Recent studies in our laboratory demonstrated that 30% of normal mongrel dogs have platelets that aggregate when stirred with sodium arachidonate (ASP), while the majority have platelets that do not aggregate under similar circumstances (AIP). In the current study, we confirmed previous observations that indicated that AIP formed aggregating products of prostaglandin metabolism (endoperoxides and/or thromboxane A2) when stirred with sodium arachidonate. We also determined that the difference between ASP and AIP was attributable to the platelets rather than the plasma. To further define the mechanism responsible for the difference between ASP and AIP, we studied the response of dog platelets to agents known to alter platelet cyclic AMP, and we measured platelet cyclic AMP. Radioimmunoassay of cyclic AMP in platelet-rich plasma (PRP) and in washed platelets demonstrated no differences between unstimulated ASP and AIP. However, agents that lower elevated platelet cyclic AMP (epinephrine, norepinephrine, adenosine diphosphate, serotonin and low concentrations of prostaglandin E2) uniformly converted AIP to ASP. Agents that elevate platelet cyclic AMP (prostaglandins E1, D2, I2, theophylline and dibutyryl cyclic AMP) uniformly converted ASP to AIP. The addition of epinephrine, following prostaglandin I2 (PGI2) restored the ability of ASP to aggregate when stirred with sodium arachidonate. In vitro PGI2 (0.14-2.1 nM) converted ASP to AIP without inhibiting adenosine diphosphate (ADP) or collagen-induced aggregation, thereby duplicating the behavior of naturally occurring AIP. PGI2 infusions produced the same result in vivo. The results of these studies suggest that cyclic AMP plays an important role in regulating the response of dog platelets to the aggregating products of prostaglandin metabolism. AIP have a cyclic AMP-dependent mechanism that modulates their response to the aggregating products of prostaglandin metabolism that is different from ASP and human platelets We postulate that AIP retain the effects of in vivo elevation of platelet cyclic AMP stimulated by exposure to PGI2. The mechanism by which this occurs is uncertain.