate cyclase in smooth muscles are of regulatory significance to the contractile response of the muscle was discussed. It was proposed as a working hypothesis that cyclic GMP and Ca 2+ might participate in reciprocal negative feedback mechanisms.