A comparative kinetic study was carried out on the anaerobic digestion of untreated beet molasses alcoholic fermentation wastewater and beet molasses previously fermented with Penicillium decumbens. Three 1 l volume stirred tank reactors were used for the study, one with freely suspended biomass (Control), and the other two with biomass supported on Saponite (magnesium silicate) and Esmectite (aluminium silicate), respectively. The reactors were batch fed at mesophilic temperature (35 • C) using COD loadings of between 12 and 140 ml of untreated molasses with a COD of 80.5 g/l and of between 43 and 304 ml of molasses previously fermented with a COD of 23.0 g/l. The anaerobic digestion process of both substrates was found to follow a first-order kinetics and the experimental accumulated methane volume (G)-time (t) data to conform to an equation of the form: G = G m [1exp(-K 0 t)], from which the specific rate constants, K 0 , were calculated. In the case of untreated molasses, the K 0 values decreased considerably from 2.87 to 0.10 (Saponite), from 2.78 to 0.10 (Esmectite) and from 2.16 to 0.07 per days (Control) when the loading of molasses applied and initial COD added were increased from 12 to 140 ml and 1 to 10 g/l, respectively; this showed an inhibition phenomenon in the three reactors studied. In contrast to this, the kinetic constants of the anaerobic digestion of pre-treated molasses were virtually constant over the COD range used (1-7 g/l) in the three reactors considered. Finally, the average methane yield coefficient for pre-treated molasses was 305 ml CH 4 STP/g COD removed , viz. 35% higher than that provided by untreated molasses.