A study based on a statistically designed set of experiments (central composite design) has been made to determine the optimum conditions for grinding of coal/water slurries. The dependent variable in these experiments was fineness of the product while the independent variables were percent solids, ball size, dispersant dosage and mill speed. The optimum conditions found from a response surface analysis not only predicted an increase in grinding efficiency but the product obtained in the region of the optimum showed other desirable slurry properties, especially improved slurry rheology. This analysis has helped to establish the significance and relative importance of the independent variables examined and has been useful in defining the interactions between the operating variables and their effect on the grinding response.