1. The effect of histamine on cellular cyclic AMP levels in cultured bovine adrenal medullary cells has been studied. 2. Histamine (0.3-30 microM) increased cyclic AMP levels transiently, with a maximal response after 5 min, a smaller response after 20 min, and no increase seen after 80 or 180 min. The EC50 at 5 min was approximately 2 microM. Histamine had no effect on cyclic AMP release from the cells over 5 min, but increased it after 90 min. 3. The cyclic AMP response to 5 microM histamine was reduced by 45% by 1 microM mepyramine and by almost 30% by 1 microM cimetidine, and was abolished by the combination of both antagonists. Cimetidine at 100 microM did not inhibit the response to histamine more than 1 microM cimetidine. The H3-receptor antagonist, thioperamide (1 microM), had no effect on the response to histamine. 4. The H1-receptor agonist, 2-thiazolyethylamine (5-100 microM) and the H2-receptor agonist, dimaprit (5-100 microM), each induced a cyclic AMP response, and gave more-than-additive responses when combined. The H3 agonist (R) alpha-methylhistamine (100 microM) had no effect either on its own or in combination with either the H1 or the H2 agonist. The response to 100 microM 2-thiazolylethylamine was unaffected by cimetidine (100 microM). 5. The cyclic AMP responses to 5 microM histamine, 100 microM thiazolylethylamine and 100 microM dimaprit were each weakly enhanced in the presence of 1 mM 3-isobutyl-1-methylxanthine. The response to dimaprit was enhanced more than 10 fold in the presence of 0.3 microM forskolin, while the responses to histamine and thiazolylethylamine were weakly enhanced.6. The cyclic AMP response to 5 microM histamine was partially reduced in the absence of extracellular Ca2 and the residual response was fully antagonized by 1 microM cimetidine and was unaffected by 1 microM mepyramine.In the absence of Ca2 , the cyclic AMP response to 100 microM thiazolylethylamine was abolished, while that to 100 microM dimaprit was unaffected.7. Reincubation of 5 microM histamine solutions with a second set of chromaffin cells, following prior incubation with another set of cells, induced a cyclic AMP response in the fresh cells. This response was reduced by a combination of mepyramine and cimetidine to the same degree as the response to fresh 5 microm histamine solutions.8. The results indicate that histamine increases cellular cyclic AMP levels in bovine chromaffin cells by three mechanisms: by acting on H1 receptors, by acting on H2 receptors, and by an interaction between H, and H2 receptors. The H1 response does not require concomitant activation of H2 receptors, is fully dependent on extracellular Ca2 +, does not depend on secreted chromaffin cell products, and is not due to reduced cyclic AMP degradation or export. The H2 cyclic AMP response is the first functional response reported for H2 receptors on chromaffin cells, is independent of Ca2 , is not due to reduced cyclic AMP export or degradation, and is likely to be mediated via a direct action through Gs. The role of these different mechanisms in the regulation of cyclic AMP-dependent processes in chromaffin cells by histamine is under investigation.