To characterize the adenylate cyclase system of the primate corpus luteum, we assayed the conversion of [alpha-32P] ATP to [32P]cAMP in preparations of luteal tissue obtained from rhesus monkeys on days 17-19 of the menstrual cycle. Basal, gonadotropin (hCG; 250 nM)- sensitive, and guanine nucleotide (5'-guanylyl-imidodiphosphate [GMP-P(NH)P]; 10 microM)-sensitive cAMP production were influenced by the pH, osmolality, and ionic strength of the assay buffer. As the concentration of Mg+2 increased from 2 to 10 mM, adenylate cyclase activity was enhanced; however, the relative stimulation by hCG plus GMP-P(NH)P was maximal when the Mg+2 concentration approximated that of ATP plus EDTA. In contrast, the presence of Ca+2 inhibited basal, hCG-stimulated, and GMP-P(NH)P-stimulated cAMP production. Adenylate cyclase activity was substrate dependent at ATP concentrations from 0.7-4.5 mM; however, higher concentrations of ATP did not alter cAMP production. The relative stimulation by hCG and GMP-P(NH)P was independent of ATP levels when the ATP to Mg ratio was constant. The rate of cAMP production was constant during 30 min of incubation at 37 C, with the ATP concentration maintained at greater than 87% of initial levels. Adenylate cyclase activity was 10-fold greater in luteal tissue from the superovulated rat than in that from the cycling rhesus monkey; however, relative stimulation by hCG and GMP-P(NH)P was qualitatively similar in the two species. Thus, we have demonstrated the existence of an adenylate cyclase system in the corpus luteum of the rhesus monkey during the menstrual cycle and have established some general properties and optimal assay conditions for the gonadotropin-sensitive adenylate cyclase.