Alcaligenes eutrophus JMP 134 was grown in a recycling-mode fermenter with 100% biomass retention on 2,4-dichlorophenoxyacetic acid (2,4-D), phenol, and fructose. The growth pattern obtained given a constant supply of substrates exhibited three phases of linear growth on all three substrates. The transition from phase 1 to phase 2, considered to correspond to the onset of stringent (growth) control as indicated by a significant increase in guanosine 5(prm1)-bisphosphate 3(prm1)-bisphosphate (ppGpp), took place at 0.016 h(sup-1) with 2,4-D and at about 0.02 h(sup-1) with phenol and fructose. In the final phase, phase 4, which was achieved after the growth rate on the respective substrates fell below 0.003 to 0.001 h(sup-1), a constant level of biomass was obtained irrespective of further feeding of substrate at the same rate. The yield coefficients decreased by 70 to 80% from phase 1 to phase 3 and were 0 in phase 4. The stationary substrate concentrations s(infmin) in phase 4, calculated from the kinetic constants of the strain, were 1.23, 0.34, and 0.23 (mu)M for 2,4-D, phenol, and fructose, respectively. These figures characterize the minimum stationary substrate concentrations required in a dynamic system to keep A. eutrophus alive. This is caused by a substrate flux which enables growth at a rate >=0 due to the provision of energy to an extent at least satisfying maintenance requirements. According to the constant feed rates of the substrates and the final and stable biomass concentrations, this maintenance energy amounts to 14.4, 4.0, and 2.4 (mu)mol of ATP (middot) mg of dry mass(sup-1) h(sup-1) for 2,4-D, phenol, and fructose, respectively, after correction for the fraction of living cells. The increased energy expenditure in the case of 2,4-D is discussed with respect to uncoupling.