Nonuniform conduction may cause block/delay, providing a substrate for the onset and maintenance of reentrant atrial arrhythmias. We tested the hypothesis that linear, triple site, bipolar rapid pacing (LTSBRP) of the right atrium (RA) generates more uniform wave front propagation compared to single site bipolar rapid pacing (SSBRP), thereby reducing and/or eliminating conduction block/delay otherwise present. Methods: Five pericarditis and 3 normal dogs were studied. Three plunge wire electrode pairs were placed 5-7 mm apart in both a perpendicular and a parallel configuration at the superior aspect of the crista terminalis (CT), and used to pace at 200 and 300 ms cycle lengths for up to 6 sec. During pacing, 380 electrograms were recorded simultaneously from electrode arrays placed epicardially on the atria, producing activation sequence maps for each pacing episode. Local conduction velocity vectors were computed for each site during each episode. Histograms of absolute velocity vector angles from the x-axis (CT) were plotted to assess uniformity of wave front propagation, and the magnitude of each vector was computed to assess the local speed. Results: LTSBRP showed: (1) more uniform linear activation wave fronts compared to SSBRP; (2) velocity vectors with a more uniform magnitude and direction compared to SSBRP; (3) a predominant absolute velocity vector angle vs. a scattered angle distribution with SSBRP; and (4) shorter RA activation time and faster mean epicardial speed than SSBRP for each pacing CL. Conclusion: LTSBRP created a more uniform wave front propagation with less or no conduction block/delay compared to SSBRP.