Efficient processing of cassava roots by wet milling requires overcoming challenges associated with disaggregation of the starch-containing parenchyma cells. These cells entrap starch granules and hinder their release during wet milling. Steeping of ground cassava in 0.75% (w/v) NaOH in combination with wet milling was investigated to determine whether and how dilute NaOH modifies cassava cell walls. Gas chromatography (GC) data of cell wall constituent sugar composition and Fourier transform infrared (FTIR) data showed that NaOH steeping caused solubilisation of the cell wall pectin fraction. FTIR and wide-angle x-ray scattering (WAXS) spectroscopy indicated that NaOH steeping combined with fine (500 ?m opening screen size) wet milling reduced cellulose crystallinity. Dilute NaOH steeping also produced pits (micropores) through the cell wall structure as shown by scanning electron microscopy (SEM). The micropores seemed to have weakened the cell walls, as revealed by increased cellular disaggregation as viewed by light microscopy. Disaggregation of cassava root cells was associated with a reduction in large (diameter > 250 ?m) residue particle size in the bagasse and consequently more starch yield. Thus, it seems that mechanistically, dilute NaOH solubilisation of alkaline-soluble pectin weakens the cell walls of starch-containing cassava root parenchyma cells. Weakening of cassava cell walls with a combination of biological (14 day endogenous fermentation under microaerophilic conditions) and dilute alkaline pre-treatment (0.75% NaOH steeping) was investigated in an attempt to further increase starch yield by wet milling. However, the combined pre-treatment resulted in approx. 11.8% more starch yield, slightly less than the 12.3% increase obtained by using endogenous fermentation alone. The absence of an additive effect was probably because although endogenous fermentation (retting) and dilute NaOH steeping weakened cassava cell walls through different mechanisms (hydrolysis/solubilisation of pectin), the resultant loss in pectin cohesiveness was similar. Solid state fermentation of ground cassava using various alkaliphilic Bacillus spp. starter cultures separately and in combination was also investigated to determine their extracellular hydrolytic enzyme induced weakening effects on parenchyma cell walls. GC and FTIR data indicated that fermentation with Bacillus akibai + endogenous microflora (EM), B. cellulosilyticus + EM, B. hemicellulosilyticus + EM and B. spp. in combination + EM caused reduction in cell wall pectin, xyloglucan and cellulose contents. Cell wall solubilisation/hydrolysis seemed to have primarily involved the amorphous constituents, as indicated by an increase in cellulose crystallinity by WAXS spectroscopy. Enzyme assay and SEM indicated that Bacillus spp. extracellular cellulase and polygalacturonase weakened the cell walls through formation of micropores and possible rupturing of cellulose microfibril structures. These modifications seemed to have aided disaggregation of the cassava parenchyma cells and consequently liberation of more starch granules as indicated by light microscopy. Fermentation with B. akibai + EM, B. cellulosilyticus + EM, B. hemicellulosilyticus + EM and B. spp. in combination + EM also resulted in less large (diameter > 250 ?m) residue particle size in the bagasse and consequently higher starch yield. Thus, dilute NaOH steeping and fermentation with alkaliphilic Bacillus spp. starter cultures are techniques capable of improving the effectiveness of wet milling in disintegrating cassava cell walls. However, with regard to the demand for environmentally cleaner production, potential utilisation of alkaliphilic Bacillus spp. and more specifically Bacillus cellulosilyticus appears more promising.