This paper investigates the hypothesis that the minimum concentration capable of exerting a selection for oxytetracycline-resistant bacteria would be the minimum concentration of oxytetracycline to have a perceptible effect on the growth of oxytetracycline-sensitive bacteria. The investigation studied two variants of Escherichia coli J 53 differentially marked with tet(r) or with kan(r) resistance genes introduced by pUT mini-Tn 5 suicide plasmids. Spectrophotometry and indirect conductance technology were used to determine the minimum concentration of oxytetracycline that produced a detectable effect of the growth or the metabolism of the oxytetracycline sensitive strain. The minimum selection concentration was determined in batch cultures for mixed populations of oxytetracycline-resistant and oxytetracycline-sensitive strains. Results show that, when determined under three different conditions, the minimum effect concentrations (MEC) and the minimum selection concentrations (MSC) were equivalent. It is therefore proposed that measures of MEC can act as proxy measure of MSC. These data represent a contribution to the theoretical understanding of the mechanisms by which antibacterial agents select for increased frequencies of resistant bacteria. They represent a first step in the development of models capable of predicting the environmental impact of these agents. (C) 1999 Elsevier Science B.V. All rights reserved.