Environmental and dietary factors are thought to be significant in breast cancer aetiology. The alkaline single-cell gel electrophoresis (‘Comet’) assay was used to examine breast milk cells for DNA damage and to measure the activity of extracts of the milk in causing such damage. UK-resident women were recruited as donors (n = 16) and provided ‘early’ (~4 weeks post-partum) and/or ‘late’ (~4 months post-partum) milk samples. Cells (79–94% viable, trypan blue exclusion) were either examined immediately for DNA damage or were cultured for 1 week prior to treatment with a breast milk extract. DNA damage in the form of single-strand breaks was quantified as comet tail length (CTL). Cell preparations examined immediately exhibited interindividual variation in median CTL (range 2.0–40.0 µm) with or without the DNA repair inhibitors hydroxyurea (HU) and cytosine arabinoside (ara-C). DNA damage decreased following culture, suggesting either DNA repair or death of DNA-damaged cells. Some donors' breast milk extracts induced DNA damage in their cultured cells and increases in median CTL were significantly greater with HU/ara-C (range 4.0–72.5 µm) than without (range 2.5–27.5 µm). Genotoxicity occurred without cytotoxicity (81–97% viability after treatment). Comparisons between cells and extracts from ‘early’ and ‘late’ milk samples did not support the idea of a progressive clearance of genotoxins from mammary lipid during lactation. Donors whose untreated cells contained the most DNA damage tended to yield genotoxic breast milk extracts. Cells isolated from milk activated the rodent mammary carcinogens o-toluidine and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP). The relevance of genotoxic exposures to breast cancer initiation requires further investigation.