Abstract: Bacillus amyloliquefaciens (strain Bs006) was previously shown to promote growth of Cape gooseberry, a plant that is susceptible to the pathogen F. oxysporum. The practical use of Bs006 for biocontrol of F. oxysporum is unclear due to the observed variability and reduced efficacy observed in field tests compared to greenhouse tests. The aim of the present work was to determine biotic factors related to Bs006 and pathogen that would affect the biocontrol activity. Experiments to determine the dose-response relationship between biocontrol agent (105, 106, 107, 108, 109 cfu/ml) and pathogen (102, 103, 104, 105, 106 cfu/g) in both sterile and non-sterile soil showed that biocontrol activity was significantly affected by the population level of the bacterium but not of the pathogen. An adequate disease control was achieved when strain Bs006 was applied at 108 cfu/ml and inoculum density of F. oxysporum at 104 cfu/g. The absence of native soil microflora negatively affected the biocontrol activity of Bs006, but this did not make the soil more conducive for the disease, suggesting that the interaction between native microflora and antagonist is important for biocontrol. Bacterial metabolites inhibited the growth of F. oxysporum at concentrations higher than 5% in vitro, but pathogen metabolites also reduced the growth rate of strain Bs006. This work demonstrates the importance of understanding the ecological fitness of microbial antagonists to design strategies to improve their performance in the field.