In order to study the metabolic differences between whole blood and isolated lymphocyte cultures, two indirectly acting mutagens cyclophosphamide (CP) and benzo[a]pyrene (B[a]P) were assessed for their potential to induce micronuclei (MN) in the presence and absence of S9 microsomal fractions. In isolated lymphocyte cultures supplemented with S9, CP and B[a]P induced a statistically significant increase in MN which was not observed in whole blood cultures. However, the directacting agent methyl methanesulphonate (which was used as a positive control) showed an increase in MN frequency in a dose-dependent manner in both culture methods. The effect of erythrocytes was then investigated by treating isolated lymphocyte cultures simultaneously with CP and S9 mix in the presence of purified erythrocyte concentrate (PEC). A clear reduction in the MN frequency was observed compared to the frequencies of MN induced in isolated lymphocyte cultures treated with CP and S9 mix in the absence of PEC. Thus, isolated lymphocyte cultures may represent a more sensitive test system for the evaluation of potential indirectacting mutagens. However, whole blood cultures may reflect the ‘real life’ situation more accurately as a consequence of the presence of erythrocytes.