An experimental direct methanol fuel cell (DMFC), designed and manufactured in-house, was used in this study. The cell is of standard filter-press configuration with parallel rectangular single-pass anode channels. The membrane electrode assembly (MEA), with a suitable Pt-Ru anode electrocatalyst, was purchased from E-TEK Inc. A 1.0 M methanol in water solution was used as the fuel and pure oxygen was used as the oxidant in all experiments. Three graphite anode plates were machined with the same Bow channel configuration but each with different depth of channels. The cathode was kept the same for all experiments. Polarisation curves and ac impedance spectra were obtained for varying temperatures and channel depths. To separate the contribution of the oxygen reduction reaction to the overvoltage from the: anode and membrane contributions, reference hydrogen electrode (RHE) measurements were taken. By comparing the RHE polarisation with the methanol-oxygen polarisation experiments, it was: found that polarisation losses at the oxygen cathode accounted for a 40-50% of the overpotential. The variation in the performance of the cell with flow of methanol/water mix, with temperature and with current density was studied. Polarisation measurements indicate that the medium channel depth flow channels performed better than either the shallow depth or deep depth how channels indicating that there is a complex relationship between the effect of Bow velocity and the influence of the: rate of production of product CO2. AC impedance spectroscopy measurements confirmed the observed polarisation results. This method proved to be able to provide a reliable indication of the performance of the cell even when the cell had not yet achieved steady-state. In the case of the shallow channel depth anode, ac impedance revealed that it required considerably longer to achieve steady-state than the time required for the medium and deep channel depths.