A linear relationship has been found under certain conditions between the 1st order flotation rate constant, k, and Sb, the bubble surface area flux produced in a flotation cell. Sb is the total surface area of bubbles moving up through a unit cell cross-sectional area per unit time and gives a measure of the hydrodynamic and gas dispersion properties in the cell. This relationship has been found to be independent of the size of the cell and, in mechanical cells, independent of the type of impeller used ie. method of Sb production, which allows scale-up possibilities. These findings are, however, only related to metalliferous flotation in conventional mechanical cells and to date no work has been carried out to study this effect in coal or non-conventional cells. The aims of this thesis were to firstly determine whether the k-Sb relationship, which has found to be an effective scale-up tool for base metal flotation, is evident in coal flotation and secondly, to determine whether it exists for non-conventional flotation technologies, commonly used in coal flotation. Testwork was performed at two industrial sites and consisted of: . • Site measurements of bubble size, superficial air velocity and air hold-up to determine the value of Sb in operating plant scale cells. This characterised the particular cell under investigation and identified the existing hydrodynamic conditions. The measurements were combined with conventional sampling of feed and products followed by analysis and mass balancing to reveal the prevailing rate constant k. . • Simultaneous operation of the JKMRC 60 litre conventional flotation cell or pilot scale Jameson cell (depending on plant cell being used) in parallel with the site cell under investigation. It was possible to make major changes to the operating conditions of the test cell (as opposed to the site cell) in order to determine the k-Sb relationship for the coal under investigation. . • Measurement of froth recovery in site cells. The effect of froth residence time and other parameters on froth recovery was determined in the 60-litre conventional cell or pilot Jameson cell. • Keeping the cell hydrodynamic conditions (as specified by the Sb value) fixed, and independently evaluating the effect of other parameters such as reagent regime and froth depth, using the 60-litre cell. Results for conventional flotation (and, it is assumed, all coal flotation) showed that particle size and feed composition effects are significant and can override the effect a change in Sb has on cell performance. Decoupling these effects indicated a distinct relationship between Sb and flotation rate. Also, the range of Sb values under which linearity is maintained was found to be limited. Further work to investigate these effects in more detail was unsuccessful, therefore additional testwork is recommended to enable firm conclusions to be made. Feed changes made analysis of the effect of the froth phase inconclusive for the conventional cell, however, the froth phase was found to be a significant factor in the performance of the plant scale Microcel. It was found that conditions inside the downcomer of the Jameson cell appear to be most significant in terms of the k-Sb relationship, indicating that, in line with popular opinion, collection occurs solely within the downcomer of the Jameson cell. Conditions in the pilot scale Jameson cell rig exhibited perfect mixing behaviour with mixing conditions in the plant cell inconclusive but most likely perfect mixing or axial dispersion with large amounts of dispersion. Improvements in gas dispersion measurement techniques in the plant cells, in particular bubble size, need to be developed for specific application to coal flotation to allow further investigation into the findings of this work.