Formaldehyde (FA) is a chemical traditionally used in pathology and anatomy laboratories as a tissue preservative. Several epidemiological studies of occupational exposure to FA have indicated an increased risk of nasopharyngeal cancers in industrial workers, embalmers and pathology anatomists. There is also a clear evidence of nasal squamous cell carcinomas from inhalation studies in the rat. The postulated mode of action for nasal tumours in rats was considered biologically plausible and considered likely to be relevant to humans. Based on the available data IARC, the International Agency for Research on Cancer, has recently classified FA as a human carcinogen. Although the in vitro genotoxic as well as the in vivo carcinogenic potentials of FA are well documented in mammalian cells and in rodents, evidence for genotoxic effects and carcinogenic properties in humans is insufficient and conflicting thus remains to be more documented. To evaluate the genetic effects of long-term occupational exposure to FA a group of 30 Pathological Anatomy laboratory workers was tested for a variety of biological endpoints, cytogenetic tests (micronuclei, MN; sister chromatid exchange, SCE) and comet assay. The level of exposure to FA was evaluated near the breathing zone of workers, time weighted average of exposure was calculated for each subject. The association between the biomarkers and polymorphic genes of xenobiotic metabolising and DNA repair enzymes was also assessed. The mean level of exposure was 0.44 +/- 0.08 ppm (0.04-1.58 ppm). MN frequency was significantly higher (p = 0.003) in the exposed subjects (5.47 +/- 0.76) when compared with controls (3.27 +/- 0.69). SCE mean value was significantly higher (p < 0.05) among the exposed group (6.13 +/- 0.29) compared with control group (4.49 +/- 0.16). Comet assay data showed a significant increase (p < 0.05) of TL in FA-exposed workers (60.00 +/- 2.31) with respect to the control group (41.85 +/- 1.97). A positive correlation was found between FA exposure levels and MN frequency (r= 0.384, p = 0.001) and TL (r= 0.333, p = 0.005). Regarding the genetic polymorphisms studied, no significant effect was found on the genotoxic endpoints. The results of the present biomonitoring study emphasize the need to develop safety programs.