Cyclic voltammograms for the reduction of 2-methyl-2-nitropropane at a mercury electrode in acetonitrile solvent have been analyzed carefully by the "global" method. The dependence of the logarithm of the heterogeneous rate constant on the potential is only mildly parabolic, but statistical analysis shows that the curvature is nonetheless genuine and cannot be explained by random errors. The sign of the curvature is that predicted by the Marcus theory and the magnitude of the effect is also compatible with theory. No trend with scan rate is observed in any of the measured kinetic, thermodynamic or transport parameters.