The combined application of ozone and hydrogen peroxide represents a kind of advanced oxidation for water treatment. The radicals that are generated during the process are used for the degradation of organic pollutants from groundwater and industrial effluents. The aim of our study was to evaluate the possible microbicidal, and particularly virucidal, efficacy of such a process, since no substantial data were available. The investigations were performed at a pilot plant installed for the elimination of perchloroethylene from polluted groundwater (reduction efficacy for perchloroethylene from 26 microg/L to 5 microg/L). To enable a reliable evaluation of the microbicidal effect, a set of alternate test organisms was used. As model viruses we chose bacteriophages MS2 (F+ specific, single-stranded RNA), phiX174 (single-stranded DNA) and PRD-1 (coated, double-stranded DNA). Furthermore, spores of Bacillus subtilis were included as possible surrogates for protozoa and Escherichia coli as representative for traditional indicator bacteria used in water analysis. The microbicidal efficiency was compared to the inactivation by means of ozone under two standard conditions (20 degrees C): (a) 0.4 mg/L residual after 4 min and (b) 0.1 mg/L residual after 10 min. Surprisingly, a good microbicidal effect of the ozone/hydrogen peroxide process was found. This was somewhat unexpected, because we had assumed that the disinfection potential of ozone would have been interfered with by the presence of hydrogen peroxide. Escherichia coli and the three test viruses revealed a reduction of about 6-log. In contrast, spores of Bacillus subtilis showed after the total process a reduction of 0.4-log. These results matched the effect of the ozone treatment (a) with a residual of 0.4 mg/L after 4 min contact time (20 degrees C). The test condition (b) with a residual of 0.1 mg/L ozone after a contact time of 10 min at 20 degrees C gave a higher reduction of the B. subtilis spores (1.5-log). The presented study revealed a satisfying microbicidal efficacy of the ozone/hydrogen peroxide process with respect to vegetative bacteria and viruses (bacteriophages). However, it has to be emphasised that intense mixing and sufficient contact time have to be optimised and tested for each individual installation.