The production rate of a bacteriocin, produced by Lactobacillus plantarum TMW1.25 and previously named plantaricin1.25, was studied during pi-I-constant batch fermentations under various growth media conditions. The growth of L. plantarum and production of bacteriocin during the retardation phase were modelled, using 11 different empirical and mechanistic approaches. The optimal pH for bacteriocin production was 4.5. Among the different nitrogen sources tested, yeast extract was the most important, on the basis of the fact that the maximum growth rate decreased 16% without yeast extract, and only 7.2% or 8.1% without meat extract or peptone respectively. However, the change of nitrogen source did not have a significant effect on bacteriocin production. The progression of plantaricin1.25 production during the retardation phase and growth of L. plantarum TMW1.25 could be described by a structured model in which the bacteriocin concentration induces its own production. Among those models not implementing bacteriocin induction, only the one with an exponential increase of bacteriocin yield per unit biomass was suitable to describe bacteriocin production. Computer-aided evaluation of experimental data appears to be helpful in elucidating the relationship between the growth of lactic acid bacteria and bacteriocin production.