Epidemiological studies have shown an association between the consumption of raw farm milk and reduced incidence of allergy. In the present study, we fed untreated raw milk, gamma-sterilised milk, heat-treated milk or water to mice and compared their responses to allergen exposure and challenge treatment in a mouse model of gastrointestinal allergy. From weaning (3 weeks old), groups of BALB/c female mice (n 8) received raw milk, gamma-sterilised milk, heated milk or water via drink bottles, with the control group receiving water. All mice were fed a standard (dairy protein-free) rodent diet. At 6 and 8 weeks, groups were given intra-peritoneal injections with ovalbumin (OVA)/alum to sensitise them to the antigen. Controls were sham immunised. At week 10, mice were fasted and challenged four times on alternate days by intra-gastric administration with 50 mg OVA or saline. Levels of bacteria and milk proteins were assessed in milk samples. Mouse serum levels of specific IgE, IgG1 and IgG2a antibodies and mouse mast cell protease-1 (MMCP-1) were determined. Cytokine responses to 48 h activation with OVA were measured in cultured splenocytes from mice. Sterilised and heated milks contained no viable bacteria and reduced detectable levels of many milk proteins, in contrast to raw milk. Mice drinking raw milk had highest serum MMCP-1 and specific-OVA IgE responses. Cultured splenocytes from OVA-primed mice produced similar levels of IL-4 in response to the antigen; however, IL-10 levels were highest from mice drinking raw milk. Overall, the present study adds to the evidence that consuming different types of milk can affect allergic responses to a non-related dietary antigen.