AbstractTen chitosan products, prepared as described in part I of this study, were evaluated in jar tests that measured their effectiveness for coagulation of suspended solids and removing turbidity from cheese whey. A polynominal regression analysis was found to be useful for determining the optimal effectiveness of each chistosan preparation, and was expressed as the percent reduction on turbidity per unit concentration of chitosan added. The effectiveness of the chitosan products was found to be inversely related to their molecular‐weight values. This situation was different from the findings described in part II of this study, in which the filterability of activated sludge was tested. Enzymatic deproteination yielded chitosan products that performed better than those produced by alkali deproteination. Demineralized products were also more effective than those that had not been demineralized. The preparations deacetylated under a nitrogen atmosphere were more effective than those deacetylated in air, but this was shown to be true only for the first 5 min of deacetylation. When deacetylated for 15 min, no differences were noted. In this study, differences in performance between the various products were largely due to the differing dosages required to achieve the maximum reduction in turbidity of cheese whey, while the maximum responses achieved by the various products tested were about the same. A commercial product, which was less effective as a sludge coagulating agent in part II of this study, was more effective for cheese whey coagulation and turbidity removal than the majority of the experimental chitosan preparations tested.