Polystyrene latex particles (PLP) and zymosan particles (ZP), two commonly employed phagocytic stimuli, were noted to bind to purified human peripheral blood lymphocytes. This interaction was not accompanied by ingestion but did lead to a marked increase in intracellular cyclic AMP. The cAMP response to PLP was proportional to the particle cell ratio which, in turn, correlated with the number of membrane-associated particles. After the addition of PLP to lymphocytes, the cAMP response occurred within 2 min, peaked between 4 and 15 min, and returned to baseline by 30 to 60 min. The cAMP response to ZP was similar in onset and duration to that seen with PLP but was less marked (2- to 4-fold vs 25- to 50-fold) and more variable in magnitude. This is probably a reflection of the smaller number of cells interacting with ZP. At high PLP to cell ratios almost all of the lymphocytes bound PLP but only 10 to 28% of the mixed lymphocyte population bound ZP. Two lines of evidence established conclusively that the cAMP response was taking place in the lymphocytes themselves rather than in contaminating cells. 1) When lymphocytes were purified additionally by filtration through a nylon wool column (99 to 100% lymphocytes), they were found to undergo a similar cAMP response to PLP. Since the nylon filtration procedure also removes almost all of the B cells, this further indicates that T cells are capable of undergoing the response. 2) Immuno-fluorescence studies with anti-cAMP antibody revealed an increase in intralymphocytic cAMP which was primarily adjacent to the site of PLP or ZP attachment. The likely explanation of this data is that PLP and ZP perturb the lymphocyte surface leading to regional activation of membrane-bound adenylate cyclase and subsequent cAMP accumulation. Although the physiologic significance of these observations remains to be determined, the results: 1) provide histologic confirmation for the concept of cAMP compartmentalization, 2) clarify conflicting results regarding the localization of cAMP accumulation during the phagocytosis of PLP by mixed leukocyte populations, and 3) suggest that this experimental system may allow an analysis of the mechanism by which perturbations of the lymphocyte surface modulate cAMP.