Pseudomonas fluorescens-mediated induction of systemic resistance in radish against fusarium wilt (Fusarium oxysporum f. sp. raphani) was studied in a newly developed bioassay using a rockwool system. In this bioassay the pathogen and bacterium were confirmed to be confined to spatially separate locations on the plant root, throughout the experiment. Pathogen inoculum obtained by mixing peat with microconidia and subsequent incubation for four days at 22 degrees C, yielded a better percentage of diseased plants than a microconidial suspension drench, an injection of a microconidial suspension into the hypocotyl, or a talcum inoculum. Pseudomonas fluorescens strain WCS374 applied in talcum or peat, but not as a suspension drench, induced systemic resistance. A minimal initial bacterial inoculum density of greater than or equal to 10(5) CFU WCS374 root-(1) was required to significantly reduce the percentage diseased plants. At least one day was necessary between bacterization of strain WCS374 in talcum on the root tips and inoculation of the pathogen in peat on the root base, for an optimal induction of systemic resistance. Strain WCS374 induced systemic resistance in six radish cultivars differing in their susceptibility to F. oxysporum f. sp. raphani. Significant suppression of disease by bacterial treatments was generally observed when disease incidence in the control treatment, depending on pathogen inoculum density, ranged between approximately 40 to 80%. Strains WCS374 and WCS417 of Pseudomonas fluorescens induced systemic resistance against fusarium wilt, whereas P. putida WCS358 did not. This suggests that the induction of systemic resistance by Pseudomonas spp. is dependent on strain-specific traits.