The objective of this work is to remove ethanol from an alcoholic drink by using the adaptable biochemistry of yeast in a Membrane Bioreactor. Two yeast types (small and normal sized Saccharomyces cerevisiae) stored with different carbon sources of ethanol and glucose were used to carry out aerobic batch culture stirred tank fermenter experiments, using ethanol as the sole carbon substrate, to obtain Monod kinetic parameter values for use in theoretical mathematical models, derived in this work for an ideal Membrane Bioreactor. Aerobic batch culture experiments on ethanol as the sole carbon substrate revealed that two growth phases and two or three substrate uptake phases were present within the time course of a single batch culture experiment. The formation, extrusion and re-uptake of ethyl acetate was seen to occur during the time course of the aerobic batch culture of yeast on ethanol as the sole carbon substrate. Mass transfer experiments on the Membrane Bioreactor were carried out to obtain the overall mass transfer coefficient for ethanol across the dialysis membrane from the beer stream into the yeast stream, for use in the design of an ideal Membrane Bioreactor, and the testing of an experimental bioreactor. Membrane Bioreactor experiments were carried out in order to investigate the removal of ethanol from an alcoholic liquor. These experiments revealed that the Membrane Bioreactor was initially rate controlled by the bioreaction of the yeast in the system, which then switched to mass transfer rate control about half way through the experiment. Comparison of the experimental data with the ideal theoretical model showed that a change in yeast kinetic values, during the time course of the experiment, caused the experimental data to deviate away from the theoretical model which used constant kinetic values for the entire analysis of the experimental run. Recommendations for future development work have been made.