The diameter of pinpoint inoculated cheese contaminates (Cladosporium herbarum, Penicillium roqueforti, Penicillium caseifulvum and Penicillium commune), isolated from either the dairy environment or directly from cheese, were inoculated 24 h after inoculation of the secondary starters to simulate contamination at the critical point of the salt brine. Pure P. camemberti had the largest inhibitory effect on the C. herbarum contaminant. Adding G. candidum in mixed cultures weakened the inhibitory effect of P. camemberti on C. herbarum. Low levels of G. candidum (10(3) spore/ml) promoted visible growth effects of C. herbarum, and this was most pronounced in the early stages of growth. The interaction mechanism of C. herbarum was not affected by the choice of the strain of P. camemberti whereas the Penicillium contaminants were very sensitive to the choice of the P. camemberti strain. The presence of G. candidum in the mixed cultures seems to decrease the suppressing effect of pour-plated P. camemberti. No correlation of any kind was found in the pour-plated spore concentration totals by the inhibition of the C. herbarum and P. roqueforti contaminants whereas P. caseifulvum and P. commune were sensitive to this.