The broad aim of this research was to evaluate a cyclic fed batch culture system for the production of secondary metabolites and to identify a physiological parameter that correlated with product induction for optimisation of the process. A cyclic fed batch culture (CFBC) system was successfully established using a constant flow of glucose limited medium into the bioreactor, and culture stability was found to be dependent on time of flow initiation. A 40% increase in product concentration was obtained using CFBC when compared to batch culture with the organism Saccharopolyspora erythraea. CFBC induced secondary metabolite production with the organism Amycolatopsis orientalis, despite the lack of production in batch culture. Product concentration was dependant on the growth rate range used in CFBC for both organisms, although the trends of productivity differed for the two. Product stability at quasi steady-state during CFBC depended on the variation in biomass concentration obtained. A number of physiological parameters were measured in both batch culture and CFBC for correlation with product induction. A decrease in protein synthesis rate was observed to precede the induction of antibiotic in CFBC with both organisms, however, this trend was not observed in batch culture with Amycolatopsis orientalis. Although complex trends were observed for levels of RNA, total protein and the ratio of the two, no correlation with induction was observed. Profiled feeding regimes were employed for the manipulation of protein synthesis rates to optimise conditions for production. Increased productivity was not obtained, although protein synthesis rate was manipulated to reduce the timing of product induction at the start of the cycle. Morphological examination of the filamentous organism Saccharopolyspora erythraea, grown in either CFBC, chemostat culture or variable volume-chemostat culture, was found to indicate the organisms' physiological state. Productivity was highest in CFBC, coinciding with a transition from exponential to linear morphological growth and decreasing growth rate through the progress of one cycle. Where the morphological growth rate remained constant, that is in variable volume chemostat, productivity was lowest. In chemostat culture a morphological distribution was observed with both exponential and linear growth resulting in a product concentration higher than found in variable volume chemostat but less that in CFBC. Future work should include molecular studies to determine events leading to secondary metabolite induction. In conjunction with these studies development and unstructured model of the physiology is required. Clarification of the role of protein synthesis rates and particular components of the mechanism are required before this parameter can be discarded.