Exposure of humans to poly cyclic aromatic hydrocarbons is an ongoing concern because of the carcinogenicity of these substances. DNA adducts are being increasingly used as indicators of carcinogen exposure. While considerable experimental evidence exists to support their use there are aspects that require further attention, especially after repeated exposure, which has led to this series of experiments. Male Sprague-Dawley rats were dosed with 10 mg/kg benzo[a]pyrene (B[a]P) i.p., 3 times/week for 2 weeks. At 1, 3, 7, 14, 28 and 56 days after the last treatment liver, lung, spleen and peripheral blood mononuclear cells (PBMNs) were sampled. The DNA adduct levels, as measured by the 32P-postlabelling technique, were significantlyincreased in all tissues, with lung having the highest levels. At day 14 total DNA adducts in lung, spleen and PBMNs were still >50% of the level at day 1. The removal of total DNA adducts was found to be fastest in liver > spleen < PBMNs > lung. A consistent correlation of total adducts between the lung and PBMNs was observed. A major adduct, designated adduct 1, was detected in all tissues, while adduct 4 was only found in liver and lung. Adduct 5 was detected only in lung, where it constituted ∼38–49% of total adducts and persisted at a higher level than either adduct 1 or adduct 4 for the entire post-exposure period. These results indicate that B[a]P induced a significant increase in DNA adduct levels in all tissues tested and that total adducts in PBMNs reflect total lung adduct levels. DNA adducts were still readily detectable 56 days after exposure ceased. Thus the results support the use of PBMNs as surrogates for estimation of B[a]P exposure in lung, the primary target organ, and indicate that samples taken days or weeks after repeated exposure will still yield DNA adductd at detectable levels. The role and significance of adduct 5 deserves further investigation, as it was detected only in the primary target organ for B[a]P-induced cancer.