What is the central question of this study? Can we experimentally replicate atrial pro-arrhythmic phenotypes associated with important chronic clinical conditions, including physical inactivity, obesity, diabetes mellitus and metabolic syndrome, compromising mitochondrial function, and clarify their electrophysiological basis? What is the main finding and its importance? Electrocardiographic and intracellular cardiomyocyte recording at progressively incremented pacing rates demonstrated age-dependent atrial arrhythmic phenotypes in Langendorff-perfused murine Pgc1β-/- hearts for the first time. We attributed these to compromised action potential conduction and excitation wavefronts, whilst excluding alterations in recovery properties or temporal electrophysiological instabilities, clarifying these pro-arrhythmic changes in chronic metabolic disease. Atrial arrhythmias, most commonly manifesting as atrial fibrillation, represent a major clinical problem. The incidence of atrial fibrillation increases with both age and conditions associated with energetic dysfunction. Atrial arrhythmic phenotypes were compared in young (12-16 week) and aged (>52 week) wild-type (WT) and peroxisome proliferative activated receptor, gamma, coactivator 1 beta (Ppargc1b)-deficient (Pgc1β-/- ) Langendorff-perfused hearts, previously used to model mitochondrial energetic disorder. Electrophysiological explorations were performed using simultaneous whole-heart ECG and intracellular atrial action potential (AP) recordings. Two stimulation protocols were used: an S1S2 protocol, which imposed extrasystolic stimuli at successively decremented intervals following regular pulse trains; and a regular pacing protocol at successively incremented frequencies. Aged Pgc1β-/- hearts showed greater atrial arrhythmogenicity, presenting as atrial tachycardia and ectopic activity. Maximal rates of AP depolarization (dV/dtmax ) were reduced in Pgc1β-/- hearts. Action potential latencies were increased by the Pgc1β-/- genotype, with an added interactive effect of age. In contrast, AP durations to 90% recovery (APD90 ) were shorter in Pgc1β-/- hearts despite similar atrial effective recovery periods amongst the different groups. These findings accompanied paradoxical decreases in the incidence and duration of alternans in the aged and Pgc1β-/- hearts. Limiting slopes of restitution curves of APD90 against diastolic interval were correspondingly reduced interactively by Pgc1β-/- genotype and age. In contrast, reduced AP wavelengths were associated with Pgc1β-/- genotype, both independently and interacting with age, through the basic cycle lengths explored, with the aged Pgc1β-/- hearts showing the shortest wavelengths. These findings thus implicate AP wavelength in possible mechanisms for the atrial arrhythmic changes reported here.