The results of a cytogenetic study carried out in a group of nine radiologists are presented. Chromosome aberrations were detected by fluorescence plus Giemsa staining and fluorescence in situ hybridization. Dose estimates were obtained by extrapolating the yield of dicentrics and translocations to their respective dose-effect curves. In seven individuals, the 95% confidence limits of the doses estimated by dicentrics did not include 0 Gy. The 99 dicentrics observed in 17,626 cells gave a collective estimated dose of 115 mGy (95% confidence limits 73-171). For translocations, five individuals had estimated doses that were clearly higher than the total accumulated recorded dose. The 82 total apparently simple translocations observed in 9722 cells gave a collective estimated dose of 275 mGy (132-496). The mean genomic frequencies (x100 ± SE) of complete and total apparently simple translocations observed in the group of radiologists (1.91 ± 0.30 and 2.67 ± 0.34, respectively) were significantly higher than those observed in a matched control group (0.53 ± 0.10 and 0.87 ± 0.13, P < 0.01 in both cases) and in another occupationally exposed matched group (0.79 ± 0.12 and 1.14 ± 0.14, P < 0.03 and P < 0.01, respectively). The discrepancies observed between the physically recorded doses and the biologically estimated doses indicate that the radiologists did not always wear their dosimeters or that the dosimeters were not always in the radiation field. © 2005 by Radiation Research Society.