► Differences in repair efficiency of four polycyclic aromatic hydrocarbons could not be related to bay or fjord region structures. ► DBPDE was found to be the most reactive of the four polycyclic aromatic hydrocarbon diolepoxides tested here. ► Regarding DNA repair and mutagenicity, DNA adducts of DBPDE behave differently than adducts derived from the other tested PAH–DE. Polycyclic aromatic hydrocarbons (PAH) are an important class of environmental contaminants many of which require metabolic activation to DNA-reactive bay or fjord region diolepoxides (DE) in order to exert their mutagenic and carcinogenic effects. In this study, the mutagenicity of the bay region diolepoxides (+)-anti-7,8-dihydroxy-9,10-epoxy-7,8,9,10-tetrahydrobenzoapyrene (BPDE) and (±)-anti-1,2-dihydroxy-3,4-epoxy-1,2,3,4-tetrahydrodibenzoa,hanthracene (DBADE) and the fjord region diolepoxides (±)-anti-11,12-dihydroxy-13,14-epoxy-11,12,13,14-tetrahydrodibenzoa,l-pyrene (DBPDE) and (±)-anti-3,4-dihydroxy-1,2-epoxy-1,2,3,4-tetrahydrobenzoc-phenanthrene (BPhDE) was compared in nucleotide excision repair (NER) proficient and deficient hamster cell lines. The 32P-postlabelling assay was applied to analyze DNA adduct levels and the Hprt gene mutation assay for monitoring mutations. Previously, we found that the mutagenicity per adduct was four times higher for DBPDE compared to BPDE in NER proficient cells. In these same cells, the mutagenicity of DBADE and BPhDE adducts was now found to be significantly lower compared to that of BPDE. In NER deficient cells the highest mutagenicity per adduct was found for BPDE and there was a tenfold and fivefold difference when comparing the BPDE data with the DBADE and BPhDE data, respectively. In order to investigate to what extent the mutagenicity of the different adducts in NER proficient cells was influenced by repair or replication bypass, we measured the overall NER incision rate, the rate of adduct removal, the rate of replication bypass and the frequency of induced recombination in the Hprt gene. Since NER turned out to be an important pathway for the yield of mutations, we further analyzed the role of transcription coupled NER versus global genome NER. However, our data demonstrate that neither of these pathways seems to be the sole factor determining the mutation frequency of the four PAH–DE and that the differences in the repair efficiency of these compounds could not be related to the presence of a bay or fjord region in the parent PAH.