Increasingly often soil residual concentrations of pharmaceutical antibiotics are detected, while their ecotoxic relevance is scarcely known. Thus, dose related effects of two antibiotics, sulfapyridine and oxytetracycline, on microorganisms of two different topsoils were investigated. The fumigation-extracted microbial C (E(C)) and ergosterol were determined to indicate soil microbial and fungal biomass, respectively. Microbial activity was tested as basal respiration (BR), dehydrogenase activity (DHA), substrate-induced respiration (SIR), and Fe(III) reduction. The BR and DHA were uninfluenced even at antibiotic concentrations of 1000 microg g(-1). This revealed that an activation of microbial growth through nutrient substrate addition is required to test possible effects of the bacteriostatic antibiotics. In addition, the effects of both antibiotics were time dependent, showing that short-term tests were not suitable. Clear dose-response relations were determined with SIR when the short-term incubation of 4h was extended into the growth phase of the microorganisms (24 and 48 h). The Fe(III) reduction test, with a 7-d incubation, was also found to be suitable for toxicity testing of antibiotics in soils. Effective doses inhibiting the microbial activity by 10% (ED(10)) ranged from total antibiotic concentrations of 0.003-7.35 microg g(-1), depending on the antibiotic compound and its soil adsorption. Effective solution concentrations (EC(10)), calculated from distribution coefficients, ranged from 0.2 to 160 ng g(-1). The antibiotics significantly (p<0.05) reduced numbers of soil bacteria, resulting in dose related shifts in the fungal:bacterial ratio, which increased during 14 d, as determined from analysis of ergosterol and E(C). It was concluded that pharmaceutical antibiotics can exert a temporary selective pressure on soil microorganisms even at environmentally relevant concentrations.