A great number of genotoxins are known to be present in the environment. These genotoxins induce many kinds of DNA damage, and may cause changes in genetic information and cancer. Therefore, evaluation of such DNA damage is important to keep genetic information stable and to prevent cancer. We reviewed here methods used to assay the damage and the importance of this DNA damage in mutation. DNA damage is categorized into two groups, strand breaks and base modifications. To assay DNA strand breaks, the alkaline elution method, pulsed-field gel electrophoresis and single-cell gel assay are being used. The alkaline elution method determines both single-and doublestrand breaks sensitively and quantitatively. Pulsed-field gel electrophoresis preferentially determines double-strand breaks, and the results of the method appeal to the eye as an electrophoretogram. The single-cell gel assay could determine both single-and double-strand breaks even in a single cell, and could evaluate susceptibility to the damage in individual cells. To assay base modifications, methods to detect differences in the physicochemical properties of the damage (physicochemical methods), immunoassays and the 32P-postlabeling method are being used. Physicochemical methods are suitable for chemically minor and abundant modifications such as those in 8-hydroxyguanine, using high-performance liquid chromatography or gas chromatography/mass spectroscopy. Immunoassays, by the use of specific antibodies against DNA damage, are highy sensitive to DNA damage such as changes in O6-methylguanine and thymine glycol, and simple once the assay systems have been established. The 32P-postlabeling method is the most sensitive for the detection of bulky DNA adducts. Indirect methods are also being used to estimate base modifications. Determination of protein-genotoxin adducts may become an alternative to detect DNA adducts wi hout handling DNA. To assay modified bases excreted in urine may also be used to estimate DNA modifications present in tissues. However, every method has shortcomings and no method is perfect. Thus, to effectively evaluate DNA damage a combination of the above methods or the most suitable method for the genotoxin of interest should be used.