Fermentation is commonly used to produce food materials (beverages, dairy products), renewable fuels (hydrogen, ethanol), pharmaceuticals (antibiotics) and industrial chemicals (acetate, butyrate). In industrial fermentation, a specialised, pure microbial culture is normally used to generate specific products. This requires expensive, sterile production conditions with high-quality raw materials. In contrast, Mixed Culture Fermentation (MCF) uses environmentally ubiquitous organisms to produce a mixture of products depending on the environmental conditions. As they are sourced from the environment, mixed cultures do not require expensive culture maintenance. In addition, they are capable of dealing with a mixture of substrates of variable composition and non-sterile feeds. This has the potential to reduce costs, increase reactor loading rates, and allow for continuous reactors, as opposed to batch operation. MCF is a preferable, flexible process in that can continuously manipulate product mixtures by changing operational condition. The key limitation to industrial implementation of MCF is the difficulty in predicting product formation based on operational conditions. This is due to a lack of understanding of how operational factors affect the various pathways, and hence the product spectrum, with pH being the most commonly manipulated process variable. This thesis attempts to further analyse the link between operational conditions, microbial community, and product spectrum. Two experiments were done, the first focusing on mode of pH manipulation in a continuous reactor, and the second focusing on batch operation at different pH levels, and with different inoculums. The continuous study varied pH from 4 to 8, with one experiment varying the pH from 4 to 8 progressively (progressive), and the other resetting pH back to 5.5 as an intermediate point (reset). The reset regime resulted in a highly dynamic community, shifting from Clostridia at low pH to Klebsiella, and a more dynamic product spectrum, with specifically ethanol being produced at high pH. The progressive regime resulted in relatively flat microbial community, with less dynamic product spectrum. Kinetic experiments done on the same reactors, varying also hydrogen partial pressure through nitrogen sparge rate, and using a membrane inlet mass spectrometer (MIMS) as measurement technique emphasised a number of time constants, including direct chemical response to pH change (1000 d-1), liquid and gas response to changes in gas flow and to pH change, likely related to mass transfer characteristics (100 d-1), and biological responses, mainly measured as ethanol (1 d-1). Changes in microbial community are even slower than this. The second batch indicated that the major factor influencing rate and spectrum response was the inoculation pH, with biomass inoculated at pH4 being very slow in batch (1 d-1), and producing mainly ethanol, and biomass inoculated at higher pH levels being pH 6 (3 d-1) and 8 (10 d-1) being progressively faster. The slowest batches were at batch pH 4 for each inoculum regardless of the inoculation pH. The results from the batch work hence contrast with the continuous work in that the primary driver is inoculum history rather than current conditions. Based on the overall thesis, microbial communities are multi-capable, with very different communities achieving the same outcomes in terms of product spectrum (with acetate-ethanol-butyrate spectrums dominating), and microbial community is a steering or filtering factor rather than the primary factor (which remains environmental conditions.