Control of fermentation processes is a complex problem due to the inherent non-lineanties and time varying characterstics of the process. The application of conventional single loop analogue controllers provides poor control due to problems in tuning individual loops and the lack of ability to implement complex controllers. The application of standard optimal techniques is compounded both by the complexity of the process and the lack of adequate models as a result of poorly understood dynamics. The lack of important transducers for product measurement coupled with the time varying parameters in a fermentation process provides a natural test for adaptive control techniques. This thesis includes details of modelling and simulation studies earned out on a Bakers Yeast fermentation process. A mathematical model of the growth of Saccharomyces Cerevisiae which desenbes oxidative and aerobic fermentative growth on glucose is presented. The parameters which influence the growth phases of the yeast organism are identified by Recursive Least Squares as part of an overall adaptive control technique. An integrated approach is presented for the on-line estimation of the state of a biochemical reactor from presently attainable real time measurments State estimation by the presented method of Kalman filtering and the above parameter estimation technique is used for the development of an adaptive control scheme. Details of the pilot-plant, instrumentation and computer systems are desenbed highlighting the practical problems m these areas and the means by which these problems have been overcome. Results are presented to show the sucessful performance of the adaptive technique and this work indicates that the application of an adaptive technique could provide great opportunities for the enhancement of conventional control of fermentation processes.