Fetal mouse hearts develop tachycardia in response both to norepinephrine and to glucagon, but although adenylate cyclase is stimulated and adenosine 3':5'-monophosphate (cyclic AMP) elevated by norepinephrine, no measurable changes are produced by glucagon. To test further the possible independence of glucagon chronotropy from the cyclic AMP system, the effects of a phosphodiesterase inhibitor were evaluated. The dose-response curve to norepinephrine was shifted to the left by the phosphodiesterase inhibitor 4-(3,4-dimethoxybenzyl)-2-imidazolidinone (Ro7-2956), but the dose-response curve to glucagon was unaltered. Thus, 10(-6) M norepinephrine produced an increase of 40 +/- 5 beats/min in hearts pretreated with Ro7-2956, as compared to an increase of 22 +/- 3 in control hearts (P less than .01). In contrast, 10(-6) M glucagon produced a rate increase of 25 +/- 4 beats/min in treated hearts vs. 26 +/- 4 beats/min in controls. These data are compatible with the hypothesis that adenylate cyclase and cyclic AMP are involved in the chronotropic response of the fetal mouse heart to norepinephrine but not to glucagon.