This project aimed to refine and validate an optical remote sensing method to predict the ultimate pH of slaughter beef animals. An existing commercial method converts muscle glycogen in a known mass of muscle sample into glucose that is determined by a diabetic’s personal meter. The method is expensive in terms of consumables and results are fraught with inadequate operator skill levels. Pilot studies showed that it may be possible to measure the mass of the muscle sample and the concentration of glucose by colorimetry. Redness was a measure of muscle mass in acetate-buffered slurry, and after addition of Fehlings solution and heating, yellowness was a measure of glucose. This was the starting point for the study. Phase 1 determined the value of individual Hunter colour a*, b* and L* values for predicting mass of meat samples by linear equations. Hunter a* was a useful predictor of meat mass, but only within animals, probably due to the different muscular origins of the meat cuts selected. However, it was proposed that if samples were taken from a single muscular site, as in the existing commercial method, among animal variability might be much reduced. In Phase 1, a digital camera was also used to extract colour data, but it proved much less useful than the Hunter meter. Its use was thus discontinued. Phase 2 showed that different concentrations of glucose did not affect the colour due to meat mass, which was a necessary condition for using colour as a predictor of meat mass. Phase 3 explored the broad relationship between glucose concentration and meat mass on colour change due to the Fehlings reaction induced in a microwave oven. As expected from prior research, the concentration of glucose strongly affected the heat-induced colour, but meat mass also affected colour presumably through the Maillard reaction which would compete with the Fehlings reaction for the available glucose. However, if the mass of meat were known, colour values could be adjusted for this effect. In Phase 4, randomly chosen but defined masses of meat, and similarly randomly chosen, defined concentrations of glucose were used with the Fehlings reaction to test the predictive value of equations relating concentration of glucose/mass of meat to various Hunter colour values. The ratio was well predicted by Hunter b* and L*, unexpectedly implying that information about meat mass and glucose could be simultaneously extracted from the same colour data. This result suggests that there may be no need to measure meat mass, gravimetrically or by colour, to get useful results. In a limited way, Phase 5 extended the Phase 4 work by using the ratio of colour values before heating (no Fehlings added) colour values after heating (Fehlings added) to see if this would improve the predictive values established in Phase 4. It did not. The results are discussed with a focus on future work required to confirm the results in Phase 4, and also describe the steps required in a hypothetical semi-automated application of the technology.