The idea that fibrillation is only possible in hearts exceeding a critical size was introduced by W. Garrey >80 years ago and has since been generally accepted. In ventricular tissue, this critical size was originally estimated to be 400 mm(2). Recent estimates suggest that the critical size required for sustained reentry is approximately 100 to 200 mm(2), whereas 6 times this area is required for ventricular fibrillation. According to these estimates, fibrillation is not possible in the mouse heart, where the ventricular surface area is approximately 100 mm(2). To test whether sustained ventricular fibrillation could be induced in such an area, we used a high-speed video imaging system and a voltage-sensitive dye to quantify electrical activity on the epicardial surface of the Langendorff-perfused adult mouse heart. In 6 hearts, measurements during ventricular pacing at a basic cycle length (BCL) of 120 ms yielded maximum and minimum conduction velocities (CV(max) and CV(min)) of 0.63+/-0.04 and 0.38+/-0.02 mm/ms, respectively. At a BCL of 80 ms, CV(max) and CV(min) changed to 0.55+/-0.03 and 0. 34+/-0.02 mm/ms. Action potential durations (APDs), measured at 70% repolarization at those pacing frequencies were found to be 44.5+/-2. 9 and 40.4+/-2.6 ms, respectively. The wavelengths (CVxAPD) were calculated to be 28.6+/-3.4 mm in the CV(max) direction and 16.8+/-1. 5 mm in the CV(min) direction at BCL 120 ms. Wavelengths were significantly reduced (P<0.05) at BCL 80 ms (CV(max), 22.2+/-1.8 mm; CV(min), 13.7+/-0.9 mm). In 5 hearts, stationary vortex-like reentry organized by single rotors (4 of 5 hearts) or by pairs of rotors (1 of 5 hearts) was induced by burst pacing. In the ECG, the activity manifested as sustained monomorphic tachycardia. Detailed analysis showed that the local CVs were reduced in the vicinity of the rotor center, which allowed the reentry to take place within a smaller area than was calculated from wavelength measurements during pacing. In 4 of 7 hearts, burst pacing resulted in a polymorphic ECG pattern indistinguishable from ventricular fibrillation. These data challenge the critical mass hypothesis by demonstrating that ventricular tissue with an area as small as 100 mm(2) is capable of undergoing sustained fibrillatory activity.